Thursday, 28 January 2021

Getting your facts straight

 


I'm not sure when the obsession with "fact-based" and "data-driven" insights began, but I do know that often enough, chasing after spurious "facts" and/or data to support whatever argument you want to make takes more time than carefully constructing your knock-their-socks-off strategy. 

Let me take a good insight - Guns don't kill people, people do

Now, who in their right mind would demand to see the data or facts supporting that? The whole thing about insight is that when it's good, it's like an a-ha, a Eureka, a leap of faith, a revelation, the obvious that you'd never really thought about that way. Your mind has done a flip from rational, step-by-step logical thinking into a more holistic way of seeing the world.

I must admit it can be tiresome to be continually asked to back everything up with facts, or data. I'm a one woman show, and if clients think it's best use of my time scouring the internet for random information that may appear relevant to the case, but on closer inspection isn't, then so be it.

For my own part, I'd offer the following advice:

What are your reasons for demanding facts and data, and are they good, or is it about ticking the "selling it in to management" boxes?

Have a look at some of the facts and data you're presented with. Do you see a % sign, an asterisk and a source and think good-o, time to move on?

Only too often, I see "facts" presented along the lines of "32% of people believe that ...."

When looking at numbers and percentages, I always ask myself whether those numbers and percentages represent people directly, or not. If not, they may well represent something like € or £ , which likely means that they are less open to interpretation.

But if the percentage represents people, you really do need to find out at least which people, when and where, to judge whether the data is relevant to your case. And best to delve into what (was meant when the word "brand" was used in the survey, for example) and why the survey was done, while you're there.

For example, you may find that those numbers regarding propensity to switch brands during the pandemic were only referring to brands of toilet paper, not brands in general.

Or that the survey of young people's attitudes to social media privacy was carried out in 2015.

Or that the sample asked about brand activism consisted only of college students from the USA, who'd volunteered to take part in a survey on the topic.

It's far more dangerous to argue from the particular to the general than it is to admit to not having 100% relevant hard data to support a particular point.

That, by the way, is my view. Not a fact. 

No comments: