In his Masters of Business Analytics blog, Tom Davenport has argued that the analytical technology environment will change dramatically over the next several years and that one key change will be a shift from multi-purpose technology environments to focused applications and data. An example would be analytical widgets or tools that are linked to a particular type of decision in a particular industry.
I’ve cross-posted my response – “Will Apps Really Help” below:
The notion of “apps” emerging from the primeval goo of the early analytics era to become the dominant life form in the field is very interesting. It makes sense since the promise of analytics is to enable intelligent and quick decision making given some relevant data – and how nice it would be if there was a button we could press to get some specific results. I think we are already on the way. For example, statistical packages keep introducing new procedures reflecting developments in statistics as well as in response to user needs. A simple human capital proc could be “turnover” where the inputs are headcount data, termination data, organization data, covariates, etc. and the output could be graphs, predicted turnover, or any other useful item. Ultimately, any app will have to tap into some sort of database, perform some sort of standard analysis and produce some sort of standard output. There may be a set of simple standards that emerge that might be somewhat useful, but the real need will be customization that takes into account data quality and the specific focus of the analysis. My contention is that the evolution of analytics will proceed apace with regard to technology, but will be limited by human capability in terms of understanding the science and limitations behind the analytics. So while apps emerge to meet needs, there is a danger that they are oversimplified and lead to sub-optimal decision making.