Instrumenting Your Enterprise for Maximum Predictive Power

By James Kobielus

Business is all about placing bets and knowing if the odds are in your favor.

As I noted in my most recent Forrester report, business success depends on your company being able to visualize likely futures and take appropriate actions as soon as possible. You must be able to predict future scenarios well enough to prepare plans and deploy resources so that you can seize opportunities, neutralize threats, and mitigate risks.

Clearly, predictive analytics can play a pivotal role in the day-to-day operation of your business. It can help you focus strategy and continually tweak plans based on actual performance and likely future scenarios. And, as I noted in a recent Forrester blog post, the technology can sit at the core of your service-oriented architecture (SOA) strategy as you embed predictive logic deeply into data warehouses, business process management platforms, complex event processing streams, and operational applications.

The grand promise of predictive analytics—still largely unrealized in most companies—is that it will become ubiquitous, guiding all decisions, transactions, and applications. For the technology to rise to that challenge, organizations must move toward a comprehensive advanced analytics strategy that integrates data mining, content analytics, and in-database analytics. Already, we’ve sketched out a vision of “Service-Oriented Analytics,” under which you break down silos among data mining and content analytics initiatives and leverage these pooled resources across all business processes.

You may agree that this is the right vision but have doubt about whether there is a practical, incremental roadmap for taking your company in that direction.  In fact there is, and it starts with re-assessing the core of most companies’ predictive analytics capability: your data mining tools. As you plan your predictive analytics initiatives, you should avoid the traditional approach of focusing on tactical, bottom-up project-specific requirements. You should also try not to shoehorn your requirements into the limited feature set of whatever modeling tool you currently happen to use.

To become a fully predictive enterprise, you will need to take both a top-down and bottom-up approach to your data mining initiatives. From the top-down, it’s all about building and integrating alternate models of how your business environment is likely to evolve internally and externally. In our recent report on advanced analytics, Boris Evelson, Leslie Owens, and I sketched out the many business processes that can be enriched by predictive analytics.

So how do you instrument your company to become more predictive? For starters, assess whether your analytics tools support the following capabilities for developing, validating, and deploying predictive models:

• Model multiple business scenarios: You should be able to build complex models of multiple, linked business scenarios across different business, process, and subject-area domains, using such key features as strategy maps, ensemble modeling , and champion-challenger modeling.
• Incorporate multiple information types into models: You should be able to develop models against multiple information types, including unstructured content and real-time event streams, while leveraging state-of-the-art algorithm in sentiment analysis and social network analysis.
• Leverage multiple statistical algorithms and approaches in models: You should be able to develop models using the widest, most sophisticated range of statistical and mathematical algorithms and approaches, including regression, constraint-based optimization, neural networks, genetic algorithms, and support vector machines.
• Apply multiple metrics of model quality and fitness: You should be able to score and validate model quality using multiple metrics and approaches, including quality scores, lift charts, goodness-of-fit charts, comparative model evaluation, and auto best-model selection.
• Employ multiple variable discovery and assessment approaches: You should be able to build and validate models using various approaches for variable discovery, profiling, and selection, including decision trees, feature selection, clustering, association rules, affinity analysis, and outlier analysis.

How is this different from predictive analytics as usual? Traditionally, most predictive modeling specialists focus on the latter three capabilities: statistical algorithms and approaches, model quality and fitness, and variable discovery assessment. Most models are built in narrowly scoped business or subject domains—such as customer analytics for marketing campaign management—and only against structured data sources (such as relational tables). Traditionally, few predictive analytics projects have entailed modeling of multiple business scenarios across diverse domains--such as sales, marketing, customer service, manufacturing, and supply chain-- though in the real world these business processes are often quite interconnected. Also, many data mining initiatives fail to incorporate information from unstructured sources—such as text in call-center logs—though this content may be as important as what comes relational databases and other structured sources.

It’s very important to build multi-scenario predictive models against complex information sets, but becoming a fully predictive enterprise demands much more. To instrument your organization for maximum predictive power, you should also tool your advanced analytics to support the following capabilities:

· DW-integrated data preparation: To speed up and standardize the most time-consuming predictive modeling project tasks, you should be able to leverage your existing data warehouse, extract transform load, data quality, and metadata tools to support a full range of data preparation features. These features include the ability to discover, acquire, capture, profile, sample, collect, collate, aggregate, deduplicate, transform, correct, augment, and load analytical data sets.

· Deep application and middleware integration: To deliver models deeply into whatever heterogeneous SOA-enabled platform you happen to use, your predictive analytics tool should deploy on and/or integrate with a wide range of enterprise applications, middleware, operating platforms, and hardware substrate. You should be able to deploy models seamlessly into your data warehouse, business intelligence, online analytical processing, data integration, complex event processing, data quality, master data management, and business process management environments. And to play well in your SOA, your predictive modeling tools should support application programming interfaces, languages, tools, and approaches such as Web services, Java, C++, and Visual Studio, as well as emerging languages such as SQL-MapReduce and R.

· Consistent cross-domain model governance: To avoid fostering an unmanageable glut of myriad models, your predictive analytics solution should support a wide range of tools, features, and interfaces to support life-cycle governance of models created in diverse tools. At the very least, your tools should enable model check in/check-out, change tracking, version control, and collaborative development and validation of models. To realize this promise, it should support a full range of tools, standards, and interfaces for import and embedding of models from other tools, as well as export and sharing of models to other environments.

· Flexible model deployment: To execute modeling functions--such as data preparation, regression, and scoring—on the widest range of data warehouses and other platforms, your tools should support in-database or embedded analytics. And to scale to the max, your predictive analytics tools should deploy models to massively parallel data warehouses, software-as-a-service environments, and cloud computing fabrics.  Your advanced analytics tools should also support development of application logic in open frameworks—such as MapReduce and Hadoop—to enable convergence of data mining and content analytics in the cloud.

· Rich interactive visualization: To deliver their precious payload—actionable intelligence—your advanced analytics tools should support interactive visualization of models, data, and results. Ideally, you should be able to visualize all of this in your preferred business intelligence tool, or in the predictive modeling vendor’s integrated visualization layer. Of course, you have every right to expect the full range of visualization techniques--histograms, box plots, heat maps, etc.—regardless of who provides the visualization layer.

As you can see, this goes well beyond data mining as usual. Forrester has a slightly different perspective on the development of the predictive analytics market than you’re likely to get from other sources. We see a robust, flexible, SOA-enabled data mining tools as the centerpiece of advanced analytics for fully predictive enterprises. The competitive stakes are too great for businesses to take the traditional silo-mired approach when implementing this mission-critical technology.

What do you think?

Informatica Evangelizes Its Increased Business Relevance With Latest Release

By Rob Karel

Last week Informatica announced the release of Informatica 9, its data integration/data management platform that continues to evolve its flagship PowerCenter and PowerExchange data integration and access technologies into a much more comprehensive data management platform going well beyond the scope of traditional, batch-oriented ETL that remains Informatica’s bread and butter.  

The three main themes Informatica has pitched for this release include:
- Pervasive Data Quality
- Business-IT Collaboration
- SOA-based Data Services

While these themes and capabilities - reusability, SOA-compatibility, real-time, business engagement - are not necessarily new to the broader data integration or data quality software markets, few organizations have been effectively able to execute on them.   For the purposes of this blog post, I’d like to focus a bit more on the DQ and business/IT collaboration parts of the announcement.

With pervasive data quality, Informatica wants its customers to extend the use of data quality software well beyond the limits of data warehousing.   Informatica 9 provides role-specific tools that can be leveraged by both the IT and business stakeholders involved in defining, analyzing, testing, and implementing data quality logic.   Informatica has also included advanced postal address verification and global matching from its acquisitions of AddressDoctor and Identity Systems, respectively.   

I’m a huge fan of Informatica’s goal of centralizing the definition and management of data quality rules that can then be delivered as real-time data services to any part of the enterprise.  But as I discussed in my “It’s Time To Invest In Upstream Data Quality” research last year and most recently with Forrester BPM analyst Clay Richardson in “Warning: Don’t Assume Your Business Processes Use Master Data”, the biggest inhibitor to enabling this is not technical.  Business process owners responsible for the capture and update of critical enterprise information are too often more concerned with process efficiency than data quality issues that may impact downstream processes. 

Similarly, I think Informatica is on the right track with its Business-IT Collaboration theme, but has a lot more work to do from evolving this from vision to reality.   Informatica 9 provides some useful business-oriented capabilities including browser-based tools designed for business analysts and data stewards to participate more actively in the data profiling and DQ rule definition phase of a data quality initiative.   And from a collaboration standpoint, the shared metadata environment ensures both business and IT have access to the same information, and the ability to bookmark and share views of analysis and business rule output, along with comments, is a great feature that eliminates the old method of exporting results to a spreadsheet and attaching to an email.   My primary concern is that Informatica has not yet created a pre-packaged interactive collaborative environment.  For example, there are no predefined approval, escalation, and validation workflows.  

Interestingly, almost every major data quality and master data management vendor I speak with is investing quite heavily in what some are trying to define as the Data Governance Technology market. Or in other words, tools used to support both business and IT stewards in the data governance process.   Of course data governance is not a technology market, it is a business responsibility that must manage the people, process and organizational challenges required to get business value from data assets.   But there are enabling technologies that can support data stewards from both the business and IT side during certain steps of the data governance lifecycle, and the jury is still out on what approach to this technology will be the most effective. 

But the biggest challenge for enterprises trying to embrace data governance has nothing to do with the tools.  It’s getting the business to participate in the first place!   In a recent Forrester survey asking MDM and DQ professionals to rate their organization’s current level of data governance maturity, 79% said they did not have a data governance organization that included active participation from all the necessary business stakeholders. Unfortunately, there is very little that a software vendor can do to change this culture, so I’m not expecting widespread adoption from the business of Informatica’s or anyone else’s data governance tooling until that maturity improves.

Process-centric DQ Services and MDM among top technology trends to watch

By Rob Karel

For the past few years, master data management (MDM) has been a hot topic among enterprise architects and data management professionals. But many of these evangelists have felt like rebels without a cause because their MDM initiatives have been primarily IT driven with minimal business sponsorship and participation.   In other words, the business expects high quality data, but hasn’t taken much accountability in delivering it. 

Data quality (DQ) management, a more mature technology and competency than MDM - and a required core capability for any MDM solution - has also struggled to be embraced by the business as a top priority.  More often than not, DQ has been relegated as a supporting downstream batch application for ETL, data warehousing and business intelligence applications, but not for a broader cross-enterprise data architecture including upstream transactional systems and processes.

But Forrester now sees this trend changing for the better, as highlighted by my colleague, Alex Cullen, who recently published important research on “The Top 15 Technology Trends EA Should Watch.”  A key theme Alex identifies is around Process-Centric Data and Intelligence. Alex believes - and I agree with him - that both MDM and real time data quality services will become critical enablers to support real-time operational and decision-making processes.  

Organizational competency in the definition and implementation of DQ rules, processes and technologies that can effectively cleanse, standardize, enrich, match, and merge critical data is a foundational step to moving towards true MDM:  the real-time delivery and synchronization of a single, trusted view of enterprise reference data to both operational and analytical environments.  Why? Because you can’t bi-directionally synchronize master data if you haven’t implemented the DQ rules required to derive a master record.

Unfortunately in many instances the after-the-fact batch cleansing of data in a data warehouse is often too little too late in ensuring high quality data can be delivered via MDM or otherwise.   This is why we see real-time DQ Services as a critical evolution for data architects to embrace and evangelize: embed DQ validation and cleansing logic within the upstream transactional processes that capture and maintain this data in the first place.   Not only will this of course improve the trustworthiness of the data used for analysis later, but even more transformational – it will improve the efficiency and quality of the upstream transactional processes themselves. 

A simple example:
A “Ship To” postal address is typically captured as part of an order management process.  Before real-time DQ Services are implemented to ensure that the addresses captured meet formatting standards and are in fact deliverable, order fulfillment runs a very real risk of being delayed – leading to potential postage penalties and negative customer satisfaction due to the delays.  But DQ Service can significantly reduce this risk by recognizing these DQ issue while the customer is still engaged in the process and best able to respond.

As further evidence of this trend toward transactional data trustworthiness, in my just-published “Trends 2009: Master Data Management” research I share the results of a recent Forrester MDM readership survey where over 64% of respondents said their planned or implemented MDM solution must support real-time subscription requirements, and not just scheduled batch loads.   Therefore it’s not too surprising that almost 79% of those surveyed that have decided upon an architectural deployment style for their MDM solution selected a Hub, SOA, Federated, or Registry approach since all of these are optimized to support real-time, transactional MDM requirements.  

But don’t expect this transformation to real-time DQ and MD to happen overnight. The complexity and invasiveness of these data services to the streamlined transactions that are managed make many EA’s wary, especially since most organizations still have extremely immature data governance programs.   In the same readership survey, nearly 46% of companies rate their data governance maturity as low or very low, while another 29% rate their maturity average, seeing significant participation from just a few key business stakeholders.   Data governance – an organizational and cultural shift – is the first trend EA’s must follow and evangelize in order to deliver the process-centric data and intelligence their organizations are demanding.