Why data-rich businesses still make weak decisions, and how category can help close the gap.

This article was originally published on LinkedIn by Simon Dunn, and is republished here with his permission.

‍Stickybeak take:
In markets with little or no volume growth, the commercial teams that win are rarely the ones with the most data. They are the ones who know which reading of it to trust, and can defend that view when a buyer pushes back. In this piece, Stickybeak partner Simon Dunn makes the case that growth is now earned through the quality of commercial judgement, not the volume of information behind it. For anyone preparing a range review or steering a category conversation, it is a useful lens on where the real advantage now sits.

Grocery has a growth problem, and it is more sobering than the headline numbers suggest. Value is still rising across many major markets, but very little of it - if any - comes from volume growth.

In the US, volume declined last year as shoppers bought physically less (value edged up, driven by price). And in the UK, what modest volume growth there was came from a bigger population, not fuller baskets. Strip out price and population and the picture is stark: on a per-capita basis, grocery volume is flat to negative across many markets.

For all the focus on optimising the P&L through improved mix and higher price per pack, very little value has been earned through rising demand, and with prices already far above their pre-pandemic level, that lever is close to spent.

Which leaves many commercial teams facing the same question: where does the next phase of real growth actually come from?

In reality, when markets are growing, mediocre decisions are often hidden by inflation, population and category momentum. When underlying demand is flat, that cover disappears, and the quality of commercial decisions is thrust into full focus.

Growth stops being something the market hands you and becomes something your judgement has to earn.

Reporting systems, and learning systems

The natural reflex, when growth gets hard, is to reach for more information. More data, more dashboards, more analysis. But almost every organisation already holds more data than it uses.

What separates teams is what they do with it. Many have built reporting systems: they describe what happened, build the deck, and explain the result after the event, sometimes over-claiming cause along the way, this went up, that went down, therefore one must have caused the other.

Far fewer have built real learning systems: they state what they expect, record the assumption, test and learn, separate correlation from cause, and carry the answer into the next decision.

Our Wave 2 research findings are pointing to a consistent weakness: most organisations do generate learning of some sort, but analytics remain surprisingly unsophisticated, and relatively few have systems that reliably carry learnings into future decisions.

Better questions, not more answers

The most effective teams I've talked to in the research share one trait, and it is not a bigger data budget. It is commercial rigour. They are genuinely data-led, not merely data-rich.

"The moment our capabilities really started to grow was when the whole organisation said 'we're going to move to a more, data-centric view.'"

— Head of Category & Shopper, EU

That distinction matters more as the tools improve. AI will not just speed the work up, but will also produce a flood of outputs and apparently credible recommendations. But more plausible answers are not the same as better judgement.

The real bottleneck is moving from creating information to validating it, from having the analysis to knowing which answer to trust.

Rigour is what closes that gap, and it is concrete. A reporting team sees penetration falling and recommends a recruitment campaign. A learning team first asks whether the decline is coming from distribution, availability, switching or lapsing users, because the answer decides whether the business needs communication, innovation, better execution or no intervention at all.

The advantage is not simply asking better questions. It is avoiding expensive action against the wrong diagnosis.

From better category to better commercial judgement

The shift that matters is bigger than category. In our research, 94% of practitioners say the traditional category process is still at least partly valid, while 90% say it has to evolve - but it's the direction of that evolution that is the interesting part.

If commercial judgement is the differentiator, the capability to build is the ability to weigh a wide evidence base and make sound decisions from it. In practice that means:

• Widening the lens beyond scan, loyalty and panel data to richer consumer, shopper and digital signals, read together rather than in silos.
• Properly diagnosing root causes, not settling for correlations.
• Planning against scenarios, not a single forecast, and testing early and cheaply rather than defending a plan all the way to launch.
• Giving someone the mandate to hold all of it and form an independent, commercially objective view of what to do next.

This is where category, redefined, can again become the driving force of growth. Not by taking territory from marketing, insight, sales or finance, but by integrating what each of them sees into one commercially coherent view.

Category will not inherit that position automatically. It has to earn it, through broader evidence, a stronger grip on internal and retailer economics, independence from any single brand agenda and a track record of turning analysis into decisions that prove out.

The bottom line

In a market of limited or no volume growth the winners are those who really earn it. And you earn it through the quality of your judgement, not the volume of your data.