sascom voices - The sascom magazine blog

It seemed like we were all settling in for some post-acquisition indigestion with a keen ear for roadmaps from the mega vendors on how they look to rationalize their bloated product portfolios. While we got some of it -- whether good, bad or nightmarish -- depends on which side of the equation you fall. And now, there seems to be a new flurry of activity from the BI vendors with freshly minted monikers, namely ‘predictive analytics.’

The word analytics was conveniently and loosely used during all the consolidation hoopla with claims that each vendor already had analytic capabilities. So why now the sudden urge to form strategic alliances with SPSS by Business Objects (an SAP company) and Cognos (an IBM company) respectively? Did we finally get it? Well…Sort of!

Analytics is more than just predictive analytics and the simple answering of the question, What will happen next? Don’t get me wrong: predictive analytics is a key component of the analytics maturity curve, but you also need statistical analysis, forecasting and optimization to answer questions like, Why is this happening? What if the trend continues? And what’s the best that can happen?

Analytics is about helping organizations reduce uncertainty, predict with precision and optimize performance. To provide these data-driven insights, you need an environment for predictive analytics, descriptive modeling, data mining, text mining, forecasting, optimization, simulation, experimental design and more. This environment needs to support a range of techniques and processes for the collection, classification, analysis and interpretation of data to reveal patterns, anomalies, key variables and relationships, leading ultimately to new insights for guided decision making.

I wish it was as simple as providing the ubiquitous ‘Easy’ Button (thanks to Staples, this is now part of every PowerPoint junkies lexicon). Or in this case should I say, it’s not as easy as providing a ‘Score’ button to traditional query & reporting BI users. Don’t shoot the messenger here, I am just quoting some of the parallels that others are starting to draw.

Shaking hands and spinning value on embedding capabilities vs. technology alliances may open doors, but the same opportunities will undoubtedly lead to integration challenges. It’s not as if these vendors have infinite resources and nothing else happening on the integration front. Quite the contrary: They have only scratched the surface on that one. Business Objects and SAP have lots of overlapping products, many of which are now destined for rationalization. And doesn’t IBM have a computational mathematics department that spearheads all the great analytics coming out of IBM, and wasn’t this department poised to work with Cognos as part of the big picture strategy?

Another interesting tidbit is that data preparation consumes 60 to 80 percent of the time it takes to complete a data mining project. Again we’re seeing ‘easy button’ answers about calling back in to SPSS for the analytical results. A little tip here for the not so faint hearted: You’re going to need a common metadata repository that defines the analytic data and automatically refreshes the data as new data becomes available, and you’re going to need a model with the ability to push that model to the metadata as an object to be re-used as a business asset by all others parts of the ecosystem.

I need to wrap this up as I have oft been told that my blog posts are way too long, which makes me rule breaker. But then, I am more a New Yorker than any anything else when it comes to my US roots and so rules, what rules? The line starts from wherever and whenever I show up.

Partnering with original equipment manufacturers is not a new strategy for SPSS. Over the years, SPSS has partnered with database vendors and embedded its analytics into third party applications. This recent flurry of partnerships, however, raises some interesting questions: Why is SPSS making so much noise about its partnerships? Why now? Is SPSS stirring up the waters among key software vendors to make itself more attractive as an acquisition target?

It’s all good from my perspective: our competitors are doing a great job of validating our long held strategy, and we are glad that the market is finally paying due attention to the power of analytics. After all, we’ve been proving this for 32 years and have more than 43,000 customer sites in 112 countries to show for it.