Why identical stores perform differently — and what actually…

Take two of your stores. Pick them to be as alike as you can make them.

Same brand over the door, same range on the shelves, same prices on the tickets. Similar footfall through the door, similar store size, staffing built off the same template. On paper they are the same shop, printed twice.

Now put their numbers side by side. One is comfortably ahead of plan. The other is quietly behind, and has been for a while. Same everything — different results, month after month.

Every multi-store retailer has this pair. Usually they have dozens of them. And the honest answer to why is worth more than almost anything else on the P&L, because one of the two explanations you can't do a thing about, and the other you can.

The puzzle: same inputs, different outputs

Here's what makes it nag at you. If the two stores really do share the same range, prices, brand, and roughly the same traffic, then most of the usual levers are held flat. You didn't give one a better product, a discount, or more people past the window.

And yet the outputs diverge, and they keep diverging. This isn't noise that washes out over a quarter. It's a settled, repeating gap between two shops that were supposed to be interchangeable. Something is happening between the same inputs and the different outputs. The whole question is what.

The easy answer — and why it's a cop-out

Ask around head office and you'll get the same reply almost every time: it's the location. Different catchment. Older demographic. Weaker foot traffic quality. The other one's just got a better postcode.

And that's not wrong. Location is real. Catchment is real. A store on a dying high street is playing a harder game than one in a booming centre, and no amount of good work fully closes that.

But notice what the location answer quietly does. It explains the whole gap using the one variable nobody can touch. You can't move the store. You can't re-pick the catchment you signed a fifteen-year lease on. So if the difference is all market, then the difference is fate, and the underperforming store is just having weather done to it.

That's a very comfortable conclusion. It asks nothing of anyone. Which is exactly why it gets reached for first, and exactly why it's usually overweighted. The market sets the ceiling a store could hit. It does not explain how far below that ceiling the store actually lands. Two shops with the same catchment can still be a long way apart — and now you can't blame the postcode, because they share it.

The part everyone underweights: execution

Here's the variable that gets skipped, because it's the uncomfortable one: whether the floor actually reflects the plan. Day after day. Not on the day of the visit — on a wet Wednesday when nobody's watching.

The same brief lands in every store, and it lands differently in every store. The planogram that's followed to the centimetre in one shop is interpreted in the next and quietly ignored in a third. The promotion that goes up clean and on time in one goes up late, half-built, in the wrong bay somewhere else. The window reset that took an hour here took "we'll get to it" there.

None of this shows up as a dramatic failure. It shows up as drift. The floor matches the plan on Monday morning and has slipped by Wednesday afternoon, and nobody decided that — it just erodes, one small compromise at a time, under real pressure with real staffing.

And the pressure is the point. A 2026 survey of 227 retail leaders found only 36% could say more than three-quarters of their initiatives execute correctly and on time. Forty-three percent could name sales they'd lost to poor execution. The single most-cited reason things don't get done wasn't laziness or defiance — it was not enough staff to do the work. The plan assumes a floor that's fully resourced and fully briefed. The actual floor is stretched. The gap between those two is execution variance, and it is enormous.

This is the part worth being precise about. Execution variance is not good stores versus bad stores, and it is certainly not good people versus bad people. The frontline is not the problem here. It's the difference between a store that has the context to get the plan right — knows what "done" looks like, why it matters, what to prioritise when the day goes sideways — and a store that's guessing. Same brief, same effort, wildly different result, because one team was set up to land it and the other was left to interpret it. Variance is stores with context versus stores without it.

Why averages hide all of this

The reason execution variance stays invisible is arithmetic. You manage the network by its average, and the average is the one number engineered to hide the thing you most need to see.

Your best store and your worst store sit in the same mean. One is carrying the other. The blended figure looks like a middling, stable, unremarkable business — and it's a lie, in the precise sense that no single store actually experiences it. You are not running a number. You are running a distribution, and the width of that distribution is the whole story.

We've watched networks spread across a four-fold gap between their best-executing store and their worst — same brand, same range, same prices. Average them and that spread vanishes into one tidy line on a slide. The line goes up two points and everyone nods. Meanwhile the real question — why is the bottom quartile a fraction of the top quartile, and how much of that is fixable? — never gets asked, because the average was never built to raise it.

The gap between your top and bottom quartile is not a reporting curiosity. It's the map of your controllable upside.

How to actually tell the two apart

Here's the hard bit. Sitting at head office, you genuinely cannot separate the market gap from the execution gap. Both produce the same symptom — a store that's behind. The postcode and the poorly-run floor look identical in the sales report. That's precisely why "it's the location" wins by default: it's unfalsifiable from a distance.

You can only tell them apart by seeing what actually happens on the floor and connecting it to the outcome — which is exactly the record most retailers don't have (the store data blind spot). Capture what the store really did — was the range set the way it was meant to be, did the promo go up, did the reset hold to Wednesday. Understand it against the result — did the stores that executed clean outperform the ones that drifted, holding traffic and catchment flat. Only then can you say, for a given store, this much is the market it's in, and this much is what happened on the floor — and act on the half you can move.

Without that, you're guessing. And the guess conveniently always lands on the variable you can't fix.

The prize

This is the cheapest same-store-sales growth in retail, and it's already inside the building.

You are not opening new stores, adding range, or cutting price. You are getting your worst-executing stores to run more like your best ones — closing the controllable half of the gap, the part that was never about the postcode.

Do that across a whole network and the maths is unlike anything a new-store programme can touch. No capital, no new leases, no new SKUs. Just the plan actually reaching the floor, and staying there past Wednesday — in the stores that were quietly leaving money on the table the whole time.

The market sets the ceiling. Execution decides how close you get. One of those you can change. Start there.

FAQ

What drives store performance?

Two forces. The market a store sits in — its catchment, footfall and demographics — and how well it executes the plan on the floor. The market sets the ceiling a store could reach; execution decides how close it actually gets. You can't change the first, but the second is usually the larger share of the gap between two similar stores, and it's the half you can move.

Why do some stores outperform others with the same range and prices?

Because "the same range and prices" doesn't mean the same floor. The planogram followed to the centimetre in one store is half-ignored in another; the promotion that goes up clean on Friday goes up late, in the wrong bay, somewhere else. That execution variance — stores with the context to get the plan right versus stores left to guess — compounds, week after week, into very different numbers.

How do you measure the gap between your best and worst stores?

Look at the distribution, not the average. Rank your stores on the metric that matters — sales per square metre, in-store conversion, or a per-store execution score — and compare the top quartile against the bottom. The width of that spread, not the blended mean, is the real size of your controllable upside.

Piing is the context engine for retail: it turns store-floor reality into a structured, real-time record and connects every action to the outcome it drove — so you can finally see how much of the gap between your best and worst store is the market, and how much is the part you can fix. See your estate come alive →