Business Intelligence in the Cloud: A GoodData Platform Overview
The GoodData Platform is a portfolio of tools, APIs, and best practices. Designed to deliver informationas-a-service to business users, the GoodData Platform enables IT to take a leadership role in enterprise BI.
The GoodData Platform: An Overview
GoodData allows us to track global KPIs for our consulting business in one place—trend them over time, and compare regional performance. GoodData is my source of truth.
Ivo Totev Chief Services Officer Software AG
GoodData enables IT to deliver high-value Business Intelligence (BI) services to the business —and create true competitive differentiation in the marketplace. In this paper, the basic features and functionality of the GoodData Platform are introduced. With pre-built analytics that measure the most significant metrics about customer interactions —including trending, waterfall, conversion, cohort and segmentation analyses—the GoodData Platform drives IT productivity and effectiveness. and explained.
The GoodData Platform: Open for Business
As an end-to-end BI platform that lives in the cloud, the GoodData Platform frees up IT to focus on the strategic use of all data including sales, marketing, social, and customer service data, and less on the underlying mechanics required to keep the BI application functional. The GoodData Platform supports IT in three ways: The GoodData Platform, with its security mechanisms, backup guarantees, and data privacy safeguards, is a fundamental building block of any IT governance plan.
Nurtures a “customer obsessed” culture Optimized for customer and partner interactions, the GoodData Platform gives you information-as-a-service that empowers you to approach, sell, and support your customers more effectively. This includes social, sales, traditional marketing support data, and human resource (HR) and finance data —anything data. Any data source. Delivers speed to value The GoodData Platform helps companies accelerate business results with on-demand analytics that enable speedy decision-making. Get up and running in just days with a platform designed for the world of software-asa-service (SaaS) applications and social media, where data changes constantly and analytics require maximum agility and speed. It gives business users real-time access and ad hoc reporting on internal and external data. And GoodData is proven to deliver 99.5 percent uptime and 84.5 percent user adoption. Translation: Deploy GoodData, and it will be always on—and always used. Enables velocity of innovation The GoodData Platform helps you build a culture of innovation with its ability to mobilize your company around innovative processes, products, and services. The open architecture of the GoodData Platform makes it easy to embed GoodData into existing applications or create new applications for your users. Address new requirements by adding data sources, transformations, data models, metrics and users. Create data mashups that correlate metrics across multiple data sources and answer questions fast in response to dynamic market conditions. And as new data sources are created, effortlessly integrate them into your BI workflow. 3
Benefits of the GoodData Platform
GoodData has helped more than 90,000 users make sense of their customer-facing data with advanced business analytics.
It’s open Thanks to its open APIs, the GoodData Platform makes it easy to integrate new data sources, change data transformations, modify the logical data model, and update metrics, reports and dashboards. All changes are instantly pushed out to users, who can use representational state transfer (REST) APIs to automate and integrate provisioning and analytical tasks. Users can also create new applications and mash-ups on the fly based on their business needs. It’s complete As an end-to-end solution, the GoodData Platform eliminates the need to buy and integrate different products for data loading (ETL), data warehousing, analytical engine, data visualization, and collaboration. The GoodData Platform does it all. It scales The GoodData Platform is a multitenant platform that runs all operations as stateless services distributed across hundreds of computers. Because of this, the GoodData Platform can scale to meet the demands of any business. Currently supporting more than a million data integrations per week, the GoodData computing fabric tracks and balances all loads, and the GoodData Platform services are partitioned to allow each service to be used independently. With a choice of Amazon Web Services™ (AWS) and RackSpace™/OpenStack™ offering a choice of public or private cloud as storage for their data and transactions, businesses have the ability to quickly deploy additional hardware to handle additional traffic or data volumes. It’s fast As GoodData users push the limits of the service, GoodData continues to optimize the platform to benefit all customers. Additionally, analytical query language abstraction enables the use of the database engine for optimal performance.
GoodData gives us an opportunity to have our clients aggregate data from different sources providing us more data to analyze and draw deeper insights into more parts of their business… [W]e can do in minutes what before we’d have to do in programming that would take months.
Robert Honeycutt CEO Market Metrix
It’s easy to manage Every management and administrative function is available via API. All these functions can be fully integrated with third-party administrative/maintenance solutions as well as other third-party applications. And because the GoodData Platform is partitioned into domains, its projects and user spaces never conflict. The GoodData Platform also enables dynamic provisioning of new users, projects, and permissions. The fact that it’s fully automated means that creating new data marts and data warehouses can be done in seconds. Finally, the GoodData Platform’s mass-management capa-
bilities mean projects can be derived from templates, and analytical objects like dashboards, reports, and metrics can be transferred between multiple projects and upgraded or updated with one command. It’s built on leading technologies Designed for the cloud from Day 1, the GoodData Platform is built on leading-edge technologies to ensure dependable performance, integration, and efficiency. Technologies such as Vertica, MongoDB, Cassandra, NetApp, and Rackspace are all integrated into the GoodData Platform.
The GoodData Platform Architecture
The GoodData Platform encompasses all the key components of a BI solution: load, store, analyze, visualize, and share. All were built from the ground up to exist in the cloud and be delivered as an end-to-end service. However, they also can be used as independent modules and integrated into BI applications an organization has already implemented.
Slice & Dice Real-time Best Practices
y ilit lab ility a Sc liab ty Re curi Se
SE OU EH AR W
Se lfse Int rv er Cu ice sto acti v m iza ity tio n
US ER IN TE RF AC
EXTRAC T, TR A LOAD NSFO RM (ET L) ,
E OR ST
VIS UA LIZ E
s ation tific No bility Mo her oget ng T nni Wi
Conn ect Sche dule Mon itor
ATE BOR LLA CO
Load: Powerful Out-of-the-Box ETL Capability
In this first stage, the GoodData Platform provides for built-in connectors to extract data from any data source —both on-premise databases, and cloud applications and databases —structured and unstructured data alike. All this is available through the GoodData Data Integration Service, a key differentiating technology from GoodData that enables visual development of logical data models, and provides extract, transform, and load (ETL) functionality. The GoodData Data Integration Service connects to any application, application program interface (API), database, or file system, and recognizes many different data formats (see sidebar). For example, the GoodData Platform can connect to a variety of sources, thanks to its generic readers such as REST, simple object access protocol (SOAP), comma separated value (CSV), and Java database connectivity (JDBC). For organizations with data stored in “big data” applications, the GoodData Platform works with Hadoop, Hive, Cassandra, and HBase.
Data Sources the GoodData Platform Has Connected To:
Salesforce (all clouds) Salesforce Chatter Zendesk Twitter Facebook SAP Google Analytics YouTube Google AdSense Google Docs ExactTarget Marketo Eloqua Netsuite Jira GitHub Platform Overview Yammer Pinterest Box.net MediaMind Scup Zuora Pivotal Tracker MindBody Harvest BrightIdea Coupa GetSatisfaction Pardot Sugar CRM Microsoft Dynamics Oracle Twilio Genius GreatVines Aurix Flurry AppFigures Campaign Monitor Amazon AWS Chargify QuickBooks Keynote Magento New Relic SendGrid and more
The GoodData Data Integration Service uses CloverETL, an open source data integration tool based upon Java, to create a flexible data transformation platform. This platform consists of two separate environments: a data transformation development environment with an attractive visual interface, and a remote execution environment. The GoodData Data Integration Service
performs more than two dozen prebuilt transformations on data, including deduplicate, denormalize, aggregate, and others. Users can also write their own customized transformations using the GoodData Integration Service. Once the data is transformed, it is loaded into a GoodData project, again using a visual tool, and connected in a logical flow.
The logical data model piece of the GoodData Data Integration Service allows the business to define the entities as well as the relationships —for example, combining data from Facebook campaigns with CRM data from Salesforce. All data models can be edited on the fly using the GoodData Data Integration Service tool. This completely eliminates the need to do low-level coding, and because the GoodData Platform has pre-built connectors. It also eliminates the
need to perform integrations using third-party tools--however, customers have the flexibility to use their ETL tool of choice. Visual and easy-to-use, this allows even non-technical users to design complex flows, and advanced transformations are performed right out of the box. For scheduling loads, the GoodData Platform possesses a data-loading service console that lets users choose the frequency for scheduling loads. If you choose to create a template-load-
ing process, at the point of schedule you can program company-specific information, scheduling parameters, notifications on load success or failure. Indeed, one of the biggest benefits of the GoodData Data Integration Services is that once a process has been defined, tested, and deployed to the GoodData cloud infrastructure, users have a visual interface where they can configure criteria such as notifications via email of the success or failure of a
load, or when a key performance indicator (KPI) has been reached. Once the designing of the logical data models is complete, all processing is performed in the cloud. Although the tool exists on a local PC or server, the data is pushed to the cloud so you don’t need to invest in hardware resources for the data load and transformation. The data flows directly from Facebook, Salesforce, SAP, or other data sources directly to the cloud.
Store: Flexible, Scalable, and Secure Cloud Database
For the storage aspect of the BI service, the GoodData Platform using the HP Vertica Database, a powerful, reliable, scalable, and secure data storage application in which data from all connected data sources is stored. GoodData’s Innovation Velocity GoodData separates the logical data model from the physical model. This means GoodData can continuously implement new innovations—and infrastructure—that increase performance and lower costs, all without disruption of the user experience.
Vertica uses advanced technologies like columnar storage, projections (instead of indexes), optimized I/O, and other features that help achieve top query performance. Businesses will get one of two possible deployment scenarios of Vertica: either as a multitenant deployment on Amazon EC2, or on a private cloud using Rackspace cloud services. In either case, businesses are provided with the highest security. With a Rackspace private cloud deployment, businesses get their own private space with customer data stored on dedicated hardware. Data is encrypted using SafeNet, a leading global provider of data protection. This important component helps GoodData implement the end-to-end data encryption—data encryptions at rest—of the shared file systems on NetApp storage. In the first stage (Load), the data model is an abstraction—a logical data model. During Store, the GoodData Platform automatically translates this logical data model to a physical one, optimizing the data as necessary by creating indexing, splitting tables, and otherwise storing it as efficiently as possible to achieve highest report rendering performance. In addition, the Store component of the GoodData platform allows users to take data snapshots of historical data at specific points in time to analyze and compare trends or examine past performance. Users get real-time access to their data, can easily provision large numbers of new users, and can be confident that their data is completely secure.
Analyze: A Robust, Extensible Analytics Engine
Once the data is securely stored in the cloud in the HP Vertica database, users get access to the GoodData Extensible Analytics Engine (XAE): a no-cube technology for flexible slice-and-dice functionality.
The GoodData XAE uses multi-level caching for exceptional performance —each multidimensional query is broken down into multiple small sub-queries that are executed and cached for reuse. The pivoted query result is also cached. Dashboards, reports and metrics are abstracted from underlying data model, freeing up GoodData to implement innovations without disrupting users’ operations. The engine is extensible with additional predictive analytics and other advanced modules. It also optimizes the underlying storage structures for multiple underlying data query and storage systems, including both relational and columnar database configurations, for optimal performance across use cases. The analytics engine also pivots and transforms the data for visualization purposes. In addition to the GoodData XAE, the GoodData Platform includes the GoodData Multidimensional Analytics Query Language (MAQL). GoodData MAQL is a powerful, easy-to-use query language similar to SQL, but optimized for multidimensional environments. This expressive language is designed for reuse, and is simpler to use than Multidimensional Expressions (MDX), the query language for OLAP databases. GoodData MAQL includes advanced expressions such as conditionals, inner/outer joins for exception reporting, ranking, transformations, multidimensional reporting, and time aggregations such as year to date (YTD). All this allows users to gain rapid insight into their data while they implement out-of-the-box analytical best practices. Users can further enrich their knowledge with custom metrics and advanced “slicing and dicing” and predictive analytics.
Visualize: A Delightful and Intuitive User Interface
When it comes time to deliver reports, the GoodData Platform offers a rich and intuitive user interface for reporting and data visualization. The interactive dashboards are editable in pixel-perfect, what-you-see-is-what-you-get (WYSIWYG) manner.
Share: Immediate and Productive Collaboration
The GoodData Platform also has advanced collaboration capabilities that allow users to share insights with colleagues at any time, no matter where they are located. They can collaborate on dashboards and reports in real time, and even trigger notifications to colleagues of important events—say when a KPI reaches a particular threshold in the data. These Collaboration capabilities support cross-departmental and virtual teams, enabling pixel-perfect PNG/ PDF export of reports and complete dashboards. Any changes to reports or dashboards are instantly propagated to all users. Mobility is also a key feature of the GoodData Platform. Users can access reports and dashboards on any mobile device or schedule them via e-mail. The GoodData Platform also enables the full embedding and white-labeling of dashboards using SAML and Okta support for easy integration. The GoodData Platform also enables exporting of raw data as well as reports and dashboards. It supports CSV and Excel for raw data, and PNG and PDF for rendered tables and charts. Because it is pixel perfect, exported reports and even complete dashboards look exactly the same in an exported PNG/PDF as they do on screen, and reports can be pushed directly to social streams such as Chatter and Yammer. All reports are also enabled with a “commenting” feature that allows users to have real-time conversations within the report. Finally, users can schedule the timing of the reports, and get results over email.
Managing The GoodData Platform
Even though the full stack of BI components allows for complex and comprehensive analyzing, visualizing, and sharing of data, GoodData Platform also has management tools that ensure people can access the service as needed, and manipulate reports, log in when they have to, and load data whenever they want. Because the entire platform is based on REST principles, every management/ administrative function is available via APIs.
People are coming out of the woodwork saying, ‘I love this. I need these reports, too!’
Bob Hammond Chief Technology Officer JumpTap
Management features of the GoodData Platform include the following: Dynamic provisioning Provisioning of new users, projects, and user permissions all can be automated. For example, you can automatically provision new users using pre-built connectors to Microsoft Active Directory. Comprehensive automation capabilities Users can create new data marts, data warehouses, or provision new users in just seconds. Partitioned domains Full integration with third-party solutions means that projects and user spaces don’t conflict or overlap with others.
Mass management tools Projects can be derived from templates, and analytical objects (for example, dashboards, reports, or metrics) can be transferred between multiple projects and upgraded or updated using the GoodData Platform API. Single sign on framework SAML-based single sign on framework enables you to integrate the GoodData Platform with many authentication systems. No double authentication is required. White labeling opportunities Dashboards and reports can be embedded into third-party SaaS or web applications and portals.
Ways To Use the GoodData Platform
It’s been humbling to see our customers slice and dice data and build custom reports and dashboards so quickly.
Mikkel Svane CEO Zendesk Your Apps An Enterprise subscribes to GoodData and builds its own custom analytics leveraging the open architecture of the cloud-based, multi-tenant GoodData Platform. GoodData Apps An enterprise subscribes to one or more GoodData Apps, such as GoodData Sales Analytics App or GoodData Marketing Demand Analytics App, and uses them in conjunction with the GoodData Platform to create analytics and reports based on industry best practices. Powered By GoodData Solution providers can partner with GoodData in two ways: First, they can embed the GoodData Platform in their own analytic offerings, which are white labeled and marketed as “powered by” GoodData. Or they can build their own applications on the GoodData Platform and bring them to market.
The GoodData Platform gives businesses the opportunity to achieve what until now has been elusive: to understand how business activities lead to cash. It enables them to nurture leads through social media marketing campaigns, sell through their CRM systems, and close the deal in the billing system. At the end of the day, businesses can understand the holistic path customers took and calculate ROI on specific campaigns, identify how campaigns are different, how many leads a particular campaign contributed, and other key metrics. GoodData helps businesses achieve these goals by offering enterprise IT the only complete, open, and agile BI solution delivered as a service available today.
Get started started today with a 30-day Free Trial