Summary Bullets:

• We’re all becoming data scientists, steeped in the meaning and value of data along with data visualization and discovery.
• Applying self-service expectations to big data could readily lead to erroneous conclusions and a resulting lessening of trust in big data and in IT itself.

I don’t play tennis, but if I did so with any passion for the game, it’s likely that I’d also be a data scientist.  Well, at least I’d be thinking more like one thanks to the current market rush toward device instrumentation and personal analytics.

The ‘Internet of Things’ (IoT) is already here, helping so many of us track our daily movements and sleep habits via personal wearable devices like Jawbone and Fitbit bracelets.  So far, however, those have not gone much further than telling us what we already know – that we’re sleep deprived and perhaps a bit too sedentary.  What has piqued my interest is the current wave of IoT devices that are designed not just to instrument and transmit data, but also to analyze that data and provide actionable insight.

That’s where tennis comes in.  The creative folks at Babolat are building a forthcoming tennis racket (Play Pure Drive Set) that encourages players to do some serious data mining while on the court or back at home in order to improve their game, or maybe just their backstroke.  (You have to start somewhere.)

Built in conjunction with Movea, which was involved in building the Nintendo Wii remote, Babolat’s new tennis racket will contain a simple gyroscope, accelerometer, Bluetooth connectivity, and some storage media.  The magic in a solution like this is rests not in the sensors of course, but in the data models built by Babolat.  Their software looks for correlations between the force, timing and direction of a series of hits to surface analytical insights such as “you’re over-swinging your return volleys.”

That’s where an improved backswing can lead to improved corporate efficiency, agility and innovation.  Let’s face it: we’re all becoming data scientists, steeped in the meaning and value of data along with data visualization and discovery.  Recognizing that fact, major data and analytics players such as IBM, SAP, Tableau, etc. are moving quickly to democratize data itself.

However, they’re doing so not just from the top down, but from the bottom up.  With both freely or affordably available, cloud-borne data visualization and discovery tools, these vendors are literally encouraging departmental heads and business owners (not just the data scientists) to plug into both local Excel spreadsheets and massive backend data stores (perhaps a Hadoop cluster), running what-if scenarios right then and there… whether on the tennis court or in the office.

As with all good things, this democratization of data comes at a price.  That price is “eternal vigilance,” as Thomas Jefferson once intoned.  In this case, it is up to IT to work with their executive team, establishing a master data management plan befitting such democratized data.  The creation and maintenance of “trust” in big data and the insights derived thereof really rests in the hands of these two groups and their use of data management and governance software.

It’s one thing to draw conclusions about your backswing using pre-built models and a closed data set, but applying those same expectations to disparate sources of data (i.e., big data) and self-service data discovery tools could readily lead to erroneous conclusions and a resulting lessening of trust in big data and in IT itself.