Colin Lewis mulls the question of whether pre-existing assumptions about retail media are hindering marketers’ ability to be successful at it.
Last week at Ascendant Network conference in London, Florian Clemens, Director Strategy, Proposition & Measurement for Tesco Media asked a deceptively simple question after a very confident statement was made from the stage:
“What assumptions are you making?”
It stuck with me for a whole load of reasons. Because once you start looking, you realise just how many unexamined assumptions sit underneath how we think about a lot of things – and how retail media is viewed by the wider world of marketing is particularly fraught with assumption.
And assumptions count.
What underpins our assumptions? In many cases, our personal biases.
Biases drive assumptions
I believe the two of the most important factors to be successful in retail media and in our business careers in general are
- being an expert in our own specific domain, for example, retail media, adtech, selling, marketing- whatever our particular job is.
- being an expert in understanding our own cognitive biases.
The two big biases that matter, in my opinion, are ‘cognitive bias’ and the ‘availability heuristic’.
A ‘cognitive bias’ is the tendency to focus on and remember information in a way that confirms our preconceptions and worldview – in other words, we see reflections of a truth we have already assumed. Our judgement is distorted by mental shortcuts, emotions, or prior beliefs – rather than by facts or logic. The craziest bit about cognitive biases is that you might feel like you’re being rational, even when you are not!
The ‘availability heuristic’ is a mental shortcut where we judge how likely or true something is based on how easily examples come to mind, rather than on actual data or probabilities. If I can think of it quickly, I assume it’s common / important / true. For example, a CMO remembers one terrible “banner ad” campaign that they saw this on their phone, so they assume all digital display is ineffective. That one story is “available,” so it dominates their judgement.
The challenge with biases
We like to believe that, over time, the evidence will prove us right. We work harder, build a better deck, find “just one more” case study. But that’s often just our old friend confirmation bias at work: we search for evidence that supports what we already believe, and quietly ignore what contradicts it.
The problem is that it is hard to clear the slate. Our worldview, our smartphones and our choice of social media feed give us the illusion that we are absorbing objective information, whereas we are, in fact, observing biased information because we tend to choose our news from customised sources. We often imagine ourselves to be paragons of impartiality, whereas we actually might be a collection of crazy beliefs and biases.
Retail media assumptions that should be questioned
So, let’s actually look at some of those assumptions and see where they are leading us.
Assumption 1: “Retail media is purely lower funnel”
If you’re a classic brand marketer, this assumption is pretty common.
The logic goes something like this:
- Retail media is based on search and is close to the point of purchase.
- Therefore it must be “performance media”.
- Therefore it can’t be “real” brand building.
The assumptions are:
- That anything close to the buy button can’t build brands.
- That the act of purchase is somehow separate from the experience of the brand.
How is getting the product into someone’s hand not brand building? The moment you try, buy, taste, wear or use something is one of the strongest possible brand moments. That is where memory structures are created and refreshed. Pretending that this is “just performance” is a just another bias – but it doesn’t make it true.
Assumption 2: “In-store screens don’t build brands”
You hear this a lot – usually from the same people who happily sign off big outdoor budgets. The assumption works as follows:
- In-store = “shopper” = tactical = not brand.
- Outdoor = “proper media” = brand.
Yet the screens might be the same size, built with the same technology, often bought via the same holding companies!
Somehow, a screen at the side of a road, glimpsed at 50 km/h from a moving car, is treated as a noble brand-building medium…but a screen in the aisle, right next to the shelf, in front of an actual buyer, is dismissed as ‘just shopper marketing’ and therefore not worthy of the status attributed to outdoor.
In-store screens today can deliver high-quality, dynamic creative, targeting by store, time of day, shopper mission and they can show brand messages at the very moment of choice. If that’s not at least capable of building brands, what is?
At its core, I would argue that this assumption is based on how the world used to look before the advent of digital instore screens with amazing 4k capabilities and powerful content management systems.
Assumption 3: “Agentic commerce will replace everything”
Right now, agentic commerce and AI agents are the hot topic. Let’s start with what is “agentic AI” in retailing and digital commerce? In simple terms, a bot that doesn’t just recommend, but acts on your behalf.
For example:
- Agents that watch prices and automatically reorder.
- Agents that plan your shop based on your goals (budget, nutrition, preferences).
- Agents that negotiate – working in the background on your behalf
You can read a variation of the following everywhere:
- “AI agentic commerce will kill retailers.”
- “AI agentic commerce will kill brands.”
- “AI agents will kill loyalty.”
- “AI will kill creativity.”
The assumption is that agents will take over all the messy bits of shopping, so none of this digital commerce and retail media stuff will matter.
Agentic commerce will almost certainly change how people discover, compare and buy. It will not magically erase the need to build brands, create desire, and be present where choices are made – including in retail media environments.
We underestimate how long old habits – and old assumptions – hang around. If you don’t believe me, ask yourself about the following:
- Amazon Dash button ordering
- Voice ordering through Amazon Alexa
- Metaverse commerce
- Drone delivery
- Just-walk-out stores
All the hype about agentic is classic example recency bias in action. The lesson is mostly to not overestimate short term (direct) impact and not underestimate long term (indirect) impact.
Assumption 4 – “Broad reach media good, targeted reach media bad”
This assumption usually turns up in the form of the statement that “TV advertising is broad reach, everything else is narrow and fragmented and cannot deliver brand brand growth”.
The broad reach vs targeted debate leans (selectively) on How Brands Grow by Professor Byron Sharp and Professor Jenny Romaniuk. It’s an excellent book with powerful ideas on penetration, mental and physical availability. One of the key tenets of HBG is that brands grow by increasing penetration, and they do this through:
- Mental availability – ensuring your brand thought of of in buying situations.
- Physical availability – ensuring your brand easier to find and by the shopper.
But, as Steve Gray of SG Retail points out ‘most media channels affect only the first. Retail Media affects both, simultaneously, and often dramatically. With Retail Media, they are not only buying awareness. They are upgrading physical availability, which drives measurable increases in category entry, trial, and repeat. This is what makes Retail Media categorically superior to TV, digital display, OOH, and social.’
If you treat How Brands Grow as a set of principles (reach, consistency, distinctiveness, memory structures), retail media works perfectly to support those principles.
Steve Gray summarises it perfectly: ‘Retail media is the only marketing channel that delivers both mental and physical availability — the two scientifically proven drivers of brand growth. Penetration grows when presence grows. Presence grows through retail media.’
Assumption 5 – “Measurement in retail media is a problem”
Often, the statement about measurement is a problem in retail media is really shorthand for: ‘this is new, I am not sure I really understand it, they want some of my budget, I don’t like that’!
This is an assumptions problem. Part of that assumption is that other media have better measurement standards. But how true is this assumption? How about how TV ratings are measured?
Audience ratings are the currency by which decisions are made on whether an audience does, in fact, exist and the subsequent media buying and selling that happens as a result. They are the established foundation for advertiser supported TV. So how are these audiences measured?
BARB (in the UK) and Kantar panels globally are the methods by which TV ratings are measured. The measurement system for TV ratings is based on a sample of households with meters that detect what’s being watched on each set. Household members have to log in and log out so the system knows who is watching. These nationally representative panels are then weighted to represent the full TV universe and generate ratings, reach, GRPs, etc.
Panels now also capture time-shifted viewing, BVOD/streaming on TV sets, and in some markets some cross-device viewing (e.g. apps on smart TVs, sometimes PCs/mobiles where integrated).
There is inevitable gap between the measured audience and the actual audience and this gap is becoming even more evident if you walk down a high street, through a mall, get onto a bus or train. What will you see? People addicted to their mobile phone screens. Everywhere. And on those screens, every touch, swipe, click and ‘buy button’ can be measured down to the pixel.
I am sure many people reading this will still believe TV is perfectly measured and retail media is wild west of measurement, proving the point that I am making about assumptions in the first place.
They may also point out that econometric measurement models are the ‘gold standard’ of measurement. They also have an underlying assumption. As I pointed out, these models are always subject to a big caveat: what was the base level of sales that would have happened anyway?
Check your assumptions at the door
The real risk isn’t that retail media doesn’t work but that our assumptions are faulty.
In an age of hot takes and AI-driven LinkedIn posts, it’s never been easier to offer a point of view that looks solid at first glance but falls apart under when you see the assumptions underpinning the statement.
For retail media in particular, it’s worth asking yourself regularly:
- What am I assuming about “brand” vs “performance”?
- What am I assuming about screens in-store vs screens anywhere else?
- What am I assuming about what counts as “real” reach?
- Where am I using “measurement” as a stick to beat people with?
As a practical test: write down your three biggest beliefs about retail media and ask, “When did I last update this – and what would it take to change my mind?”
