Context : This note from Bill Franks
Bill has aptly put analytics outsourcing in perspective and I could not stop myself from expanding it further to my earlier post.
Spot on Bill Franks! I wonder how the third party analytics (so called knowledge powerhouses) companies will take it as they claim to manage end-to-end analytics (but mostly end up in creating just the reports, huh !).
I have personally witnessed the gigantic mistakes made by some global corporations (some are still walking that path), only to painfully course correct after extensive damage done by blind outsourcing. In my view, apart from the obvious cost benefit (only on paper) narrative of outsourcing; the outsourcing mistakes are also made due to the ‘generalists given responsibility to manage analytics’ within the organization. On one hand these decision makers are clueless about the mechanics of setting up analytics on the other they find it very convenient to outsource ‘everything’. This buys them significant time to get incarnated as ‘analytics evangelists’ and at the same time the outsourcing partner gets all the crap of not been able to meet the expectations.
By the time the mistake is realized, the generalists now converted into analytics evangelists remember enough analytics jargon to sell themselves into some bigger role or get into some other organizations again to repeat the same modus-operandi and outsourcing business continues !
Originally published in Energy Central
“we didn’t do anything wrong, but somehow, we lost”
— Nokia CEO
Context : study reveals that most companies are failing at big data
There is a big problem hampering the evolution of analytics as a strategic function in many organizations – Worldwide. Many analytics leadership and ‘resource approving authorities’ roles are filled with people who have been generalists all their professional life and due to their vintage (aka influence) in the organizations are given the task to build the analytics and data strategy capabilities. Even worse, the CTO’s are assuming roles of Chief Data Officer’s due to their proximity to technology and huge misconception that analytics is a first cousin of business intelligence. Third category is lead by ‘analytics cum technology cum services’ vendors who are pushing / influencing the boardrooms to deploy their quick-fix analytical suits at every given opportunity. As a result organizations are overdoing the data with little focus on science. Quite aptly illustrated in this blog here.
There are more and more horror stories coming out from such organizational experiments. These endeavors either end up in massive investments in misplaced strategies (e.g. thousands and millions of $$ spend in building MIS/reporting platforms with limited actionable output), failures (e.g. ‘unused big data platforms’ or ‘dead data ocean’s’ ) or just a long cherished dream (e.g. ‘waiting for those millions of $$ to even start’). The short term damages due to these phenomenon are the waste of – efforts, $$ and time; but more devastating long term damages are the drag effects of these failed experiments. There failures are resulting as big roadblocks to even restart the analytics journey in short to medium term !
This situation is almost like letting a civilian administrator take charge of military operations and then blaming the soldiers for the defeat!
It’s a no-brainer that the days of taking broad-based business decisions on the gut feel in any customer facing business are over (long back), and it is inevitable for any organization to make the data driven decision making as core part of their long term strategy. There are ample examples where organizations ceased to exist just because they relied too heavily on the gut based decisions and chose to ignore the changing times or were too late to change the course.
On the other hand people who have really had their hands dirty doing analytics job, precisely know where to start. To the least, these people understand the pre-requisites of technology and business to make analytics happen. Unfortunately in most of the organizations these people are tasked with jobs of managing data chaos (something that burn’s them much faster) or are simply pushed to corners (due to political and/or legacy reasons) as white elephants.
Organizations need to understand that analytics is neither a 100% business competency nor is a 100% technology area; it is an art which mixes the colors of business and technology in certain combination to bring out the best shade. And every analytics initiative has a unique combination of these areas which only an analytics artist can deliver !
Please don’t put blame on analytics people for failed strategy when you have kept a generalist or a pure IT person or an external vendor as an authority to drive your analytics initiatives ! You are likely to fail – sooner or later.
For a change; empower analytics artists and let them paint your analytics and data strategy roadmap!
The most ironical fact is to explain what it means to be a data scientist to almost everyone who think that it is Version 2.0 of reporting analyst or even worse a database administrator. Even the people who put money on these unusual professionals expect them to sort out the data shortcomings and produce reports every now and then !
Understanding data and being capable to make unruly data worthwhile for decision making does not necessarily mean that it’s a data scientist job to bring discipline to the data chaos. In fact it’s a virtue that often gets misread as the core capability of data scientists and the real benefits of using these minds and skills to solve business problems or identifying new business opportunities often take backseat.