Retail Media: Not everything that can be counted counts, and not everything that counts can be counted

Retail media are easily measurable

In retail media we are lucky that there are great opportunities to measure our effectiveness. CPG and FMCG retail media buyers are working in partnership with the people who are selling our wares, our retail media supplier. We can ask them how much they are selling of what and to whom. It is possible for us to get real insight into how our customers are behaving and what the effects of our retail media spend are.

Those who know me will know that I am fanatical about measuring results and then taking those measurements seriously. It comes as something of a surprise to find myself saying that I wonder if retail media experts are taking this too far? It may be that this leads to discarding some things simply because they cannot easily be measured. Might we be deciding that retail media that are easily measured will allow us to look good by demonstrating a real return? With the major economies of the US and UK entering what looks like a significant downturn being able to demonstrate a return is very important for job security. Does this lead to throwing out good retail media simply because they are hard to measure?

Except the difficult ones

As Max says "In an increasingly quant-driven marketplace, it’s easy to obsess on what you can count and disregard the rest." retail media contains a few examples of the "hard to measure" variety. Measurement of sales shows that adding them to your media mix gives a measurable sales uplift, but it is not currently possible to measure their actual effectiveness because there is nothing show tell who has viewed them and who has not. Worse than that there is an observer’s paradox here.

The Heisenberg uncertainty principle says there are limits on what can be measured because the act of measuring affects the thing being measured. This doesn’t only apply to subatomic particles, it can apply to retail media as well. Taking floor graphics as an example, there is a provable sales uplift through using them correctly but there are still many customers who are not even aware they exist. If one tries to measure who has viewed the floor graphics this will make these people aware they exist and so will affect the customer experience. At this point many marketers are switching off and saying that they cannot be measured and so it isn’t possible to reliably change their effectiveness. This is exactly the sort of thing that Einstein might have been talking about in his quotation about what really counts.


But even there retail media can tell us something

Instead of taking the path of least resistance it is necessary to go further down the path of trying to measure them. Granted this will affect the measurement so it will not be safe to rely on generalising it without thinking, but it may still prove useful. Even if it does not say anything definitive there is the fall back of the known return from floor graphics.

So let us assume that a survey to find out who has seen the graphics is conducted. Where does this lead? One can measure the total sales uplift before and after the survey and then examine the difference. The difference is likely to be positive as those who did not know floor graphics existed now do, from the questions they have been asked, and so are more likely to notice the graphics in future. Regardless of the scale and direction of the difference made by the survey it is now possible look at how the people asked compare to the whole customer population in terms of uplift in subsequent campaigns. You can gather a lot of data that may or may not make floor graphics entirely accountable but will tell you some more about how you should use them and when. Who knows this may even prove that for the surveyed sample uplift has improved. If that is the case the next step is to look at how to mirror this across the whole population of potential customers.

Einstein was right and yet it is still possible to make a difference to the less measurable retail media.

Rufus Evison
With thanks to Max Kalehoff who provided the Einstein quotation that was so apposite to this article that I made it the title.