We all love to make decisions based on data. It’s objective, mathematical, undisputable. But not always. We can make numbers tell the story we want to telle depending on how we get and present them.
For example, let’s talk about the impact of robots on jobs. Many reliable organizations tell us that this impact will be small, other serious organizations tell us that we are going to face a massive job destruction. And, in this second case, some say that 18% of jobs will disappear, others 42, others 27. If it clearly points at a real trend, I’m curious about the methods used to claim that within 10 years 10,20 or 30% will have disappeared. Who’s right ? The matter is so complex that I’m interrogative regarding the methodology used.
One can always find numbers to prove he’s right
The same thing happens when a company chooses a technology. X% more productivity. Y% more engagement (as if technology alone was able to build engagement). Meanwhile, everybody agrees that very few projects meet the results promised by the cases made on similar projects in the past.
That’s what I call the “McKinsey syndrome”. When McKinsey says, for instance, that using digital technology at work, a 25% raise in productivity can be expected, is it a prediction (and based on what) or the what businesses who actually did it have measured. In this case, knowing that McKinsey mostly surveys its clients, is it caused by technology or by McKinsey’s impact on organizational change ?
And there are the goals that are assigned for the wrong reasons. I’m building a brand new e-commerce site and expect 20% more sales. I’m rolling out a new HRIS and expect HR’s productivity to be 5% higher. Why 20% and not 22 or 19 ? Why 5% ? What study was conducted to get these results ? Most of times absolutely none. But these are the numbers that will help to get the budget approval. And what if sales increase by 17% only ? The project will be seen as a failure while it’s actually a success. 17 is a good performance but such a promise would not have helped to get the budget.
Using numbers to please others…
It reminds me of a discussion I had with a friend who “crashed” his startup. Why ? He overestimated the market size. The point is not that his model was wrong but that it was not profitable enough. So the investors said “stop”. He could have run his business with only half the funds he raised but investors told him : no ticket under x millions. So the market size needed to be consistent with the x millions.
So numbers are good but we should always use them carefully. Sometimes it’s better to understand the meaning, the trend rather than to trust numbers. “Yes, things will improve but we cant’ say to what extent”. As a matter of fact pretending we know and find the numbers to back our assumptions has a consequence : business cases are built on lies and in the end either projects crashes or successes are seen as failures. Decision makers should also understand they should hae reasonable expectations : they will always find someone who will come with the numbers they want to ear, no matter it’s realistic or not.
Who measures the risk of doing nothing.
And since we’re discussing ROI, let’s talk about a number no one uses : the RONI. Return on non investment. What happens if we do nothing. You’ll see it’s very easy to find out what is going to happen if nothing is done and it’s sometimes so scary that decision makers start to have lower expectations on the ROI and accept more realistic numbers.
But we can continue telling stories and pretending we have the right numbers…and wonder, year after yearn, why 99% of projects are underperforming compared to expectations.
Photo Credit : Fotolia