Direct Marketing

Investment Banking Team Industry Reports

FALL/WINTER 2005

Direct Marketing 2005: The Emergence of Convergence

FOREWORD
In June 2005, under the new leadership of John Greco, the Direct Marketing Association released its “Ten Commandments” for how DMA members should conduct their marketing and customer relationship activities. We think it very positive for the DMA to have boiled down its tenets from the previous list (totaling 51 rules), since it is easier for members to follow fewer guidelines and, frankly, we think the ten new Principles to live by are appropriate and adequate. The ten DMA Member Principles are: A DMA member: 1) Is committed to its customers’ satisfaction; 2) Clearly, honestly and accurately represents its products, services, terms and conditions; 3) Delivers its products and services as represented; 4) Communicates in a respectful and courteous manner; 5) Responds to inquiries and complaints in a constructive, timely way; 6) Maintains appropriate security policies and practices to safeguard information; 7) Provides information on its policies about the transfer of personally identifiable information for marketing purposes; 8) Honors requests not to have personally identifiable information transferred for marketing purposes; 9) Honors requests not to receive future solicitations from its organization; and 10) Follows the spirit and letter of the law as well as the DMA’s Guidelines for Ethical Business Practices. We at The Keystone Equities Group (a DMA Member ourselves) wholeheartedly endorse the DMA Member Principles, and believe that all Members should strive to follow them. The Direct Marketing industry has been through some very turbulent periods of adverse public opinion and legal and regulatory setbacks due to the problems created by a few bad apples. We are pleased and honored to consider ourselves part of the Direct Marketing industry, and we think that if firms follow these ten simple rules, public perception will improve, government regulatory efforts may slow, and Direct Marketers will have more (and happier) customers.

The Keystone Equities Group Marketing Services Investment Banking Team October 2005

Fall/Winter 2005 Direct Marketing Industry Report

2

TABLE OF CONTENTS

Foreword ................................................................................ 2 Introduction............................................................................ 4 State of the Direct Marketing Industry — 2005 Update ...... 6 Outdoor ............................................................................. 10 Direct Mail.........................................................................11 Telephone Marketing......................................................... 13 Energy Prices .................................................................... 15 Identity Theft & Phishing .................................................. 16 Internet Sales Tax .............................................................. 19 Online Marketing .............................................................. 20 Hispanic Marketing........................................................... 28 Industry Meta-Trend: New Media Convergence................ 32 Digital Video Recorders .................................................... 34 The Next Generation of Television — IPTV ..................... 38 Pay-per-Call ...................................................................... 44 Online/Offline Print Synergy............................................. 46 About The Keystone Equities Group .................................. 49 The Marketing Services Investment Banking Team........... 49 Disclosures ............................................................................ 50

Fall/Winter 2005 Direct Marketing Industry Report

3

INTRODUCTION
The Direct Marketing industry has experienced a substantial improvement in fortunes since our first industry white paper was published in late 2003. Major regulatory changes occurred that led many to claim the potential doom of some or all of Direct Marketing, such as Do-Not-Call and the Can-Spam Act, but it appears the industry has generally taken the obstacles in stride. Indeed, outbound telemarketing has lost substantial mindshare and market share since the initiation of Do-Not-Call, but we believe the industry is ultimately stronger for it. The changes incorporated by the federal government instituted a (generally) level playing field for all Direct Marketers, and it appears that industry participants are willing to play by the new rules. The telemarketing sector hasn’t experienced the “millions of lost jobs” predicted back in 2003, but some have been lost and others have been refocused on inbound calls. As for Can-Spam, we have personally experienced a reduction in certain types of email solicitations (pornography, get-rich-quick schemes), but that has been replaced, and then some, by massive blasts of online dating and replica product spam. We think we are receiving substantially more spam email than a year ago, and it is harder for our IT personnel to create an effective firewall to let in good emails while keeping out the spam. In essence, we are lacking the tools to make an “intelligent” differentiation between desired and undesired electronic correspondence. Overall, Direct Marketing clients have come back after the 2001 recession and are spending well — and expect to spend even more in coming quarters. However, the fortunes of marketing services providers appear to be increasingly linked to their channels of specialization; multi-channel marketing is occurring but online, DRTV and direct mail campaigns seem to be benefiting at the expense of print, radio, and outbound telemarketing. The concept of “direct mass marketing” coined by Stan Rapp, and explained in detail in our 2004 industry white paper, seems to be playing out most effectively in the use of television campaigns that drive the interested masses to an Internet microsite created specifically for that campaign. In this white paper, we explore in detail the convergence of media channels and its potential impact on Direct Marketers. Evolution of the television, Internet, and telephone has created a marketing revolution, providing the ability to create and effectively track true “multi-channel” marketing campaigns. The lines between TV, Internet and phone are rapidly blurring, as evidenced by the introduction of video-on-demand, television over DSL or fiber-optic cable (Internet Protocol TV or IPTV), and mobile phones capable of displaying streaming video downloaded in real time. In addition, the ongoing proliferation into the home of digital video recorders (DVRs), such as the ubiquitous TiVo, have had advertisers worried for several years. It now appears that DVRs, pushed by cable and satellite providers as incremental revenue generators, are crossing over from early adopters to the population at large. What does this mean for marketers? Strong interest continues to emanate from private equity groups to invest in the
Fall/Winter 2005 Direct Marketing Industry Report 4

“marketing services” sector. There are two types of interested investors: one that doesn’t know much about the industry, but thinks it’s an increasingly important sector in which to participate; the other is knowledgeable in marketing services, and seeking either complementary snap-ons to existing portfolio companies or wants to expand into new sectors (for example, they own a direct mail company but are also interested in data/ analytics or online service providers). The first group wants low valuations, while the second group is concerned about future adverse regulatory changes and technological evolution (“TiVo will eliminate DRTV”). One problem common to the larger private equity players: they want revenues over $50 million and EBITDA (earnings before interest, taxes, depreciation and amortization — a common indicator of profitability of an enterprise) over $5 million. Experienced participants are also raising smaller amounts of funds to do a number of smaller acquisitions in niche sectors. These roll-up strategies have the benefit of having more companies to target, but integration of multiple entrepreneur-owned companies may provide ultimately quite difficult without a seasoned executive team and a commitment to a well-defined plan. Again, we thank you, the reader, for taking the time to review our annual white paper. We received a significant number of positive responses to our past white papers, and even had the good fortune to have our 2004 Keystones report incorporated into a course syllabus at the Annenberg Graduate School of Communication, University of Southern California. Comments and feedback are always appreciated, as are ideas for future issues. If your company is seeking advisory or placement agent services for capital raises, mergers and acquisitions, or corporate consulting, we would be very pleased to discuss your needs and goals. Contact us at (800) 715-9905 or visit our website at www.keystoneequities.com. We also thank Greg Bruner and Gail Enault of Michael Edwards Direct, a marketing services agency based in Chicago, for again being kind enough to supply us with the Quark files for our cover pages. To obtain more information on Michael Edwards Direct and their services and experience, please call Greg Bruner at (312) 944-0606 x15 or visit their website at www.michaeledwards-direct.com.

IMPORTANT NOTE: Inclusion in this report is not intended to be, nor should it be construed as, an investment recommendation. In our view, the information provided about companies whose securities are publicly traded is not information sufficient upon which to base an investment decision. Readers who are considering an investment in any company identified in this report should obtain additional information before making an investment decision. See additional disclosures on page 50.

Fall/Winter 2005 Direct Marketing Industry Report

5

State of the Direct Marketing Industry - 2005 Update
Direct Marketing is not going away, even with the significant regulatory and technological changes that have occurred in the United States Direct Marketing industry during the past couple of years. In fact, the industry continues to grow at a faster rate than either the US economy or total retail sales. The industry association Electronic Retailing Association (ERA) had estimated that total US sales in 2004 attributed to all direct marketing channels reached $2.3 trillion. The Direct Marketing Association (DMA) estimated in September 2005 that American companies would spend more than $161 billion on Direct Marketing campaigns in 2005, which could generate $1.85 trillion of resulting sales this year (7% of US GDP). The DMA estimates that Direct Marketing sales will accelerate in the second half of the decade, growing at a projected 6.4% per year from 2004 to 2009, up from 5.3% from 1999 to 2004. In comparison, total retail sales are actually expected to slow down from about 4.8% in 1999-2004 to 4.5% in 2004-2009. Direct Marketing is expected to support about 10.6 million jobs in 2005, rising about 2% per year through 2009.

Estimated Annual Sales Growth 1999-2009

Direct Marketing Sales
6.50% 6.00% 5.50% 5.00% 4.50% 4.00% 3.50% 3.00% 1999-2004 2004-2009
projected

Total Retail Sales

Source: The Direct Marketing Association.

The Direct Marketing Association (DMA) estimated that total media spending for Direct Marketing initiatives rose to $216.9 billion in 2004, up 5.9% from 2003 and up more than 12% from 2002. Moreover, Direct Marketing campaign expenditures accounted for 47.9% of total advertising in 2004, up from 46.9% back in 1999 as advertisers continue to seek better information relating to returns on marketing investment, and as more large advertisers realize that Direct Marketing advertisements also have a positive branding effect. The DMA’s Quarterly Business Review for the second quarter of 2005
Fall/Winter 2005 Direct Marketing Industry Report 6

(ending June) found a majority of respondents claimed to have growth in revenue for that quarter relative to the same period in 2004. In addition, the QBR’s industry-wide revenue index reflected growth for the eighth straight quarter — with an expectation for even stronger growth for the September quarter.
Year-over-Year DM Industry Revenue Growth Index
75 70 65 60 55 50 45 40 35 30
Q42002 Q12003 Q22003 Q32003 Q42003 Q12004 Q22004 Q32004 Q42004 Q12005 Q22005 Q32005 proj.

>50 = growth; <50 = decline

Source: DMA Quarterly Business Review.
1H2005 Annual Advertising Spending Growth, By Media
Spanish Language TV Cable TV Internet Ads Local Magazine National Magazine Outdoor Total Netw ork TV B2B Magazine Local New spaper National New spaper Coupons 0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 16.00%

Source: Nielsen Monitor-Plus.

Fall/Winter 2005 Direct Marketing Industry Report

7

Growth in 2005 for all advertising has been relatively robust, even after a healthy 2004. According to Nielsen Monitor-Plus, advertising spending rose 6.3% in 2004, well above overall economic growth, and driven substantially by spending for the Summer Olympics and for the November 2004 political campaigns. For the first six months of 2005, Nielsen reported 5.7% year-over-year growth, and the second quarter had faster spending growth than the first quarter. Strongest growth occurred in the media channels of Spanish-language TV, cable TV, Internet (online CPM display ads only), and local and national magazines. Nielsen claimed the top advertisers for the first half of 2005 were General Motors, Procter & Gamble, Ford Motor, Johnson & Johnson, and Time Warner. The 29th annual Advertising Ratios & Budgets study by Schonfeld and Associates, released in July 2005, forecasted that 2005 advertising spending would be up 7.7% over 2004, with sales growth up 9.8% over the same period. The estimated advertising spend for 2005 was 2.0% of total sales, which ranged from 6.2% of sales for consumer products to 0.3% for oil, gas, and chemicals. The Schonfeld spending estimate was significantly higher than the 3.4% growth rate for 2005 estimated in June by TNS Media Intelligence (to $145.3 billion); growth for the second half of the year was estimated to be just 2.7%, mainly due to hard year-over-year comparisons with the Olympics and Presidential election campaign last year. ID Media estimated that political spending on television alone in 2004 was about $1.5 billion.
2004 DMA Response Rate Survey, by Channel
DR Radio DRTV FSIs M agazine Newspaper Telephone Coupons Inserts Email Catalog Dimensional Mail Direct Mail

Median Average

0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

6.00%

Source: 2004 DMA Response Rate Study. Our discussions with agencies, marketing service providers, and direct marketers of products and services have generally found strong growth in 2005 over 2004 — which is corroborated by the generally positive results of the DMA’s Quarterly Business ReFall/Winter 2005 Direct Marketing Industry Report 8

port during the past several quarters. A TNS Media Intelligence report on the first half of 2005, released in early September, found that “direct response” was by far the fastest growing category of advertising, up 19.9% over the prior-year period. That was more than double the growth rate of any other advertising category (restaurants were second at 9.2% growth). According to the 2004 DMA Response Rate Study, the direct marketing channel with the highest response rate was dimensional mail, at 5.49%. The other top-performing channels were telephone (5.45%), direct mail (2.73%), email (2.31%), catalog (2.23%) and coupons (1.77%). The worst-performing channel for response rate was DRTV, at 0.14%. Email’s response rate increased from 1.87% in 2003. We have provided a chart below which incorporates a mean (average) and a median by channel. We added the median, because the “average” is skewed by high response outliers and a relatively small total number of responses for each channel. The 2005 Multi-Channel Retail Annual Trend Report from Abacus found that Abacus Alliance members (catalog/online retailers) had an average 8% increase in annual household spending in 2004 (10% growth through call center and website sales conversions), and 55 million households purchased from the members 243 million times for $23.4 billion in direct response sales. The two biggest growth categories for the cataloguers were gardening and backyard (18% growth) and children’s products (14% growth). The study had a benchmark group of multi-channel marketing members, and 38% of those companies’ sales came from their websites, up from 32% in 2003.

Top Priorities of Target Marketers for 2005
Acquire New Customers Increase Program Response Rates Improve Target Marketing ROI Improve Customer Longevity Increase Wallet Share Reduce Target M arketing Costs

1st Priority 2nd Priority 3rd Priority

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Source: Harte-Hanks/CSO, 2005.

Source: Harte-Hanks/CSO Insights, 2005. The recently announced CSO Insights/Harte-Hanks report on target marketing (a survey of 281 firms, of which 189 had annual revenues in excess of $10 million and 92 had revenues above $500 million) found that more than 45% of respondents were spending at least 30% of their total marketing budget on “target marketing,” and more than 22% of respondents were spending more than 45%. Perhaps more specifically and more important to our readers, almost 80% (n = 231) performed some sort of one-to-one or perFall/Winter 2005 Direct Marketing Industry Report 9

sonalized marketing on a formal or informal basis, and more than 80% of users of such one-to-one marketing found it improved campaign success. The acquisition of new customers was by far the top targeting marketing priority of respondents for 2005, followed by the improvement of return on investment in target marketing campaigns. We see the relatively small number of respondents seeking a reduction in target marketing costs (and the significant focus placed on higher ROI and better response rates) as an opportunity for marketing service providers to grow revenues if they provide better methods and technologies to help Direct Marketing campaigns improve performance. To those who believe that multi-channel marketing is a waste of time, we submit a very interesting ClementDIRECT study done in 2004. The company found that, of the single-channel (retail store, online or catalog) retailers, the online retailers enjoyed the highest average annual sales per customer at $90, versus $69 for catalog and $43 for retail. Companies using catalog+website had average sales per customer of $99, about 10% higher than website alone; catalog+retail presence had sales of $124 per customer and online+retail had average customer spend of $127. In other words, using a catalog+retail or website+retail approach resulted in substantially higher sales than just combining the average results of the two channels used separately. Marketers using all three channels in an integrated format realized $243 per customer per year on average. While having a retail presence substantially increases the fixed cost overhead of an entity, these numbers suggest that retail-only merchants may enjoy a significant boost in sales and exposure via an ongoing multi-channel direct marketing presence. Outdoor Weakness in terrestrial radio, print, and network TV have led advertisers to search for other means of capturing eyeballs for branding and Direct Marketing campaigns. We see this as the evolution of direct mass marketing — trying to get as many people to the top of the sales funnel as possible. One mass media channel that has seen substantially higher demand recently is outdoor advertising. According to the Outdoor Advertising Association of America, $3.2 billion in revenue was spent on outdoor media buys in the first half of 2005, up 7.0% from the same period in 2004. According to TNS Media Intelligence, outdoor accounted for 2.4% of total media spending in 2004 in the United States. It appears the “outdoor” segment has grown from billboards to encompass any sort of advertising “outside your front door”. An Alloy Media + Marketing executive was recently cited in MediaDailyNews highlighting their very rapid growth in washroom advertising. We suppose washroom advertising is a pretty decent means of ensuring that marketers are capturing the attention of a targeted gender, though. The same MediaDailyNews report stated that Clear Channel Communications, following weak results in its radio broadcasting operations, would be spinning off its outdoor division in an initial public stock offering. Clear Channel’s outdoor segment is its fastest-growing and most profitable division. The New York Times reported in July 2005 that the WPP units mediaedge:cia and Mindshare would be sharing offices in a joint
Fall/Winter 2005 Direct Marketing Industry Report 10

venture (called Kinetic) with leading outdoor agency Poster Publicity in New York. Kinetic identified more than 200 different formats now encompassed within the outdoor sector — including putting advertisements on manhole covers in urban areas. New York City has tendered a 20-year contract, worth an estimated $1 billion, to have advertisers build and oversee outdoor advertising on city transit shelters, public washrooms, and newsstands. An adage.com report in September 2005 stated Cemusa, a small Spanish company, had won the contract.

U.S. Outdoor advertising media spending 1990-2004 ($billions):

Source: Outdoor Advertising Association of America. Direct Mail Direct mail continues to be a significant beneficiary of the recovery in direct marketing budgets and the reduction in availability of telemarketing as a targeting media. The channel continues to be the single largest marketing or advertising medium, and is growing rapidly. Robert Coen of Universal McCann, in his semi-annual marketing campaign spending forecast in June 2005, projected that direct mail spending would rise 8.5% in 2005 (more than twice the rate of the overall economy and almost double the TNS Media Intelligence estimate for US advertising spend growth for the year) to $56.6 billion. A presentation by a USPS executive in the summer of 2005 reported that 95% of American households collect, assess, and sort their mail on a daily basis, while only 57% of households check their email on at least a weekly basis. However, this email result is obviously skewed by age; 81% of households headed by someone under the age of 30 check their email at least weekly. The USPS also claims that Americans under the age of 39 (“Generation X” and “Generation Y”) still appreciate snail mail — 75% of such respondents apparently said so. Approximately 74% of Gen Xers and 68% of Gen Yers have used direct mail coupons, and 63% of these whippersnapper respondents said they liked receiving catalogs. In fact, 64% had ordered an item online after seeing it in a catalog. A speech by the Postmaster General in early 2005 announced that consumers
Fall/Winter 2005 Direct Marketing Industry Report 11

who received retailers’ catalogs tended to view 22% more pages on the retailers’ websites and spent 16% more money than those who did not first receive a catalog.

USPS Mail Volume, By Category (billions of pieces) First-Class/Priority
108 106 104 102 100 98 96 94 92 90 2000 2001 2002 2003 2004

Bulk/Direct

Sources: United States Postal Service, CFO Magazine (April 2005). Vertis’ Customer Focus 2005 survey, released back in March, found that 24% of adults who read incoming direct mail had visited a retail store due to the direct mail solicitation. Half of the respondents said they were more likely to open the direct mail solicitation if the envelope highlighted a special discount or offer inside. However, the survey found that direct mail (other than catalogs) tend to be more effective on older people within the household. This makes sense to us, since the heads of households are more likely to sort and dispose of incoming mail than the kids. A survey by Decision Direct in summer 2005 of Internet shoppers from 37 multichannel merchants (including 47,000 completed customer surveys) found that catalogs were powerful tools in driving consumers to a retail website, but that email was actually more effective. Approximately 81% of respondents stated they were likely to visit a retailer’s website after receiving an email, while 78% said they were likely to go there after receiving a catalog. Are catalog companies simply cannibalizing their offline sales through online sales? According to a Pitney Bowes study released in July 2005, about 30% of online sales for catalog companies are incremental, and online buyers that previously received a catalog in the mail tend to be “over twice as likely” to purchase from the catalog company’s website and spend $114 more for their first purchase than firsttime visitors who show up at the site without a catalog in hand. The DMA estimated that the annual number of catalogs mailed in the United States rose from about 12 billion in 1993 to more than 17 billion in 2003. Hurricane Katrina is likely to have eliminated any hope of a delay in postal rate inFall/Winter 2005 Direct Marketing Industry Report 12

creases by the USPS. In April 2005, the USPS requested 5.4% increase in all rates, which was to be implemented in about March 2006. It is now expected that the USPS will institute the rate increase in January 2006 to help cover more than $100 million in projected repairs to the postal infrastructure along the Gulf Coast and corresponding loss of revenue from the area. In addition, fuel is a major expense component for the USPS, given its massive delivery infrastructure (212,000 vehicles). A single penny increase in the price of fuel is estimated by the USPS to add about $8 million to their annual expense. Another hike in mail pricing for 2007 is already being discussed. Telephone Marketing Telemarketing continues to be negatively impacted, especially given the ongoing and enormous success of the national residential Do-Not-Call registry. According to the Federal Trade Commission, the national registry broke the 100 million number level in August 2005. Of those, approximately 90 million came from the national registry (65% via Internet registrations and 35% by phone registrations) and about 10 million from various state registries. The FCC continues to fine telemarketers who break the rules (numbering almost 100 investigations and 15 citations through the end of fiscal 2004, which is the most recent available data), and Verizon Wireless filed lawsuits in September 2005 against two telemarketing firms using autodialers to call mobile phones. To its credit, the DMA issued a statement supporting the “vigorous enforcement” of regulations preventing unsolicited mobile phone telemarketing campaigns. Universal McCann’s June 2005 updated projections for the American advertising industry noted that, while total direct response advertising was up sharply in the first half of the year over the same period in 2004, restrictions on telemarketing were also driving significant portions of marketing budgets to other direct response channels. In particular, Coen claimed that direct mail was a primary beneficiary of budgets formerly spent on telemarketing.

Total Global Outsourced Contact Center Agents, 2004-08
1,400,000 1,200,000 1,000,000 800,000 600,000 400,000 200,000 0 2004 2008 (est.)

Rest of World US

Source: Datamonitor.
Fall/Winter 2005 Direct Marketing Industry Report 13

A report released in February 2005 by Dimension Data found that inbound call centers worldwide were having issues accommodating the needs of consumers. Inbound callers were willing to wait less time to talk with a live representative (65 seconds versus 73 seconds in 2003), while representatives felt they were overworked and lacked the necessary information to confidently answer customers’ inquiries. Consequently, call center agent turnover continued to rise (23% in 2005, versus 19% in 2003). However, it appears inbound call centers continue to enjoy rapid growth, as more than 58% of surveyed call centers were having 20+% annual call volume growth. The labor arbitrage opportunity of moving contact centers to outside the United States also continues unabated. According to a Datamonitor report released in September 2005, the percentage of call center agents located in the United States is expected to fall from 315,000 in 2004 to 291,000 in 2008 — reducing the American labor pool’s share of global contact center positions from 37% in 2004 to 25% in 2008. Is outbound telemarketing dead? No. Business-to-business (B2B) telemarketing is still alive and kicking, and inbound telesales driven from other channels is still a vital component of direct marketing campaigns. There are also tens of millions of available numbers still not on the Do-Not-Call list. The CSO Insights report mentioned at the start of this report found that more than 50% of responding marketers had not changed their telemarketing budget (but no mention of whether that budget was big or small or zero) after the Do-Not-Call rules, but more than 30% had shifted media budgets away from telemarketing. Moreover, in 2008, a lot of these names will come available again, as consumers will have to remember to go back into the national registry and sign back up again. But the FTC continues to raise the fees from telemarketers. Effective September 1, telemarketers accessing the do-not-call registries to scrub their lists found the annual rates per area code rose from $40 to $56, and $15,400 for a group of 280 area codes or more (up from $11,000). However, the first five area codes requested by telemarketers are free — highlighting the interest of the FTC to provide free access to telephone marketing for local marketers or agents. While there are still opportunities for cold calling to households, outbound business-toconsumer (B2C) and even B2B outbound telemarketing seems to be shifting away from a cold outbound call to an outbound follow-up after originating the lead via another channel. We talk extensively about media convergence in this white paper, but several recent articles in the marketing trade journals have highlighted the use of online lead generation that obtains consent for a follow-up sales call (known as “permission-based telemarketing”). According to a May 2005 report by The Artemis Group, a consultancy, about 80% of online leads for B2B products thought positively of a follow-up phone call by a sales agent after online lead generation — and virtually no respondents were unhappy with the follow-up phone call. However, receptivity falls dramatically unless a call is made within 24-48 hours after the lead is generated. By 8-10 days after lead generation, less than 10% of leads responded on the first outbound call (versus 35% on same-day calls).

Fall/Winter 2005 Direct Marketing Industry Report

14

Later in this report, we explore the emerging and intriguing pay-per-call segment, which combines online paid search advertisements with the ability to drive incoming calls to a business or an agent to ask questions and complete the transaction. Energy Prices The rise of the price of crude oil and its refined variants is having a major effect on the American economy. Hurricane Katrina shuttered a significant portion of rigs in the Gulf of Mexico, supertankers could not unload at the Louisiana Offshore Oil Port (the only terminal in the United States that can handle fully loaded oil supertankers), and a considerable portion of the refining capacity of the nation remains under water as of the date of writing this document. Several million people and businesses in the southern United States have been directly hurt by the hurricane, and its effect on the price of oil and gasoline is impacting the entire world.
Unleaded Regular Gasoline, U.S. City Average Retail Price (Cents per Gallon, Including Taxes) 300 280 260 240 220 200 180 160 140 120 100 80 1990 January 1991 January 1992 January 1993 January 1994 January 1995 January 1996 January 1997 January 1998 January 1999 January 2000 January 2001 January 2002 January 2003 January 2004 January 2005 January

Source: Energy Information Administration, US Department of Energy. In September 2005, UPS announced plans to raise its fuel surcharge cap to 12.5% on next day air, second day air, and three day select domestic services as well as all international deliveries to and from the United States. The rise is tied to a spike in the cost of jet fuel, which has been in short supply as refiners shifted available capacity to production of automobile gasoline. In comparison, the fuel surcharge for UPS Ground shipments are tied to a governmental average of diesel fuel costs, and that surcharge was only 3% in September. FedEx had a fuel surcharge of 12.5% on Express (aircraftbased) shipments for August 2005 and 13% for September, versus a 2.75% surcharge for FedEx Ground in August and 3% in September. E-tailers and other Direct Marketers are expected to feel the pinch of increased costs of shipping, as most are unlikely to be able to pass along the full cost increase to their cusFall/Winter 2005 Direct Marketing Industry Report 15

tomers. Free or low-cost shipping promotions could eventually be at risk, which would likely reduce overall orders for e-tailers or cataloguers. However, the appeal of free shipping has been a significant sales driver for online shoppers, and it appears that the upcoming holiday season will continue to see a broad swath of e-tailers offering free shipping — even with higher shipping costs. An October 2005 survey by Shop.org and BizRate Research found that 79% of surveyed online shoppers said that free shipping offers were a significant factor in choosing where to buy online, and 79% of surveyed retailers were expecting to offer free shipping promotions of some sort for the 2005 holiday season. Our discussions with online retailers in September 2005 suggested that outbound shipments are generally not yet being hit with large surcharge increases, given their great importance to the expedited shippers. However, they expect a quite significant increase by the end of the year or in early 2006 unless oil prices retreat. The retailers also highlighted efforts by their suppliers to pass through their own higher fuel costs. We had originally neglected to consider this potential threat on operating margins for the Direct Marketing industry. One of our clients, PetFoodDirect.com, recently added a logo to their loyalty program emails that encourages people to: “SAVE GAS...SAVE MONEY...BUY DIRECT!” After sampling 37,000 households, ACNeilsen reported in September 2005 that consumers were adapting to higher energy costs, with identifiable differences in behavior between affluent and lower-income households. Low-income respondents were more likely to state they were using more coupons (23% versus 14% of affluent homes); buying lower-cost brands of groceries (20% versus 10% of affluent); or were buying cheaper grades of gasoline (19% versus 12% of affluent). More affluent respondents were more likely to go to Costco and other price clubs to get bulk bargains. We were a bit surprised to find that 7% of affluent respondents were trying to save money by buying more products online (versus 3% of lower-income households). Americans have been very aware of the current cost of gasoline, especially since we just ended the summer peak driving season. But the larger effect on households could come this winter, given that heating oil and natural gas prices are expected to be up significantly for the second winter in a row. An Associated Press report from late August 2005 suggested that heating oil costs per month for high-use homes (people home for most of the day) in the New England area could be as much as $800, which would be about 35% higher than in 2004. In many northern states, those homes unable to pay their natural gas or electric heating bills get a reprieve during the winter months. But heating oil firms are normally exempt from those rules, so holiday spending money could be burnt in the furnace rather than used for presents. Identity Theft & Phishing — Not Your Father’s Catch of the Day Another concern for the Direct Marketing industry is the ongoing expansion of identity theft and the corresponding (and inevitable) regulatory backlash against companies that
Fall/Winter 2005 Direct Marketing Industry Report 16

collect, store, use and resell consumer information. There have been a multitude of media articles reporting the theft and misuse of large banks of information, from information sellers such as ChoicePoint, transaction processors including CardSystems, banks such as Bank of America, and retailers such as the discount shoe chain DSW. In August 2005, the latest Experian-Gallup Personal Credit Index survey found that 18% of respondents had been victims of identity theft. Identity theft victims also tend to be younger than the average population. Twenty-five percent of respondents under age 30 had had their financial information stolen, versus 18% of people 31-64 and 11% of those over 65. The results also were geographically concentrated; 26% of respondents in the Western US, 20% in the East, 15% in the South, and 12% in the Midwest had reported theft of personal information. Sixty-two percent of respondents were concerned about identity theft online, 55% from their mail, 53% at a retail store, and 47% at a restaurant. If a July 2005 survey by Reconnex, a California-based enterprise risk management tool vendor, is anything to go by, we have not seen the last of these announcements — and the loss could be an inside job. Reconnex claimed that 91% of Fortune 1000 companies sampled had employee and/or consumer Social Security numbers entering or leaving their network unencrypted via email or peer-to-peer networks, and 82% also had credit card numbers. A whopping 80% of those companies also had previously-undetected peer-to-peer software running on their network, which can introduce spyware, key loggers, and other malicious software used to capture information or hurt the corporate network. We note that only 25% of respondents to the Experian-Gallup survey discussed above were concerned about having their financial information stolen at work or school. Phoenix Affluent Marketing Service, an industry analysis firm, has an ongoing survey of more than 1,000 affluent households . Their June 2005 survey found that 48% of respondents listed identity theft as a significant financial risk; it was actually among the top five financial concerns for affluent American households (along with maintaining existing wealth, the US economy, retirement planning and minimizing tax). The fear of identity theft was higher in female respondents (51%) than males (45%). The mostoften suggested remedies for the problem were:

Remedies for Identity Theft Among Affluent Households Provide immediate cancellation of account(s) once account has been breached Impose stricter guidelines on opening new card accounts Help clients to assess their risk level for becoming a victim of identity theft Proactively educate clients on the risks of identity theft Industry should offer additional ID theft protection for a fee

Strongly Agree 86% 79% 67% 62% 20%

Source: Phoenix Affluent Market Service. A June 2005 survey of 5,000 adults by Gartner, an international industry analysis firm, led the company to estimate that the number of American consumers receiving online “phishing” scams (in which an email designed to look like it is sent from a legitimate
Fall/Winter 2005 Direct Marketing Industry Report 17

company — such as a bank or brokerage — causes respondents to give out personal financial codes and information) rose 28% over the same period in 2004. Gartner extrapolated that more than 70 million Americans had received a phishing email in the past twelve months. We believe that number is likely on the low side, as we have received phishing emails at work and through home emails from groups posing as SunTrust, Wachovia, PayPal, eBay, Bank of America, Harris Bank, Smith Barney, Citigroup, and others. As we have stated before, the most common giveaways are that the entire email message is a clickable picture rather than conventional text and links, and that the grammar usually is slightly strange (especially the ending text and signature). Gartner’s survey found that 40% of respondents had received a phishing email within the prior two weeks, suggesting an extremely pervasive method of fraud. [Note: you can let the FTC know of a phishing attack by forwarding the offending email (without opening or clicking the message) to the FTC at [email protected].]

FTC Sentinel Online/Telephone Fraud Complaints 400,000 350,000 300,000 250,000 200,000 2002 2003 2004

Source: United States Federal Trade Commission. According to a June 2005 Federal Trade Commission report (The US Safe Web Act: Protecting Consumers from Spam, Spyware, and Fraud), about 16% of identity theft and fraud complaints sent to the FTC by Americans are against foreign sources. Of course, the federal government seeks to do something about these overseas scammers. The proposed US SAFE WEB Act is intended to reduce the incidence of identity theft and fraud from foreign sources and improve cooperation between the FTC and foreign police agencies. Unfortunately, a lot of people get caught at least once by these criminal attacks. Gartner reported that an estimated 2.4 million Americans have lost money through a response to a phishing scam, including 1.2 million people in the past year (with estimated losses of more than $900 million). The FTC report noted above found that almost 400,000 people reported outright fraud complaints to the Commission in 2004, 64% since 2002. Most domestic victims received back their lost funds from the credit card provider or bank.
Fall/Winter 2005 Direct Marketing Industry Report 18

This has direct implications on Direct Marketers using online commerce, especially those firms that are less well-known or without a multi-channel (retail stores and online) presence. Three of four respondents to the Gartner survey said they are now more cautious about where they buy goods online, and one of three stated they are buying less goods online than they would otherwise directly because of increased security concerns. A September 2005 article in Direct explored another issue for e-tailers caught in the web of identity theft and fraud. According to the article, federal law limits consumers’ fraud liability for any transaction to $50. But the merchant is also responsible for up to a maximum of $100 per fraudulent transaction, and could also eat up to $40 in fees levied by the banks and transaction processors for a fraudulent transaction. Since an Internet retailer doesn’t have the ability to do a face-to-face verification of a credit card before processing, all they can do is utilize more sophisticated fraud-flagging software or implement increased security. We think e-mail prospecting by relatively unknown firms is becoming increasingly difficult, as response rates decline due to lack of trust in the sender. Gartner’s report stated that more than 80% of Americans are estimated to have lower trust in incoming emails arriving from unknown parties, and more than 85% delete a potentially suspicious email rather than opening it. In fairness, if 15% of respondents opened a prospecting email (one that followed current regulatory requirements and disclaimed itself as an advertisement), it would probably be judged quite a successful campaign. Online commerce is beginning to slow its phenomenal growth rate, but it is still well outperforming the overall rate of retail sales increases. The June 2005 Gartner report discussed above found that 77% of Americans with access to the Internet had actively shopped online in the prior 12 months. But it also found that 40% of respondents that had shopped online had reduced or plan to reduce their e-commerce activities because of concerns over phishing and identity theft. Gartner believes this increased fear of theft and fraud via e-commerce channels could reduce total online commerce activity by one to three percentage points per year through 2007. The “Internet Sales Tax”: A Threat to All Direct Marketers? Another issue for Direct Marketers is the impending launch of the regulated charging and collection of sales taxes by online merchants for customers in other states, also known as the Internet Sales Tax. A 1992 United States Supreme Court decision prevented states from forcing a business to collect sales taxes from a customer unless the company had a physical presence in the state in which the customer resided. The ruling against the states was based on a vast array of differing jurisdictions, each with different tax rates on different items. However, the National Governors Association and the Streamlined Sales Tax Project brought states and local jurisdictions together to try and harmonize the tax rates across states to make it easier to get Congress to approve a law to circumvent the existing Supreme Court ruling and permit the states to collect taxes. As an official of Amazon.com stated in a September 2005 New York Times article,
Fall/Winter 2005 Direct Marketing Industry Report 19

“Certainly [we] at Amazon.com have no plans to volunteer [to collect the taxes].” So far 13 states have harmonized their sales tax structures (Indiana, Iowa, Kansas, Kentucky, Michigan, Minnesota, Nebraska, New Jersey, North Carolina, North Dakota, Oklahoma, South Dakota and West Virginia), and five more are approved to join the Streamlined Sales Tax Project within the next few years. The New York Times article stated that there is a possibility that the states could eventually attempt to seek back taxes for online purchases from merchants, and that any online merchant that registers to begin collecting the taxes will be granted an amnesty from a back tax grab. The states plan to start requesting (cajoling?) companies to begin collecting the sales taxes this fall. The harmonized state sales tax campaign has generally been focused in the media on the effect on online retailers, but the issue reaches throughout the entire Direct Marketing industry. DRTV advertisements, DR radio, catalogs, direct mail solicitations, online, magazines, newspapers, inserts — they would all require the collection of sales taxes for out of state customers. We presume there is price elasticity for goods first purchased through direct response channels (but believe continuity products have decreasing elasticity the longer the product is used and the more important such product is to the consumer). The introduction of sales tax, which increases the price of a purchased good, will reduce the demand for the product, and therefore is expected to reduce the total amount of goods sold through direct response channels. We also presume that the majority of Direct Marketing products sold are purchased by out-of-state consumers that currently do not pay a sales tax on the product. For residents of states with no sales tax on certain items like apparel, such as Pennsylvania and Delaware, the introduction of “harmonized” sales taxes, which force consumers in these states to pay sales tax of the other jurisdiction, is likely to cause those consumers to buy at home through local DR or bricks-and-mortar merchants. We believe this initiative could have a substantial negative effect on the overall growth rate of sales achieved through Direct Response channels (other than local stores or agents). According to the DMA’s 2003 Economic Impact—US Direct Marketing Today report, estimated Direct Marketing order sales (excluding orders from DM lead generation or DMdirected traffic to retail stores or local agents) were more than $640 billion in 2003 in the United States. A 5% sales tax on all of these purchases would have amounted to more than $32 billion in 2003 alone. This is an enormous pot of money not being captured by cash-starved states and local jurisdictions. It is very reasonable to assume that harmonized sales tax collection will arrive very soon. Moreover, imagine the potential legal costs and penalties imposed on the Direct Marketing industry if just one or two states came calling to Direct Marketers and wanted back taxes on previous sales to their states’ customers. Could it be time for Direct Marketers to relocate to a tax-free nation? Online Marketing According to Jupitermedia, approximately 68% of American households have Internet access at home, or about 77.5 million homes. The estimates of homes with online connectivity seem to have quite a wide range; a Nielsen//NetRatings survey in February 2004 found almost 75% of households were online at that time, but Jupitermedia beFall/Winter 2005 Direct Marketing Industry Report 20

lieves about 68% households are online now and that this number will only grow about six percentage points — to 74% — by 2010. There is expected to be an ongoing connectivity gap in terms of income (98% of households with income over $75,000 are expected to have online access at home by 2010, versus 43% of households with income under $15,000) and by ethnicity (in 2010, 64% of African-American and 64% of Hispanic households are projected to have home Internet access versus 83% of Asian homes). Age is also an issue; the 2004 Nielsen//NetRatings survey found 63% of homes inhabited only by people over age 55 had home Internet connectivity. A March 2005 report from the Pew Internet & American Life Project found that only 26% of seniors over the age of 65 reported having Internet access (up from 22% in 2004). About 57% of the over-65 respondents in the Pew survey had bought a product online, versus 72% of respondents aged 50-65.
United States Home Internet Access, February 2004
Total All 55+ All 18-24 All 2-17 Males 25-34 Females 25-34 Males 35-54 Females 35-54 50% 55% 60% 65% 70% 75% 80% 85%

Source: Nielsen//NetRatings, March 2004. The online access saturation of America has been overshadowed by a frenzy of demand for broadband access during the past couple of years. The FCC reported in July 2005 that the number of consumers and businesses that had broadband Internet service rose 34% in 2004 to nearly 38 million lines, with 5 million new cable subscribers (to 21.4 million) and 4.3 million new DSL lines (to 13.8 million). In the second half of 2004 alone, 5.4 million broadband subscribers were added. Even so, broadband penetration rates in the United States remain substantially lower than many other industrial nations, and its ranking is falling — from 11th in 2002 to 16th in 2004. Websiteoptimization.com has calculated that the current broadband penetration rate of all homes and businesses in the United States was about 28.7% in July 2005. An August 2005 report from the Free Press, and funded by the Consumers Union and the Consumer Federation of America, suggested that the concept of “broadband” in the United States remains weak and is far behind minimums available in other nations, and is also much more expensive on a throughput basis. The FCC uses a standard miniFall/Winter 2005 Direct Marketing Industry Report 21

mum of 200 kbps (kilobits per second) to determine if service is “broadband” or not. This is barely enough to receive low-quality streaming video feeds under optimal conditions. The report claimed that Americans pay, on average, 10 to 25 times more per megabit of throughput versus Japan, and that the “digital divide” is in full effect for broadband service; 60% of households with income in excess of $150,000 per year have broadband service, while 10% of households with income of less than $25,000 do. This is not surprising, given that the incremental cost per month for broadband is about $500 per year more than dial-up service, on average.

Source: International Telecommunications Union.

Moreover, the Pew Internet & American Life Project released a study in September 2005 that found that 53% of American households had broadband connection in the first six months of 2005, up from 50% in the previous six months — the slowest rate of growth since the Pew group began evaluating broadband penetration rates. In fact, both Comcast and Verizon had fewer net new subscriber additions in Q2 of 2005 than in the prior year period. The director of research for Pew stated to CNETnews.com that the “low-hanging fruit of early adopters is gone” and a Strategy Analytics researcher was quoted in the same CNETnews.com article as suggesting that aggressive pricing (low introductory pricing, or fees comparable to dial-up for slower broadband) would be
Fall/Winter 2005 Direct Marketing Industry Report 22

needed to drive the rest of the American population to move to cable or DSL or fiber Internet service. It’s obvious that offline homes aren’t seeing value in moving directly to broadband, since the Pew report found that only 23% of households that just received Internet service for the first time in the past year ordered broadband, while 77% ordered dial-up service. We think the Free Press report hits the nail on the head; people need to get more throughput for their broadband service dollar in order to sway more people from dial-up, especially when a lot of people already have broadband access at work. We provide these statistics because the level of penetration of broadband, and the speeds available for data transfer from broadband connections, are vitally important for the ongoing growth of online commerce. Web sites are becoming increasingly sophisticated, and are often graphics-heavy and focused at broadband users. If broadband access growth rates do not begin to accelerate even further, then a lot of potential customers could be turned off by slow loading times for websites. According to JupiterResearch, 34.7 million American households are expected to still have dial-up service at the end of 2005 (down from 42.7 million at the end of 2004). That being said, online commerce continues to be the shining star of Direct Response in the eyes of American consumers. Estimates for 2005 growth of online advertising (including search engine marketing) have ranged between 23% (Forrester) and 34% (eMarketer) over 2004. Forrester also estimates online advertising and marketing could account for 8% of total advertising spending by 2010, similar to current spending levels on cable TV and radio. Forrester released a projection in September 2005 that forecasted online sales (excluding travel) of $210 billion in 2010, or about 9% of all American retail sales, up from about $110 billion in 2005. Forrester and Shop.org have estimated that total e-commerce sales in the United States would be up 22% in 2005 to $172 billion, from $141 billion in 2004. Even with consumer confidence waning and disposable incomes shrinking as a result of Hurricane Katrina and higher energy prices, eMarketer announced in September that it expected online sales for the 2005 holiday season to rise almost 22% over 2004 levels, which would result in at least five consecutive years of 20+ percent annual growth in online holiday season purchases.

Source: eMarketer.com. Direct magazine’s 2005 online marketing survey of 130 companies found that respondents expected to spend 41% of their total marketing budgets for online campaigns, up from 25% in 2004. The primary spending areas were search engine marketing, analytics
Fall/Winter 2005 Direct Marketing Industry Report 23

and website customization services. Moreover, respondents were aggressively moving their budgets from offline to online channels; the offline channel portion of budgets are anticipated to decline from 75% to 59% on average in 2005. Almost 60% of respondents expected to increase their online budget again in 2006. Online sales were expected to be 25% of total revenues in 2005, versus 20% in 2004. Given the everincreasing demand for knowledge of return on marketing investment, and since online marketing tends to have quite a rapid response, we are surprised that the respondents expected only a 5 percentage point swing in total revenue mix coming from online sales after shifting 16 percentage points of their marketing budget to online campaigns. The article discussing the survey did not mention if the total marketing budget for the respondents was stable, rising or falling over 2004 levels. As a potential explanation, we point to a December 2004 study by Comscore, which found that in the consumer electronics and computers vertical, 92% of conversions that were initiated through online research by a consumer actually occurred offline. Consequently, the multichannel marketers may be seeing substantial total revenue growth as a result of their shift in marketing channel spend to online, but the primary conversion channel isn’t actually online. An Electronic Retailing Association survey in June 2004 found that 90% of respondents definitely or probably expected to buy at least something online in the future, versus 61% through DR radio and 50% via DRTV. A total of 89% of respondents that had purchased goods or services online rated their shopping experience as excellent or very good, versus 64% for DR radio and 61% via DRTV. Search engine marketing continues to be the 800-pound gorilla for the online marketing segment. JupiterResearch estimated that paid search engine marketing campaigns cost $2.6 billion in 2004, or about 31% of total online spending. In comparison, online display ad spending has been estimated by JupiterResearch to reach $5.1 billion in 2005, up about 25% from 2004. By 2010, JupiterResearch expects annual paid search advertising expenditures to rise to $7.5 billion. According to a press release from Nielsen//NetRatings for the week of August 21, 2005, paid search links accounted for 34% of 13 billion total online ad impressions (13 billion in a week!). The top five paid search link advertisers were eBay (379 million impressions), Shopping.com (138 million), InterActiveCorp. (98 million), MyCashNow.com (94 million) and Yahoo! (91 million). Only eBay had more than 5% of total ads (actually 8.6%), suggesting use of online links by a very wide breadth of marketers. As evidenced by the Nielsen//NetRatings numbers shown in the chart above, Google continues to grow its market share of all searches, which rose fractionally in Q2 of 2005 to 47.4%, far ahead of Yahoo! in second place at 21.8%. The launch of MSN’s new proprietary search engine failed to capture much new interest, as their market share declined by more than a percentage point to 12.4% (in a sector that grew 5% quarter-over-quarter overall).

Fall/Winter 2005 Direct Marketing Industry Report

24

Nielsen//NetRatings Search Engine Queries, Q1-Q2 2005:
Search Engine Google Yahoo! MSN AOL Ask Others Overall Q1 2005 (millions) 5,737,097 2,576,473 1,659,235 562,816 216,656 1,511,488 12,263,765 Q2 2005 (millions) 6,088,343 2,798,123 1,590,049 646,641 250,869 1,475,096 12,849,121 Change (millions) 351,246 221,650 -69,186 83,825 34,213 -36,392 585,356 % Change 6% 9% -4% 15% 16% -2% 5%

Source: Nielsen//NetRatings, www.searchenginewatch.com.
Fathom Online Keyword Price Index Average (Monthly)

$2.00 $1.90 $1.80 $1.70 $1.60 $1.50 $1.40 $1.30 $1.20 $1.10 $1.00 Sep- Oct- Nov- Dec- Jan- Feb- Mar- Apr- May- Jun- Jul- Aug04 04 04 04 05 05 05 05 05 05 05 05

Source: Fathom Online. According to Fathom Online, which tracks a price-per-click index of 500 generic keywords in eight broad product and service categories for the top five positions on selected search engines, the average keyword price fell for the second straight month in August and the third month in the past four. Prices dropped about 9% on average in August from July. The largest decline came in the area of mortgages, where related keyword pricing fell 41% month-over-month to $3.81 per click. Consumer services keywords fell 12% (to $0.99 on average) while retail and investing fell 6%. The only sector showing a month-over-month price increase was the automotive sector. Fathom Online attributed the declines to seasonality, as consumers do relatively less
Fall/Winter 2005 Direct Marketing Industry Report 25

shopping in the summer months. We think it also has to do with the end of the refinancing boom in mortgages (rates have risen). But it could also represent an extremely rapid maturation of the sector. We have noticed aggressive moves by Yahoo!, MSN, and Google to enter into overseas markets (China, India) with higher user growth rates, and Google recently raised more than $4 billion through a secondary stock offering — with the proceeds rumored to be earmarked for everything from IPTV systems to a Skypelike computer-based voice telephone network. Meanwhile, as of the date of writing, Microsoft was in negotiations to buy Time Warner’s stake in AOL and merge it into MSN, while Google also was rumored to be bidding for the ISP/portal. Permission-based or loyalty email marketing continues to be a means of reaching targeted customers in a cost-effective manner, even though people still hit the delete key much more often. According to the 2Q2005 DoubleClick Email Trend Report, the company’s clients (who used DoubleClick to send 2.5 billion emails in the single quarter) had an average email bounceback rate of 7.9%, down from 10.5% in the same period in 2004 and 13.7% in Q2 of 2002. While the average open rate fell to a new low for the survey’s history of 27.5%, down from 30.2% in Q1 and 36% in the same period in 2004, and the click-through rate fell from 7.7% to 7.2% year-over-year, the click-to-purchase rate rose again for the fifth straight quarter to 4.6% in Q2 (from 4.1% in Q1 and 3.6% in the prior year period). Marketers sent out about 385 emails in Q2 per eventual order received, which was similar to the prior quarter and a marked improvement from the roughly 450 emails sent per order in both Q1 and Q2 of 2004. Marketers saw their average order decline to $80 in Q2 of 2005 from $93 in the prior year period, but DoubleClick interprets this result as consumers purchasing slightly smaller orders but in greater frequency; we think this actually might represent net deflation in pricing in many of the top online purchasing categories over the past year (electronics, DVDs, travel). We also think the DoubleClick data suggests that there remains a significant need to have a clear, informative and comprehensive e-commerce site, and a pervasive presence on influential third-party sites, to woo consumers and complete a sale once they show up at the website. The 2005 DoubleClick Touchpoints III consumer survey of more than 2,000 consumers (from a Greenfield Online panel) found that websites (both company websites and third party sites) were by far the most influential channel for consumers’ decisions to purchase a product or service for travel, automobiles, telecommunications, banks and credit unions, mortgages and investments, electronics, and home products. TV ads influenced movie decisions the most, print ads and retail store evaluations influenced home care product decisions, and the consumers’ doctors had the only real effect on prescription drug purchases. The CAN-SPAM Act, which came into effect in early 2004 and was examined in more detail in our 2004 report, requires that commercial email senders provide recipients with a easily seen opt-out notice for future marketing emails and a link for such an opt-out request to be completed. In mid-2005, the Federal Trade Commission developed a list of 100 large online retailers to see if the companies were complying with the opt-out requirements. The FTC staff then signed in with three different emails to see if their opt-out requests would be honored. A total of 93% of companies honored at least one
Fall/Winter 2005 Direct Marketing Industry Report 26

request, and 89% honored all three. However, we don’t think the problem for opt-outs comes from large and relatively well-known and trusted merchants, but rather the optout for spam emails from unknown firms. We receive dozens of spam emails per week for products and services, and the last thing we want to do is click the opt-out link — because we think it just confirms an active email address for the spammer.

Revenues of Top 100 American Direct Retailers, by McKinsey Business Model (2003)
50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Efficiency Machines Niche Leaders Traffic Drivers Triple Plays Companies in Top 100 % of Total DM-based Revenues (all channels) % of Total e-Commerce Revenues

Other

Source: McKinsey & Co. An August 2005 report from the Center for Media Research highlighted a report from McKinsey & Co., which found that Direct Marketers with physical retail stores accounted for 52% of all online sales in 2003, while those without stores captured 31% of online sales. Online retailers captured $90 billion in revenues in 2004, up from $8 billion in 1998. McKinsey placed online merchants in four categories: 1: Efficiency machines: Retailers without physical stores that are primarily sellers of relatively low-margin products such as CDs, books and computers. Efficiency is required because the Internet provides global reach but these relatively undifferentiated products require a lot of brand marketing spend, effective e-commerce sites, and speedy and effective sourcing and fulfillment systems. The large fixed cost involved in developing an effective efficiency machine requires annual revenues of more than $750 million to be profitable. Of the top 100 online retailers, only seven are considered efficiency machines, including Dell and Amazon.com, but they account for a quarter of total online retail sales; 2: Niche leaders: Retailers without physical stores that are primarily sellers of higherpriced, high-margin products through a combination of Internet and catalog sales. Niche leaders build a loyal customer base by offering high quality products, great cusFall/Winter 2005 Direct Marketing Industry Report 27

tomer service, or both attributes. These firms don’t generally generate enough revenues to afford brand marketing campaigns, and focus on search engine marketing, affiliate, loyalty or prospect emails, or direct mail campaigns. There were 28 niche leaders in the top 100 online retailers, including L.L. Bean and jeweler Ross-Simons, accounting for $15 billion in annual revenues (6% of total online revenues); 3: Traffic Drivers: Retailers with a primary physical store chain with relatively low product margin and large scale who use the Internet to draw customers to their bricksand-mortar stores and to offer shoppers a wider selection of goods and greater convenience. Examples include Home Depot, Target, and Wal-Mart. The retail stores generally provide the majority of revenues for traffic drivers, but these companies are so big that their online sales represent 35% of total e-commerce sales; 4: Triple-Play Retailers: Store-based retailers who use a combination of stores, catalogs, and the Internet to maximize their share of total customer spending. These companies sell higher-margin products than traffic drivers (home goods, clothing) and attempt to truly achieve seamless multi-channel marketing. Catalogs attract new prospects, drive repeat orders (especially for products or sizes hard to find in the store); their Internet site provides ability to order quickly, comparison shop, and get short-term promotions; and stores allow for handling and trying out of products before buying them. Triple-play retailers, such as JC Penney and Williams-Sonoma, were estimated by McKinsey to have generated $6 billion in online revenues in 2003 or 17% of total online revenues.

Hispanic Marketing The “Hispanic” market has been one of the most talked-about segments of the direct response and advertising industries during the past few years. The US Census Bureau estimated that there were more than 41.3 million Americans of Hispanic or Latino origin in July 2004 (14.1% of the total United States population), up from 14.6 million in 1980 and 22.4 million in 1990. The Census Bureau stated in 2003 that half of the entire population increase in the nation from 2000 to 2003 came from growth in the Hispanic subset. A recent column by Barbara Tulipane, President and CEO of the Electronic Retailing Association, projected that the Hispanic market could grow to nearly 94 million people by 2025. It is also estimated that Hispanics will represent almost 67 million of the 138 million new members of the American population between 2000 and 2050. Tulipane claimed that the Hispanic market currently has a buying power of about $700 billion per year in the United States, and is projected to grow to more than $1 trillion by 2009. The $1 trillion number is pretty large, but it is still a relatively small piece of total American consumer spending. Moreover, Hispanic people can be captured through general advertising or direct marketing campaigns, so is it worthwhile to focus marketing resources on Spanish-language campaigns to reach this subset? The answer is increasingly yes, primarily because of the demographic anomalies of the Hispanic market. Of
Fall/Winter 2005 Direct Marketing Industry Report 28

the 41.3 million Hispanic people in the United States, more than 67% of those people live in just five states: California, Texas, Florida, New York and Illinois. The Center for Media Research reported in September 2005 that Texas, California, New Mexico and Hawaii were states with minorities as the majority of the population, and the majority of the “minority” population in Texas and California is Hispanic. The American Hispanic population is also resoundingly skewed towards younger people and towards urban areas: a March 2002 report by the Census Bureau found one in three Hispanics was under the age of 18 and more than half of all Hispanics lived within the center city portion of large urban areas. While Hispanics are now considered the largest visible minority group in the United States, they are also very highly concentrated in a small number of urban areas that are easily targeted. On August 29, 2005, it was reported by Nielsen Media Research that Univision’s WXTV station became the first Spanish-language TV station to win a sweeps rating period in prime time in the New York City market. WXTV won in the 18-49 and 25-54 demographics, and beat the second-place CBS affiliate by ten percentage points in the 18-49 group. The station also had the highest-rated prime-time show with 18-49 viewers, which was a soap opera called La Madrastra. The strong results are even more remarkable, considering that Hispanic households account for only about 24% of all households in the New York market. In addition, the station also had the highest-rated late evening newscast for the first time in August 2005. The Los Angeles market has also been a strong success for Univision, as its KMEX TV station has been the number one prime-time station in every reporting period since the Nielsen Local People Meters (LPMs) were introduced in July 2004 in that market. Interestingly, Univision actually sued Nielsen to try and block the LPMs from being introduced in the Los Angeles market, due to “flawed methodology” that was claimed to underrepresent the effect of Hispanic viewers versus the ratings collected via traditional paper diaries. In late August 2005, Nielsen Media Research released its most recent estimate of total American households possessing a TV, which was up to 110.2 million from 109.6 million in 20042005 (up 0.5%). Nielsen stated it also had expanded its count for certain minority groups (essentially increasing the effect of minority viewers), which led to an increase in estimated Asian TV households with a TV by 3.2% year-over-year, a 0.8% increase in African-American TV homes and a 2.9% increase in Hispanic TV homes. The Electronic Retailing Association claims that Hispanics respond much more effectively to Spanish-language media than English-language media, which seems likely. The growth in the Hispanic population each year is split about 50/50 between natural population growth (more births than deaths) and immigration. The high rate of immigration suggests to us that many Hispanics in the US are much more comfortable with Spanish-language media than English. According to a presentation by Michael Saray in 2004 (and who is currently the Chair of The DMA Directo: Council for Hispanic Marketing), 64% of Hispanics are more inclined to buy products if advertised in Spanish, and commercials in Spanish are 4.5 times more effective for the Hispanic population. A survey completed in June 2005 by research firm ICR of 1,001 Hispanic adults found that Hispanics were much more likely to have premium cable subscriptions than the national average (62% versus 46%). We interpret the results as a desire to have access to SpanFall/Winter 2005 Direct Marketing Industry Report 29

ish-language channels (including sports), which are mostly available only through premium-cable packages. A New York Times article on August 31, 2005, stated that all advertising aimed at Hispanic Americans accounted for about 3.5% of total advertising spending for a population comprising more than 14% of the national population. In other words, Hispanic people were not being targeted enough for advertising campaigns. Direct Marketing doesn’t seem to be doing much better in this regard. An August 2005 report from Multichannel Merchant suggested that Hispanic folks are “starving for someone to reach them.” The report included information from the 2005 DMA Hispanic Market Report that claimed only 55% of Hispanics receive six or more pieces of direct mail per week. The 2004 Saray presentation, as a corroboration, stated that Hispanics only receive about one-tenth of the direct mail deluge of the American population at large, and that “59% of Hispanics want more direct mail.” The DMA Hispanic Market Report also revealed that 65% of Hispanic survey respondents open each piece of direct mail received and either read or skim each one. We find the lack of mail unsurprising, particularly since the Hispanic population skews young, skews below the national average in years of education, average income, and live in center city urban areas (rather than suburbs). The large percentage of the Hispanic population that has arrived in the past five to ten years has been through immigration, which suggests these people have English as a second language, may have been illegal immigrants, and are not as likely to show up on a mailing list. The DMA Hispanic Response Rate survey for the 2005 report found that a majority of the respondents had been born outside of the United States. A lot of direct mail drops are for regional or national chains (in-store coupons), which may not have stores in center city areas. Perhaps a better contact method is via the online channel; the response rate survey found 82% of respondents had made a Web-based purchase in the past year (and 36% made six or more purchases). The respondents of the survey that identified themselves as purchasers through Direct Marketing campaigns in the previous twelve months had higher average income, much higher average education, and were either more likely to be born in the United States or have been in the United States for more than ten years. We interpret the data to mean that Direct Marketing channels are a learned phenomenon for immigrants; the longer the people are in the United States, and the more they assimilate into the American Englishlanguage consumer culture, the more likely they are to be receptive to Direct Marketing campaigns. Since the concept of Spanish-language marketing has only really taken off in a big way in the past decade, those Hispanic people most likely to have responded to a campaign are those that are already comfortable interacting with the English language media. However, since the Hispanic segment of the population is growing so much more rapidly than the overall American market, and so much of the Hispanic population is first-generation American, we expect the response rates of the Hispanic market to be very strongly skewed towards Spanish-language campaigns for the foreseeable future. But is a Spanish-language campaign universally essential? Probably not. It depends
Fall/Winter 2005 Direct Marketing Industry Report 30

largely on the demographic requirements for a successful sale of the product and the location of the campaign. We believe some Direct Marketers may prefer to capture people by zip code or demographic makeup (income, self-reported hobbies, past purchases), regardless of race or ethnicity. In the New York, Los Angeles, Chicago, Dallas, Houston, San Diego, Miami, and Phoenix markets, it makes a lot of sense to have Spanishlanguage campaigns since the percentage of the population self-reporting as Hispanic is substantially higher than the American national average. But Direct Marketers need to be careful not to lump all Hispanics together, because Spanish dialects differ throughout Latin America (Mexican Spanish is different from Cuban or Puerto Rican Spanish). So while one English-language campaign creative may work nationwide, regional Spanishlanguage variations may be needed for different regions of the United States. As a footnote regarding the WXTV ratings win, it was surmised in multiple media reports that WXTV showed first-run programming against repeats from the major networks. However, ICR did a survey of 832 American Hispanics in April 2005, and found that 44% of respondents had watched a Spanish-language telenovela in the prior week (53% of women, 35% of men; 51% of foreign-born Hispanics and 33% of American-born). The high rate of Hispanic population growth, combined with the significant demand for Spanish-language entertainment, suggests to us that advertisers and marketers in these urban areas may want to sit up and take notice.

Fall/Winter 2005 Direct Marketing Industry Report

31

META-TREND: NEW MEDIA CONVERGENCE
(Or “How I Grew Up and Learned to Love Streaming Video”) We have read an awful lot of online newsletters and articles discussing the end of the world for TV as a marketing channel due to the growth of digital video recorders (DVRs) including the early leader, TiVo. However, we also have seen an increasing amount of information regarding spending by media companies and by manufacturers and retailers on “new media” and offline/online marketing. The infrastructure continues to be in its nascent phase, and most Americans lack the capability to view streaming video on anything other than through their TV. But times are slowly beginning to change (with emphasis on slowly), and these new opportunities provide Direct Marketers with interesting new ways to capture prospects and entice customers back for more. In our 2004 white paper, we wrote a section on Stan Rapp’s view of the future for Direct Marketing, which revolved around “Direct Mass Marketing” that created a huge funnel (which we have coined the “Rappian funnel”) into which prospects were initially captured by large retailers or service providers through a wide variety of media channels that permitted much more rapid response by interested prospects. Mr. Rapp explained that new channels of marketing, such as online, email, interactive TV and cell phones, provide prospects with a means to become involved in the marketing campaign. But these new channels have not just been an evolution — they are flowing together into a revolutionary ectoplasm of wires, data packets, color and video. The common denominator is a better visual and auditory experience to bring the offer to life. We think definition of what we commonly know as “online” and “TV” will become an exercise in semantics, as the two media channels blend and merge. But there is also the expansion of higher-speed data transfer to the mobile/cellular phone. In combination with broadband data rates, multi-inch color screens, global position system (GPS) and other data tags, and the omnipresent cell phone in metropolitan areas, the opportunity for marketers is rapidly moving towards the ability to capture the attention of anyone, anywhere, any time, for any product or service. The only catch — and it’s a big one — is that you cannot do outbound “telemarketing” or unsolicited commercially-oriented short message service (SMS) text messages to a cell phone number. In the current TV environment, the biggest news for marketers has been the ongoing rise of cable, and the corresponding inability of the top four networks (NBC, ABC, CBS, and FOX) to recapture the viewing public even with blockbuster shows such as American Idol and Lost. The heart of the matter is that people just have more choices. In the 1950s, most urban areas homes received four channels. A strong show (I Love Lucy, Ed Sullivan) had the ability to capture an enormous percentage of TV households. Today, according to a 2005 survey by Magna Global USA, the average American household has more than 90 TV channels, double the number in 1995. We ourselves have Comcast cable service at home with the high-definition digital cable package which offers more than 250 channels and the opportunity to further purchase additional sports and pay-TV
Fall/Winter 2005 Direct Marketing Industry Report 32

packages that could drive the total available channels to more than 300. Flipping of a remote control during a commercial can take minutes to get through our full list of available channels, and it’s a decent chance that we will stop on some new channel that piques our interest. Cable channels are also fracturing to capture as many microsegments as possible. We don’t just get ESPN, but rather ESPN, ESPN2, ESPN Classic, ESPN HD, ESPN2 HD, and ESPNNEWS. Personal interests are also very strong pulls, and the essence of today’s TV environment is that a lot of the “personal interest” cable channels are outpulling traditional broadcast networks. TLC’s home renovation shows, as well as Spanish-language telenovelas, have been able to garner higher TV ratings in prime time than the broadcast networks. The growth of advertising in TV has primarily come from the launch of new cable channels. However, cable channel expansion has slowed substantially over the past two years (as our personal cable contracts can attest), and therefore the growth of advertising revenue on cable is slowing. Moreover, the “channel for every individual” TV universe actually makes it quite difficult for specific cable networks to grow their audience base. According to Magna Global, not a single cable network has been able to get above a 1.0 rating (1% of US TV households) in prime-time. Nine networks capture more than a 0.5 rating, up from six in 2001.
Top cable

Top broadcast shows – September 19-25, 2005 Source: Nielsen Media Research
Rank 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Show title CSI: Crime Scene Investigation Desperate Housewives Lost Criminal Minds CSI: Miami Grey's Anatomy Survivor: Guatemala Law & Order: SVU Network CBS ABC ABC CBS CBS ABC CBS NBC Viewers in millions 29.0 28.4 23.5 19.6 19.2 19.0 17.0 16.8 16.4 16.4 16.3 16.1 15.5 15.3 15.2

Top cable shows — September 19-25, 2005 Source: Nielsen Media Research
Rank 1. 2. 3. 3. 5. 6. 7. 8. 9. 9. 11. 11. 11. 14. 15. Show title Giants/ Chargers Nip/Tuck Hannity & Colmes LAX Landing Larry King Live WWE Raw WWE Raw Giants/ Saints O'Reilly Factor NFL Prime Time O'Reilly Factor On the Record O'Reilly Factor Hannity & Colmes Paula Zahn Now Network ESPN FX FNC FNC CNN Spike Spike ESPN FNC ESPN FNC FNC FNC FNC CNN Viewers in millions 9.8 5.3 5.2 5.2 4.9 4.7 4.6 4.2 4.1 4.0 3.7 3.7 3.7 3.6 3.4

Extreme Makeover: Home Edition ABC Invasion 60 Minutes Monday Night Football NCIS Destination: Lost My Name Is Earl CBS CBS ABC CBS ABC NBC

shows Sources:

USA Today, Nielsen Media Research.

The above television ratings tables, which represent dates around the start of the new fall prime-time season, reveal that the top cable series-based show (Nip/Tuck on FX) captured only about one-third of the 15th ranked broadcast show. Moreover, there was
Fall/Winter 2005 Direct Marketing Industry Report 33

not even another series-oriented show on cable that broke the top 15 cable shows. The cable results also show a strong interest in NFL football, and in news talk shows on Fox News and CNN (this time period also fell during the peak of interest in the devastation caused by Hurricanes Katrina and Rita). However, the convergence of phone, computer and TV — driven by advances in technology, broadband access and usage, and Internet Protocol technology — are expected to quickly transform the television and telephone into significantly different business models and provide enormous leaps ahead in marketing capabilities and opportunities. Digital Video Recorders — “You Don’t TiVo the Weather Channel” Our private equity clients have been interested in the Direct Response TV space, given the recent success of many DRTV buying agencies, but they are very concerned about the effect of digital video recorders (DVRs) on the viewing habits of Americans. We raised this topic to a couple of our friendly DRTV agencies, and one CEO promptly responded that DRTV revenues and demand were up sharply. He also noted that DRTV consumers tend to flip through the channels until something interesting catches their eye. The DRTV agency CEO suggested to us that it would be a decade or more before TV could be truly and adversely affected by DVR technology, and the effect may not be that big, after all. The CEO has history on his side. VCRs became an essential part of American households by the mid-1990s, but DRTV media spending and demand actually rose throughout that time period. Families that figured out how to program the VCR would often tape programs, but those shows tended to be either first-run and in prime time, or soap operas. There were still an increasing number of channels available to households, and each channel introduced more available space for DRTV commercials, so marketers had the opportunity to both expand their reach across more channels and target their message to certain demographics watching a specific channel. Is the DVR technology so much different from VCRs that the outcome will be different this time? A DVR is a digital TV recorder, with the show stored on either a computer-like hard drive or on recordable DVDs. They can be either standalone boxes or integrated into a cable box or DVD player. According to a Leichtman Research Group survey, approximately 8% of US households had a DVR as of June 2005. The most well-known and proliferated type of DVR is the TiVo box, sold by the eponymous company. TiVo introduced the technology which scares DRTV agencies the most: software that “learns” the viewer’s watching habits and both offers future alternatives to watch and automatically figures out programs to record that are similar in some way to what the viewer has recently watched. TiVo also introduced time-shifting, which would permit viewers to pause a real-time show (and start recording it) so that the viewer could come back from the kitchen or the washroom and continue watching as if the show were live. Since the program is recorded, the viewer could also skip over the commercials. TiVo has evolved from selling hardware to licensing its technology to electronics manufacturers, and have struck deals with satellite and cable providers to expand their DVR reach. As
Fall/Winter 2005 Direct Marketing Industry Report 34

of the end of the April 2005 quarter, TiVo had about 3.3 million subscribers (doubling in the past year), and rising about 300,000 per quarter. Such growth had come primarily from a deal with satellite pay-tv company DIRECTV; the satellite company’s TiVo subscribers numbered 2.1 million with 247,000 new subscribers in the quarter, while TiVo’s internally-generated subscriptions grew only 72,000 in the quarter to 1.2 million. Unfortunately for TiVo, DIRECTV has decided to go it alone in DVR technology and canceled its contract with TiVo; but a new agreement with Comcast (the nation’s largest cable provider with 21.5 million households) has renewed the company’s prospects. Comcast currently offers integrated DVRs in their Motorola high-definition set-top boxes, but beginning in mid-2006 will launch new set-top boxes with the TiVo DVR software included. TiVo is also working on TiVoToGo, which will permit subscribers to view TV content from Internet-based services, and permitting the transfer of TV shows to a laptop or other portable media player. As the company shifts away from being primarily a seller of its own hardware and software to a licensor of software, it needs to rapidly grow its base, because the software revenue generated via Comcast or DIRECTV is less than $2 per month per subscriber versus almost $9 per month in software revenue per internally-generated subscriber. The skipping of the commercials is what scares DRTV agencies and industry analysts the most. If viewers don’t watch the commercials, marketers will see a lower ROI on their DRTV spending and potentially reduce their demand (and certainly demand a lower media cost). The report that has led for calls for the end of DRTV as we know it is an April 2003 evaluation by TiVo and Starcom of 14,000 TiVo households. The study found that 77% of TiVo users would fast-forward through the commercial space during a recorded program. This percentage was extrapolated to the point that eventually nobody would watch TV in real-time, and nobody would watch commercials — branding or DRTV. In a subsequent study, Forrester evaluated about 600 TiVo users in mid-2004 and found that about 60% of the time spent on watching TV was spent on time-shifted or recorded programs, and that 92% of the respondents would fast-forward through the commercials on these time-shifted or recorded shows. Forrester also expected DVR penetration to grow to about 40% of US households by 2009. We think there are two very important aspects to this discussion that tend to be overlooked. First, the content that is most often TiVo’d generally includes first-run programming and tends to positively correlate with the rating of the show (so prime-time shows are TiVo’d more often). These shows are normally on the big networks or top cable stations and the advertising in these spots are usually expensive, national in scope, and reserved well in advance. These spots are rarely the home of DRTV campaigns, given the cost. DRTV, especially infomercials, usually take up remnant time — mornings, late night and weekend afternoons — and often purchased on a regional or local basis. Consequently, the biggest risks to DRTV from this point of view are (1) DVR users will watch a taped show and fast-forward through a DRTV spot, and (2) the DVR user will watch something saved on the DVR rather than a show with DRTV spots or an infomercial. The first risk is significant and increasing; TiVo will begin to introduce a service in 2006 to permit advertisers to put a logo on the screen when the subscriber fast-forwards through commercial time. Consequently, not only are the commercials
Fall/Winter 2005 Direct Marketing Industry Report 35

being skipped, but another “master” advertiser is overriding your commercial. The second risk appears to us to be an issue of product value and production quality; DRTV marketers need to successfully capture the viewer when the opportunity arises. According to a 2004 study by the American Association of Advertising Agencies and the Association of National Advertisers, which was released in September 2005, the audience falloff is about 5% on average when television programs move to commercials. About 4% of 15,585 programs evaluated lost more than 20% of their audience (some lost more than 50%) and 45% lost 5-20%. Syndicated shows tended to lose less viewers than first-run cable or broadcast shows. Syndicated shows tend to run in early evening and late night slots, suggesting that the remnant time available for short-form DRTV spots within these shows could have relatively more receptive viewers. The TiVo/Starcom study found another interesting factoid: while 77% of viewers would skip the commercials on shows they had previously taped, only 17% of viewers would skip commercials when they were watching a time-shifted or live program. We think this relates to a few factors. People like watching interesting commercials (look at the commotion around Super Bowl commercials), they also don’t really overthink the matter (it’s easier to leave the show running than to make the effort to time-shift and skip the commercials), and live programming doesn’t lend itself as well to recording. As Erwin Ephron recently stated in his newsletter (Ephron on Media), “[y]ou don’t TiVo The Weather Channel.” This makes a lot of sense to us; even though the timing of the outcome of a football game or the news or the weather really affects the viewer very little, people tend not to tape or time-shift these items. In our opinion, these are the time slots that could have the highest ROI for short DRTV spots, since there is a broad audience, the ability to target on a local or regional basis, and a large percentage of people will visit the network at least once per day. From a DRTV perspective, we would also say that “You don’t TiVo the Home Shopping Network.” The ability of other DRTV marketers to introduce some sort of time-sensitivity to the offer (as with offers provided on TV shopping channels) could significantly improve response rates. ID Media, one of the nation’s largest DRTV media buyers, recently did a study of DRTV performance by time slot, and found that daytime audiences were significantly more responsive than other dayparts (18% higher response on average). Cost of DRTV airtime for the daytime slots were also 7% less than the 24-hour average. Prime-time DRTV cost per call was 2.6x more than campaign slots run in the daytime. This corresponds very positively to the findings listed above that TiVo users generally record prime-time shows rather than daytime shows, so it can be rationalized that DRTV will remain as a viable medium for a long time outside of prime-time. Moreover, a 2004 study by the same agency found that 78% of DRTV media executives surveyed suggested that shifting DRTV campaigns to an infomercial format that integrates programming (edutainment) with product sell would keep people’s interest and be an effective response to ad-skipping. The 2003 TiVo/Starcom report also was completed at a point when TiVo had 703,000 subscribers (according to an SEC 10-Q document filed by TiVo in June 2004), or apFall/Winter 2005 Direct Marketing Industry Report 36

proximately 0.64% of all American households. In the world of consumer electronics, we would consider this subscriber base to be “early adopters.” Early adopters tend to have different habits and usage patterns than the average population, and so we think it may be premature to consider TiVo as the end of the DRTV world as we know it based on the usage patterns of early adopters two years ago. For example, in a 2005 study of 2,500 homes in Omaha, NE, by The PreTesting Company, approximately 22% of those TV households had DVRs. PreTesting reported that the DVR households did not skip through commercials or change the channel any more frequently than just the channelskipping done by non-DVR households. Ninety-seven percent of TV households have a remote control, according to PreTesting, and so they consider channel flipping to be more of a challenge to marketers on TV than the introduction of TiVo. In today’s 350channel universe, we would agree. We would be careful not to overstate the ability of DVRs to change the TV world when VCRs are already ubiquitous and still failed to do so. A Response magazine article from April 2005 discusses the future of DRTV, and suggests that broader interest in DRTV is the biggest risk to direct marketing success on TV, rather than the proliferation of DVRs. There are a large and rising number of Direct Marketers seeking success on television, so there is competition for a limited amount of airtime, driving up the media cost. In addition, some larger companies, such as retailers (Home Depot) and consumer product manufacturers (P&G) are using DRTV as a hybrid between branding and direct response. The article points out that these new entrants to DRTV use it to increase brand awareness and to test new products, and have discovered that the marketing ROI can be substantially higher than just a branding commercial. If Home Depot launches a DRTV campaign for a new tool or service, and they capture any DRTV revenue, it’s an immediate offset to the cost of the campaign — whereas a branding campaign brings no immediately identifiable revenue. These companies have high visibility and credibility, which raises the credibility of DRTV overall. However, they also are likely to be preferred by cable networks and time resellers over more traditional DRTV marketers competing for the same time slot. These hybrid DRTV marketers also increase demand for time, and since they aren’t driven by true campaign ROI as much as traditional DRTV marketers, the pricing model may squeeze out the more traditional DRTV spots. It will take a while for TiVo or other DVR/Video-on-Demand systems to have a meaningful aggregate effect on DRTV ROI. There is some evidence to support the abovementioned DRTV agency CEO’s hypothesis that DRTV is doing very well. According to TNS Media Intelligence, Q1-2005 short-form DRTV media billings rose 27% yearover-year, and 14 of 17 product/service categories experienced annual growth. The total number of unique DRTV campaigns in Q1-2005 was almost identical to the prior year period, suggesting that DRTV marketers are spending more per campaign. This would seem to corroborate the claims of rising DRTV campaign costs from the April 2005 Response article mentioned herein. TiVo is also attempting to offer its own marketing intelligence and delivery platforms. The company captures and retains the television watching ratings and patterns of subFall/Winter 2005 Direct Marketing Industry Report 37

scribers, and provides marketing intelligence to advertisers and networks in a similar vein to ACNielsen. According to an Advertising Age article in November 2004, TiVo and IRI joined up to launch a panel of 2,300 TiVo subscribers to evaluate DVR technology and its effects on advertising, with support from a “secret consortium” of consumer products companies. In addition, in August 2005, TiVo launched an “interactive” advertising solution for its Series2 (most recent) set-top boxes. Five companies — Ameriquest, E-Trade, Nautilus, Novartis and Johnson & Johnson — signed on as initial clients. The technology permits the advertisers to insert a response tag into their commercials, and when the viewer clicks on the tag, they get transported to a long-form commercial or a request for information page. These advertisements create “more visibility” in TiVo subscriber homes, whether in fast-forward mode or in real time. Earlier in 2005, TiVo also tested software to place banner ads on the screen from advertisers while subscribers fast-forwarded through taped or time-shifted content.

The Next Generation of Television — IPTV Distribution of television signals has evolved significantly from the original, over-theair transmission method. For the first 30 years of cable, operators had to pipe the signal through amplifiers along every 1,000 feet or so from the central office. Each reamplification of the signal would add noise to the signal and reduce picture quality, and many homes would wind up receiving a signal that had been reamplified more than 30 times on its way. If one of those amplifiers failed, then every home after that point on the cable chain would lost its signal. However, by the mid-1970s, the cable operators began using fiber-optic cables for the main delivery lines (the trunks) of the system, which dramatically improved the quality of the cable system and corresponding image quality. Introduction of satellite feeds to a local headend (which is the central signal distribution office for a neighborhood of about 10,000 people) also spawned a broad number of new national free and pay-tv channels (the first pay-tv channel was HBO, which started in Wilkes-Barre, PA in the mid 1970s). This “analog” cable system still requires a large number of demodulators and receivers within the local headend and has both high initial capital expense and substantial upkeep requirements. In the late 1980s, General Instruments (now part of Motorola) revealed its new “digital cable” system. Each analog cable channel takes up about 6 megahertz of spectrum. GI had figured out how to convert analog feeds to a digital signal (through MPEG compression) and squeeze about 10 channels into the same amount of bandwidth. Each cable system has about 550 megahertz of spectrum available, so the total number of potential channels was expanded to a theoretical 1,000 channels! The digital signal is sent via data packets to the home, and then reconverted back to an analog signal at the cable box. Error correction circuits in the cable box also permit digital picture improvement, resulting in a better overall picture quality than with analog cable — and digital cable also shifts the locking out of non-subscribers from scrambling to encryption. Encryption requires a “key” data string to break the code and release the available data feed. Cable systems have converted 30 million American households to digital cable. HowFall/Winter 2005 Direct Marketing Industry Report 38

ever, cable operators have faced very strong competition from national satellite providers, called direct broadcast satellite (DBS) systems, including DIRECTV and Dish Network. The DBS operators have doubled their number of subscribers nationwide since 2000 to more than 28 million by March 2005. Cable still maintains a wide lead in subscribers over DBS, but they have been significantly impacted by the rise of an alternative service that can provide specialized services (NFL Sunday Ticket on DIRECTV) and an lower average cost than cable. In addition, DBS has been very effective in capturing rural households with limited or no cable service access.
US TV Households, by Delivery Method (Millions)
Other (Antenna, No TV), 12.0

Satellite, 25.0

Analog Cable, 43.0

Digital Cable, 30.0

Source: TiVo, July 2005. To combat satellite, and to expand their arsenal of service provisions, cable operators have spent more than $100 billion to improve their network capacities. The infrastructure laid by the cable companies provides enough throughput to permit the use of the cable connection beyond the TV. Broadband cable Internet connections have grown to more than 22.5 million homes and offices as of the end of March 2005, from zero five years ago. Using a single 6 MHz cable channel, a cable modem can achieve throughput as high as 6 megabits per second (Mbps), as much as 30 times faster than DSL services provided by telephone companies. The next generation of cable technology protocol, called DOCSIS 3.0, is expected to provide download speeds as high as 200 Mbps — hundreds to thousands of times faster than basic DSL service. The advanced fiber infrastructure laid by the cable operators have now placed them in a highly advantageous position to capture data customers from the Bells and other ISPs. In addition, cable companies have embraced Voice-over-Internet Protocol (VoIP) technology to offer voice telephone service that almost completely circumvents the national “Bell System” telephone grid. According to a Merrill Lynch report in December 2004, Cablevision was adding more than 1,000 new VoIP phone subscribers per day across their network, roughly an 8% annualized take rate. The National Cable & TelecommuFall/Winter 2005 Direct Marketing Industry Report 39

nications Association, America’s cable operator industry association, states that there are at least 2.8 million VoIP phone subscribers nationwide, up from less than 200,000 in the beginning of 2000. The cable companies have therefore been able to offer a “triple play” of voice, Internet, and television — and offer a “bundle” of the Internet and telephone services at a significantly lower rate than the local telephone companies. The telephone companies enjoyed years of revenue growth from the rise of cellular phones, but the cellular phone subscriber base growth is slowing (more on this later). Facing a disadvantageous data throughput position, and significant pricing pressure and loss of customer base for home phones (not to mention cannibalization of home phones by cell phones as mobile pricing improves), the Bells have responded by announcing massive capital infrastructure programs to lay residential fiber-optic systems across the United States. Between just Verizon and SBC, more than $10 billion is to be spent in the next five years to bring throughput to the home that exceeds even cable. Verizon is spending approximately $12 billion per year to gradually upgrade to fiber-to-the-home (FTTH) delivery (or about 1-2 million homes annually). The Verizon FTTH network permits downstream Internet connection of up to 30 Mbps, far above the 1.5 Mbps of current DSL systems. SBC has earmarked $4 billion in spending by 2007 to lay a fiber-to-the-node (FTTN) network to serve 18 million customers, in which individual homes then have upgraded DSL (VDSL2) connections into the home. The per-home installed cost of the FTTN system is about one-fourth of the FTTH system, with an attempt to reach speeds of 20-25 Mbps. The regulatory environment is also improving for the Bells; in Texas, where Verizon first launched its FiOS IPTV service this fall, the state legislature has eliminated the need for IPTV operators to secure licenses from each local cable regulatory commission. The cable companies have used lawsuits against these commissions to delay competitors’ TV rollouts in other states in the past, and Texas’ smoothing of the regulatory landscape has eliminated such roadblocks in the second most populous state in the Union. But what to do with that new capacity? The phone companies believe they can maintain existing phone/Internet customers by upgrading to VoIP phones and ultra-high-speed data transfers, and bundle those services with wireless phone plans to offer their own “triple play.” But that’s not enough — they want to also offer cable-like television service through the phone/fiber-optic line, creating a “four-play.” Television through the phone line is all digital, using data packets. The system is called IPTV, or Internet Protocol Television. Within an IPTV environment, a headend is located in a neighborhood, and then video streams are sent out as demanded to each individual set-top, which may number one or more per household. This selection of programming by the household for IPTV channels is a “multi-casting” environment, quite different than the “broadcasting” environment of conventional cable. An adverse side effect is a longer delay when changing channels, since the information has to be beamed back to the headend rather than being done immediately within the cable box. A convention digital TV (standard definition TV) channel requires about 3 Mbps of dedicated throughput, which means that most DSL installations in the United States are currently
Fall/Winter 2005 Direct Marketing Industry Report 40

incapable of handling a real-time TV video feed. High-definition TV (HDTV) requires 19 Mbps, eliminating all but FTTH or FTTN systems for telcos to consider when launching a modern TV service — thus the billions to be spent on new infrastructure. The user-driven multi-cast nature of IPTV permits the IPTV network operators to radically improve their content protection capabilities. Conventional cable requires all the streams to be sent out, some encrypted and some clear, but IPTV permits the headend to determine if the requesting party has the appropriate subscriptions before sending out content. Providers of content, such as movie and television studios, have been concerned about pirating of the signal over an IPTV platform, because the signal is beamed out in an MPEG compression format and could be taped and distributed to millions of people for free over a Napster/BitTorrent-like system. Consequently, the IPTV hardware and software developers have had to create substantial end-to-end encryption to protect the feed until it reaches the set-top box (or preferably the TV). IPTV also permits television signals to be displayed on any monitor, be it a television, computer monitor or even cellular phone. However, the interaction between computer and TV tends to work much better on newer TVs. Older TV sets only received an interlaced picture, which means that every second horizontal line of the picture was refreshed per cycle. Computer monitors are progressive-scan, so that each line is refreshed in sequence per cycle. Interlaced TVs do a poor job at displaying computer text, which made early attempts at “WebTV” essentially doomed from the start. However, new HDTV-capable TVs are essentially large computer monitors, receiving and displaying information being driven from a set-top box. Consequently, IPTV viewers with progressive-scan TVs can surf the Net from their TV with clarity similar to their PC monitor (but much bigger). As for phones, we are currently working with an IPTV service provider in the southern United States that has signed multiple wireless carriers to be able to remotely stream a home subscriber’s IPTV network to a cellular phone or other portable media device. The current cellular data transfer speeds in the United States generally aren’t high enough to sustain TV, but next-generation systems probably can. Crown Castle, which owns and leases cell-phone towers to the network operators, is expected to begin building out “broadcast” antennas to exclusively drive video to cell phones before the end of 2005, with initial phonecasts within 18-24 months; Qualcomm (a maker of semiconductors for cell phones) expects to launch their own cell-phone broadcast network beginning in late 2006. So what does this mean for the marketer? Multi-channel marketing has already incorporated a mixture of offline and online capabilities, such as the listing of websites on DRTV or branding commercials to provide information and an e-commerce transactional opportunity. The use of IPTV integrates TV and the computer, since the same network infrastructure is used for both Internet access and TV. People’s usage of the Internet and watching TV also are substantially intertwined, suggesting that a seamlessly integrated IPTV/Internet system could be very useful. BIGresearch’s Simultaneous Media Survey in 2004 found that 60% of people focused on watching TV at a given
Fall/Winter 2005 Direct Marketing Industry Report 41

time also surfed the Internet at the same time, while 63.5% of people focused on using the Internet at home also had the TV on in the background. A broader delivery of video-on-demand (much easier in a multi-casting TV environment) is also expected to cause massive fracturing of the television audience. For example, less than four million viewers watched the real-time TV broadcasts of the Live 8 concerts this summer across three channels (VH1, MTV, and ABC), while five million watched the AOL online video stream of the concerts the same day and eight million more watched online over the next week. Napster and BitTorrent peer-to-peer downloading software have also permitted users to share and download TV shows to watch at their leisure. It is no surprise that the top-downloaded shows are aimed at younger viewers and tend to be “cult favorites,” as three of the top eight downloaded shows on BitTorrent in mid-July 2005 didn’t even break the top 100 TV ratings. It would be very interesting to see which shows are downloaded during the new fall season, and whether there is a spike in downloading of those shows (which would suggest that viewers are downloading rather than watching or TiVoing the shows). Top Downloaded TV Shows via BitTorrent, Week of July 12, 2005:
BitTorrent Rank 1 2 3 4 5 6 7 8 Show Downloads in Week of July 12, 2005 374,809 360,538 360,130 358,459 286,939 282,502 279,914 278,809 2004-2005 Nielsen Ranking 15 1 3 173 112 4 5 101

Lost American Idol CSI Smallville The O.C. Desperate Housewives Survivor Family Guy

Source: Wired Magazine (September 2005). We are seeing opportunities such as immediate purchase capability for a product shown on a TV show (the zenith of this was parodied in the movie The Truman Show), or to have the capability to immediately surf to a website upon seeing an offer and web address. The IPTV systems’ multi-casting setup also permits marketers to target on a oneto-one level or ZIP+4 grouping, and the information captured by the operator — including location, time of day, channel(s) watched, whether a subscriber watched a commercial or changed the channel — can all be fed back to the marketers for improvement of their campaigns. Two households watching the same show next door to each other could get totally different advertisements, with one household receiving gardening and
Fall/Winter 2005 Direct Marketing Industry Report 42

pet food advertisements and the next home receiving ads for sports equipment and diapers. Cable cannot do this because all channels in a local area are broadcast in an identical feed to all recipients. There has been some early discussion about video-on-demand systems, and how advertisers can immediately determine how many people are watching and from where, but IPTV takes the customer interaction to the next level. TiVo also can do some of this information collection at present – number of viewers, average ratings, limited interactive TV capabilities, time watched, time-shifted or not — but the introduction of IPTV may circumvent the need to have a TiVo-branded DVR device as the home set-top box could provide limited DVR functionality. In the future, the ability to drive an IPTV signal to a mobile device adds even more targeting and information gathering opportunities. Every new cell phone needs to have a location-based service for e-911 service, but the locator can be combined with IPTV on a mobile device to determine where the consumer is at the time of viewing. This gets a bit tricky, since it’s essentially illegal to do unsolicited advertisements, telemarketing or SMS messages to a cell phone, but if people opt-in for real-time coupons and offers, the IPTV system could send video advertisements embedded in the IPTV program that correspond to nearby merchants or services. Technology in Finland already permits people to purchase sodas from a vending machine using wireless phone signals, and SMS messages also provide coupon codes to consumers that can be inputted online or in a retail store for discounts or other promotions. Moreover, the introduction of ultra-high-speed wireless networks with broad footprints (such as WiMAX) could theoretically build a net for multi-service systems outside of cable or telcos. It was reported in late September 2005 that Google is in the midst of beta-testing their new WiFi network service for launch in the San Francisco area. Google could, theoretically, install a WiMAX mesh network in major urban areas to provide free cable-like wireless speeds (3 Mbps) and low-cost high-speed (20+ Mbps) within a 10 mile radius from a single WiMAX tower. Nokia and other companies have already developed WiFi-based cordless and mobile phones, which seamlessly switch off and on the WiFi network as availability permits. Such speeds would also permit a WiMAX-enabled phone or home to enjoy IPTV delivery. The introduction of these alternative television delivery systems will mean additional choice of service provider for the household, and since each system basically offers the same content and channels, the differential likely will come down to price. Great for the consumer, not so great for the cable and satellite network operators. Price could also be affected by the amount of advertising or type of advertising that can be introduced. A lot of speculation related to this again centers around Google; they have been developing a Google Video search capability. Google — or any other search engine company — can offer a massive video-on-demand index to IPTV subscribers, with paid search results or advertising banners attached to the search results. The introduction of advertising attached to the search results could be a significant revenue generator for the IPTV network, as well as a new targeted marketing channel. The paid search results could be based on psychographic, locational, and demographic information of the searcher as well as the type of video requested.
Fall/Winter 2005 Direct Marketing Industry Report 43

IPTV still has some teething pains. Integration and installation remain stumbling blocks to large-scale rollouts of the systems. Telstra, the national telephone company in Australia, has canceled two IPTV trials using Siemens and Microsoft systems, and a Belgian trial was also cut short as performance was relatively limited. The US Bells’ IPTV systems have been using a wide variety of components from different companies, and there have been problems in making the whole system as user-friendly and effective as cable or satellite. IPTV lines are very susceptible to electronic interference, and can require substantial installation work to shield the lines inside the home. Some early installations have had problems with electrical noise coming from other AC motors within the home (exercise equipment, appliances) or from electrical storms, and some also get fuzzy screens when the phone rings. So it’s a work in progress, but we strongly believe IPTV will be a significant component of the TV universe by the end of the decade. Pay-per-Call Telemarketing, both outbound and inbound, has been a highly effective tool for selling products in large part because of the human interaction between buyer and seller’s representative. The agent can answer questions, coax an uncertain buyer, and upsell a caller to more (or just more expensive) products or services. Inbound phone calls to a business are also highly qualified sales leads, since the customer has sought out a phone number from the Yellow Pages or other advertising and made the conscious effort to call the company. A new online paid search system, called pay-per-call, is an intriguing mixture of inbound or outbound sales calls and pay-per-click paid search ads that provides marketers with the ability to drive online searchers to a live sales agent to close the sale. The first generation of pay-per-call has been launched by AOL and LookSmart. On these sites, the marketer determines which keywords should trigger the appearance of his or her ad, then creates an advertisement, and bids for ranking of the ad on each keyword search results page. When the searcher arrives at the search results, he or she can click on the ad — which drives to an information page with the company’s information and local or toll-free number. The number is usually a special number or a code so that the search engine and the marketer can track the source of incoming calls and pay accordingly. Other systems permit the searcher to input their own phone number and be contacted by a sales agent. The company then pays the search engine company for each new lead (as identified by a unique customer phone number). The concept is very intriguing for direct marketers, especially those with regional agentbased or franchise-based networks such as insurers or real estate agents. If a person is actively searching for a product or service, then it is logical to presume that connection to a live salesperson could significantly increase the sales completion rate and provide solicitations for immediate upsells and cross-sells, just like incoming calls driven to a contact center by DRTV. Moreover, the use of the online medium creates a massive Rappian sales funnel driving leads to closers. For companies not engaged in ecommerce, the pay-per-call system also provides a way to immediately drive an interFall/Winter 2005 Direct Marketing Industry Report 44

ested party to their business. Most small businesses do not engage in any sort of paid search advertising online, and a significant proportion want to use local search the same way as they use the Yellow Pages, giving the consumers the option to call or visit the storefront (according to Findwhat.com’s Vice President of Sales in a December 2004 article in searchenginewatch.com). Pricing averages $6-10 per call range on average at present, but proponents of pay-per-call claim a significantly higher conversion rate than pay-per-click so that the average cost per sale is similar. The concept appears to have legs both at a large scale and at the small business level. A Direct article in late August 2005 announced that “local search” website Superpages.com (which combines user-inputted locational information with the keyword search request), owned and operated by Verizon, will begin to offer pay-per-call for both national and local advertisers on its site by late September. The local advertisers don’t even need to have a toll-free number, for Verizon provides the option for the small business to simply have local calls routed to their main business number. We believe pay-per-call is also valuable to service providers, such as our own investment bank, that have a highly targeted audience. Keystone Equities spends about $100 per month on Google paid search, using keywords primarily associated with direct marketing and marketing services. We use it to gain exposure to executives seeking an investment bank to raise capital, or use our consulting and M&A advisory services. However, we have no e-commerce platform and our sales cycle requires a significant amount of human interaction before we sign up a client. Consequently, our Google ad drives searchers to our website, where they can email us or phone our 1-800 number. But the percentage of executives to total direct marketing searchers that click through to our site and consequently contact us about our services is very, very small. Our lead generation from the Google ad is very small on a percentage basis. Keystone Equities receives about 40,000 ad impressions per month, but only about 100-200 clickthroughs — and perhaps one to three contacts from interested parties that could be considered qualified leads. Given our monthly Google paid search cost of about $100 per month, our average cost per lead is between $33 and $100. For Keystone, one lead turned into a sale covers the Google cost for years, but it effectively is more of a “branding” tool than a direct marketing tool. If we had a pay-per-call program in place, the landing page for the keyword would provide substantially more information, and our cost per lead may fall to $10-20. Consequently, pay-per-call for Keystone is likely to be a more useful sales creation tool than Google paid search. We plan to try the system in the near future and we will update our progress in future reports. The next generation of pay-per-call could dramatically improve the speed and effectiveness of the system, driven by the widening use of broadband-based VoIP voice connections. Websites have already introduced link buttons to call centers or to instant messaging ports to provide a human touch to the online world. Future pay-per-call could modify search engine results to provide an immediate computer-based voice link to a live salesperson. The advertiser would pay a fee for each “call” connected to a salesperson.
Fall/Winter 2005 Direct Marketing Industry Report 45

The basic infrastructure already exists, thanks to VoIP technology that we discussed above. Recently, eBay acquired the European VoIP call-by-computer (or “soft phone”) technology company Skype (rhymes with “hype”) for $2.6 billion, a lot of money for a firm with approximately $60 million in annual revenues and not yet making a profit. However, Skype has had more than 54 million subscribers and 170 million downloads of its software that permits people to make free voice calls over the Internet to other Skype users. More importantly, Skype also offers extremely low-cost outbound calling worldwide ($0.15 per minute to India from the United States, $0.02 domestic, $0.02 to the UK) to regular phones, and accepts inbound calls as well — without the monthly charge associated with Vonage or other hardware-based VoIP phone plans. The large online players have almost all acquired at least one “soft phone” company. eBay believes it can both integrate the Skype software into its auction system (to freely connect buyers and sellers) as well as provide more business for its PayPal online payment system, which can be used for the paying of Skype bills. Google and Yahoo! can introduce pay-per-call technology, as well as integrate free calling in return for viewing an online ad. The introduction of WiFi phones also permits these entities to offer such capabilities after hearing a short audio ad or seeing a short ad on the phone’s screen. Again, there are a lot of new targeted marketing opportunities created from technological changes to the traditional media channels. The Online/Offline Print Synergy — Environmentalists Rejoice! Even print media is feeling the positive effect of “synergistic” media channels (to paraphrase a line from a caricature of News Corp.’s Rupert Murdoch in the recent movie In Good Company). The Newspaper Association of America recently reported that total newspaper ad revenues (print and online) were expected to be $12.2 billion in the second quarter of 2005 — annual growth of 2.8%, or a bit slower than overall economic growth for the nation. Print ads rose 1.9% from the prior year to $11.7 billion, while online ad revenue for newspapers rose 28.6% over the same period to $501 million (4% of the total). In other words, the incremental increase from the online ad revenue was almost half of the growth of the print ad revenue, but online revenue was only 4% of the total sales for these newspapers. The North American newspaper market is dominated by a relatively few number of large companies, such as Gannett, Knight-Ridder, and Tribune Co. Just as these firms consolidated the print newspaper industry and centralized copy and ad sales to improve the bottom line, they also have been able to have a network of news sites that have rapidly refreshing information of great importance (both on national/global and local levels) to individual viewers. Again, centralized and targeted ad sales (by geographic region, web site news section or article, or national) can be accomplished. The promoters of Internet advertising had the concept right — the more eyeballs, the better — but the old-school newspaper giants appear to be in the right place to serve the “local” need that has yet to be well-served by Google or Yahoo!. These print media giants have also been spending money on bulking up their online capabilities, as evidenced by the recent purchase by United Business Media of Light Reading and TechOnLine; News Corp’s acFall/Winter 2005 Direct Marketing Industry Report 46

quisition of Scout Media, IGN, and Intermix; Gannett’s purchase of PointRoll; New York Times Co.’s acquisition of About.com for $410 million (10x revenues); Dow Jones’ acquisition of MarketWatch.com; and the Washington Post Co.’s purchase of online magazine, Slate. In addition, CNN and Time Inc. announced in September 2005 that they would merge their online properties under a single “CNNMoney.com” homepage, in hopes of offering users a broader range of related content under a single site. Of course, marketers will be able to buy across the entire site. A June 2004 study by the Pew Research Center for the People and the Press estimated that 29% of Americans went online at least three times per week to obtain news (up from 23% in the 2000 study). Fifty percent of respondents who were college graduates did so in 2004 (and 74% of college graduates 18-39), up from 40% in 2000. The education gap is substantial here, as only 18% of high school graduates and 8% of high school dropouts regularly surfed the Net for news. Age is also a profound influence. Thirty-six percent of 18-49 year olds obtained news online regularly, versus only 8% of people over 65 (and that 8% number has been steady since 2000). Differences in online news sleuthing by ethnic group shrunk substantially from the 2000 to the 2004 Pew studies. Hispanic respondents using the Net for news rose from 21% in 2000 to 32% in 2004 (outpacing Caucasians at 29% and African-Americans at 25%). We wonder if there is a multi-channel marketing opportunity here between news sites and Hispanic TV, or if the Hispanic people prefer to access English-language media. The survey also found that people spent far less time using the Internet for news than other media, a fact that we think represents a more effective and efficient use of time for people. For example, we can access the Weather Channel online immediately to get tomorrow’s forecast, but we have to wait up to ten minutes to get a local forecast on the Weather Channel TV station and up to 20 minutes or more for a forecast on a local affiliate of a national TV network. A BURST! survey of 6,400 online users 14 and over in the summer of 2005 reported that 62% of respondents were seeking news stories when they go online, followed by 35% seeking entertainment content. A large number of magazines and newspapers are beginning to charge web-based subscriptions to their sites, which combine real-time updates and the columns and in-depth articles from the print versions. We subscribe to several web editions instead of print versions, including the Wall Street Journal and Canada’s Globe and Mail, largely because it is more convenient and more focused, and these editions provide the opportunity to read and search archived material. We may not read today’s paper and may miss out on a good article, but if we subscribe to the online version we can go back and read it at a later date. Zinio, a San Francisco-based digital publisher, accounts for more than 62% of all digital magazines including BusinessWeek, PC Magazine and Playboy. Zinio reported that their estimate of the total number of digital magazines in the United States rose more than 33% from June to December of 2004 alone, and circulation rose almost 26% in the same period. Approximately 80% of the total digital titles are businessoriented, but the 20% share of consumer titles accounts for 45% of total digital circulation. Zinio alone delivers more than 40 million copies of over 300 different digital magazines per month, to a paid subscriber base numbering more than two million in over 190 countries.
Fall/Winter 2005 Direct Marketing Industry Report 47

Why shift to an online version? In the June 2005 edition of The Advertiser, Ford Motor’s Global Media Manager (and Chair of the ANA Print Advertising Committee) suggests that the evolution of magazines into a personally-controlled compilation of articles and information from a wide variety of sources — essentially an online magazine version of the iTunes success story. The ability to offer an online version can expand the exposure and reach of a magazine beyond its print readership, partially because the articles can be picked up by other entities and redistributed throughout the Internet news network, and also because the electronic articles can be copied and forwarded to friends, family and other interested parties, which is essentially consumer-driven viral marketing. In addition, a material portion of total printing and distribution costs for publishers is for discount subscriptions or free copies. These lower-margin customers or prospects can be converted to digital subscribers, and for whom “printing” and “delivery” become nearly costless. Marketers may be skeptical at the concept of paying an advertising rate based on a compilation of online and print readers, since online formats may be less conducive to capturing the online reader’s attention, and there could be competition on the page from banner ads or paid-search ads. However, Zinio has reported a small but rapidly growing number of interactive advertisements incorporated into their digital magazines that can target specific subscribers and bring a more interesting experience to print advertisements.

Fall/Winter 2005 Direct Marketing Industry Report

48

ABOUT THE KEYSTONE EQUITIES GROUP
Based in suburban Philadelphia, The Keystone Equities Group is a full-service brokerdealer with institutional trading, sales, and investment banking capabilities. Founded in 2003 by several seasoned brokerage principals, we serve an institutional and corporate client base seeking information and transaction services focused on small-cap and midcap companies in our industry knowledge sectors. Our Investment Banking department specializes in Direct Marketing companies, capitalizing on more than 40 years of combined Direct Marketing industry experience.

The Marketing Services Investment Banking Team
John Harrison is a founding Partner of The Keystone Equities Group. John was President of DiMark for 15 years, and led the company’s growth to the sixth-largest Direct Marketing agency in the world before overseeing the sale of DiMark to Harte-Hanks in April 1996. He also held senior Direct Marketing management positions with CUNA Mutual, RLI Insurance and CNA Insurance. John is on the Board of Directors and Past Chairman of the Professional Insurance Marketing Association and is on the Advisory Board of DePaul University’s Interactive and Direct Marketing Institute. He sits on the Boards of The Credo Group, a digital insurance agency; Authtel, a marketing compliance and prospecting technology company; Solutionary, a 24x7 security program management company (SPMC); and IXI Corporation, a database marketing company that uses proprietary wealth management techniques. John is also an active Member of NAPES and MI2, associations that serve the Worksite Marketing industry. Jack Freeman is a founding Partner of The Keystone Equities Group. He was previously Senior Vice President, Capital Markets, for Emerging Growth Equities, a Pennsylvania-based investment bank, where he focused on serving Direct Marketing firms. Prior to EGE, Jack was President & CEO of CCM Consulting, a boutique M&A advisory firm for Direct Marketing companies. His Direct Marketing industry experience includes Director of Investor Relations and Strategic Planning and Development at DiMark from 1992 to 1996, and Director of Investor Relations for ADVANTA from 1990 to 1992. Jack is also on the Board of Directors of PetFoodDirect, a Direct Marketing firm specializing in pet food, medicines, and supplies. Noel Atkinson, CFA joined The Keystone Equities Group in August 2003, and is Keystone’s senior analyst. He has more than seven years experience in investment analysis on both the buy-side and sell-side, covering equities, bonds, and derivative instruments. This report is the fourth authored by Noel to support the industry exposure of Keystone in the Direct Marketing sector. From 2001 until 2003, Noel was a sell-side institutional Research Analyst for Emerging Growth Equities. He has also held financial and economic research positions for a international bond and currency fund and in the management consultancy field. Noel is a CFA Charterholder and a member of both the CFA Institute and the Financial Analysts of Philadelphia. He holds an M.A. (International Affairs) from Carleton University, Ottawa, Canada.
Fall/Winter 2005 Direct Marketing Industry Report 49

DISCLOSURES
The Keystone Equities Group (“the Firm”) is engaged in the investment banking business and seeks investment banking assignments from a wide range of companies, including the companies identified in this white paper. Therefore, readers of this white paper should assume the Firm intends to seek compensation for investment banking services within the next three months, and thereafter, from the companies identified within this white paper. This white paper is intended for informational purposes only. Inclusion of companies in this white paper is not intended to be, nor should it be construed as, an investment recommendation or a solicitation to purchase securities of those companies. In our view, the information provided herein about companies whose securities are publicly traded is not information sufficient upon which to base an investment decision. The Firm does not issue sell-side research coverage, or issue investment opinions, on stocks, indices, or industries. Readers who are considering an investment in any company identified in this white paper should obtain additional information before making an investment decision.
This publication is neither an offer to sell nor a solicitation to buy any securities mentioned herein. The information contained herein is based on data obtained from recognized sources that are believed to be reliable. The Keystone Equities Group has not independently verified the facts, assumptions, and estimates contained in this white paper. Accordingly, no representation or warranty, expressed or implied, is made as to, and no reliance should be placed on, the fairness, accuracy, completeness, or correctness of the information and opinions contained in this white paper. The information contained in this white paper is not and does not purport to be a complete analysis of every material fact representing any company, industry or security. Any statements nonfactual in nature herein constitute only current opinions or estimates, represent only the current judgment of the author(s), and are subject to change without notice. The Keystone Equities Group (or one of its affiliates) or their partners, officers, directors, analysts, employees or customers may have an interest in the securities mentioned herein and may make purchases or sales as principal or agent of these securities, or their derivatives, while this white paper is in circulation. © Copyright 2005 The Keystone Equities Group, L.P. All rights reserved.

Fall/Winter 2005 Direct Marketing Industry Report

50

The Keystone Equities Group 1003 Egypt Road Oaks, PA 19456-1155 (800) 715-9905

www.keystoneequities.com

Member NASD & SIPC • MSRB Registrant