Blog post

Getting More with Doing Less – The Productivity Puzzle

By Andrew White | February 25, 2017 | 0 Comments

ProductivityInformation Yield CurveInformation ValueInfonomics

Productivity (the lack thereof being the bane of many advanced economies) can basically be summarized as getting more with less.  More specifically improved productivity comes about when the ratio of the factors of production (land, capital, labor etc.) changes positively in favor of output.   All CEOs are concerned with this most basic of ideas.  Irrespective of their organizations’ mission or goal, if it is achieved with lower productivity than, say, previous efforts or compared to an alternative approach or competitor , the result will be second class and possibly damaging.

Users of information or technology (I suspect we ought to split “I” and “T” from IT) such as business leaders or IT leaders like the CIO, should also be concerned with getting more with less.  But there are some interesting and perhaps infrequently thought implications with this idea.  For example, sometimes we can get more output from the same input- which is a dialog about efficiency.  More challenging is the idea that you might be able to get a greater outcome from increased inputs, compared to an alternative approach or mix of inputs and outputs.  Finally, sometimes we actually invest without a goal or specific output defined in order to lay a foundation for a subsequent or later investment and change in output.  In financial terms this investment might be thought of a “lost leader”.

So given these complexities why is it that not more investments in information (data and analytics) or technology not evaluate the implied, likely or desired change in inputs and outputs?  Let’s look at some examples.

  1. Option A is to increase investment ‘x’ in an established data warehouse and analytics platform in order to generate better insights to drive improved outcome ‘a’, versus option B that is to dis-invest in the current systems and instead invest ‘y’ in a data lake and new tools to drive better outcome ‘b’.
  2. Option F is to invest ‘g’ in an MDM and information governance program to assure increased business process integrity versus not invest, as in ‘not g’, in MDM or information governance and just hope outcomes improve.

Both examples are similar but I have rarely seen any organization look at either common opportunity using a ‘get more out’ lens.  So what are we looking at?

In the first example we are trying to explore a change in ratio of productivity from two competing investment approaches: more of the same (classic BI/analytics, option A with investment x) and a new approach (data lake/data science, option B with investment y).   Do we realize sufficiently the increase in outcome (difference between b and a) given the change in inputs or costs (difference between y and x)?  Is the difference in outcomes between the two options greater than the net increase (or change) in inputs?  If so, we have improved productivity.

In the second example we are trying to compare results from making an investment versus not making an investment.  This might seem to be simpler than the first example but in fact it is the exact same calculation.

So the question is- what are the inputs and the change in outputs from each investment (or non-investment) approach?  It turns out that most organizations have a limited set of tools to help here.  Globally and for many years numerous analysis suggests that around 50% of projects, give or take, have any formal ROI.  But an ROI is just an analysis of a given project, not a comparative analysis of the difference between action and non-action or alternative actions.  An ROI simply seeks to determine a rate of return (or yield) on a specific investment.  An ROI cannot be used to gauge the change in yield with different inputs.  That concept we call information yield analysis and today this is mostly a pipe-dream for CIOs, CDOs, CFOs and CEOs.  At best the most innovative firms are just morphing to a unconditional practice that we call Infonomics.  Infonomics is a practice that seems to determine the value of data.  This is a prerequisite before you can do any comparative analysis of alternative investment options.  If you don’t know the value of your assets (at rest), how can you consider what and where to increase or change investments?

Finally there is one last idea that needs to be reviewed: does it really matter what you invest in?  Think about it: If on average only about 50% of investments have any form of real ROI analysis, and the rest of investments are pretty much taken on face (or prayer) value (what consultants call, ‘strategy’), maybe we don’t need to think too much more about this stuff?  After all, we have all done reasonably well, right?  It’s not like the lack of ROI or more advanced analysis has led to consistent or repeated failure or disaster.  And therein lays the bigger challenge: if we agree we don’t need to do too much more complex analysis, what is the minimum thinking (and planning) we need to do in order to reap ‘good enough’ improvements over the other guy?

Such a plan requires a road map detailing the sequence and order of all your information and technology investments.  But what does this look like? Does your road map look like mine?  Should it?  Why not?  Is there a “preferred”, low-risk/easy, or high-risk variant?  Sounds like fodder for another juicy blog.  But let me muse on this: that ‘bare bones’ level and choice of investments probably changes over time.  Just as the evolutionary arsenal of tools used by living creature changes over time in response to its environment and its competitors, so too the set of tools (people, processes, data and technology) needed by a firm to survive.

Comments are closed