In value retail, floor area is inventory and inventory is revenue. Anything that occupies floor without displaying product — a feature wall, a seating pocket, a wider aisle, a demonstration stage — has to be paid for out of display capacity.
That is the fundamental tension in every retail project we work on, and it is usually argued about badly. One person says the store feels cramped. Another says the store needs the stock. The discussion resolves on whoever is more senior, or on whoever is more tired.
The way out is to make display density an explicit, agreed number at the very start — before anyone has drawn a plan or fallen in love with a concept. We call it the display ratio, and we set it in the first fortnight of every store project.
Why it changes the conversation
Setting the ratio early does not constrain the design. It does the opposite: it makes ambitious moves defensible.
When we proposed a circular demonstration stage at the centre of a 30,000 sq ft flagship, the question stopped being "does this look good" and became "what does it cost in display capacity, and can we recover it elsewhere?" Because the model existed, the answer was arithmetic rather than opinion — the stage costs a known amount, adjusting the perimeter recovers most of it, and the net effect was small enough to accept. That conversation took an afternoon.
Without the model, the same conversation becomes a series of meetings in which a designer insists the stage is important and an operator insists the store needs the stock. Nobody is wrong. Nobody can prove anything either.
A design argument that can be settled with a number should never be settled with a meeting.
The part everyone gets wrong
Display ratio is a ratio, which means it has a denominator — and the denominator is where almost every version of this metric quietly falls apart.
Measure display area against the sales floor alone and you get a flattering number that cannot be compared between stores, because the amount of back-of-house space varies enormously with the shell, the landlord's riser positions and how much stock the format has to hold. A store with a cramped stockroom will look denser than one with a generous stockroom even when their shopfloors are identical.
Measure it against the whole tenancy and you get a more conservative figure that is genuinely comparable — and one that aligns with the commercial reality, since the business pays rent on the whole tenancy, not just the part customers see.
We use the second. It is the less flattering choice and the only useful one.
If you take one thing from this: a display ratio quoted without its denominator is not a number, it is an opinion. Whenever someone reports a density figure — a landlord, a contractor, a previous consultant — the first question is always what it was measured against.
Models drift
Display-ratio models are usually built as spreadsheets, and spreadsheets accumulate hand-typed exceptions. This is the most common defect we find when we audit an existing model, and it is worth naming because it is so easy to miss.
A model that computes consistently can be verified by recomputing it. A model with even one undocumented override cannot — from that point on, checking it requires knowing which exceptions were deliberate, which usually means asking someone who has left the company. And because these sheets get copied from store to store, a single wrong constant propagates silently across a whole portfolio.
Sometimes an override is genuinely correct; some fixtures are not well described by geometry alone. The answer is not to forbid exceptions. It is to make every one of them visible and explained, and to keep the model tied to the same module standard the drawings and the costings use. A ratio model that has drifted from the drawings is worse than having no model at all, because it is trusted.
Where this sits in our work
The display-ratio model is part of how we run the strategy phase. We build it against your fixture standard and your format, calibrate it to your existing stores, and use it to test every subsequent design decision — including ours.
The specific method, the calibration and the model itself are studio work product, and they come with the engagement rather than out of an article. What is worth taking from this piece is the principle: agree the number, agree what it is measured against, and do it before the first plan is drawn.
This piece describes YOI Design's approach in outline. Detailed methodology, benchmarks and the models themselves form part of our client work.