When a store underperforms, the post-mortem usually lands on something visible. The graphics were weak. The lighting was dim. The layout was confusing. Those are real problems, but they are symptoms. The decisions that actually determined the outcome were taken months earlier, often by people who were not thinking about design at all.
This is a list of the six we see most often. None of them are exotic. All of them are cheap to get right at the correct moment and expensive to fix afterwards.
1. The lease is signed before anyone looks at the shell
A tenancy is chosen on rent, footfall and adjacency to an anchor. Those are the right criteria. But the physical shell — the column grid, the slab-to-slab height, the position of the risers, the shape of the frontage — determines what kind of store can exist there, and it is almost never assessed before signature.
A column landing three metres inside a narrow frontage will compromise every layout you ever draw in that unit. A slab height that will not accept a suspended feature ceiling removes an entire category of design response. A riser in the middle of what should be the main run forces circulation into a shape customers will not follow.
When to take it: before signing. A half-day survey and a test-fit plan cost a fraction of one month's rent, and they tell you whether the unit can hold the format you intend to put in it. If the answer is no, you either negotiate the shell works into the deal or you walk.
2. The category plan is decided in a spreadsheet
Category plans arrive as a list: so many bays of hardware, so many of homeware, so many of toys. The list is derived from sales data, which makes it feel objective. It is objective — about the past, in a different store.
What a spreadsheet cannot encode is adjacency behaviour: that a customer buying paint will look at brushes if they are next to the paint and will not go and find them two aisles away; that putting the highest-margin impulse category at the back guarantees it is seen by fewer people; that a family group splits at the entrance and reconverges at the till, so the categories each member wants need to be reachable from a shared route.
When to take it: as a first draft in the spreadsheet, then immediately tested in plan. Bay counts and adjacencies should be resolved together, in space, before the layout is fixed. This is the single most valuable week in a retail project and it is routinely skipped.
3. The shopfront is treated as signage
Most shopfront briefs are really signage briefs: how large is the logo, how bright, what colour. But the shopfront's job is not to be seen — it is to convert someone walking past into someone walking in, and that is a threshold problem, not a graphics problem.
The questions that matter are about permission and legibility. Can you see into the store from an angle, at walking pace, or only head-on? Is there a physical or visual barrier — a step, a heavy frame, a dark zone just inside — that makes entry feel like a commitment? Is the first thing visible from outside something a passer-by actually wants?
On the flagship we documented at The Mines, the frontage runs 14 metres with no doors and no raised threshold, and the illuminated ring ceiling is drawn to be legible along the mall from either direction rather than only from directly in front. Those are threshold decisions. The wordmark is the easy part.
When to take it: at concept, alongside the store plan — because the shopfront determines what the first ten metres of the plan have to do.
4. The lighting budget is set as a percentage
Lighting is frequently budgeted as a fixed share of the fit-out cost, then value-engineered when the tender comes back high. Both of those are mistakes, and the second is worse.
Lighting is the highest-leverage money in a retail fit-out. It determines whether merchandise reads from a distance, whether a feature registers as a feature, and whether the store looks expensive or cheap — which is largely a question of contrast, not of finish quality. Cutting it is the fastest way to make an otherwise good scheme look like a budget one.
The practical defence is documentary. We draw concealed LED on its own dedicated layer, in magenta, so it cannot be quietly dropped from a package without somebody noticing. It is a small convention that has saved more design intent than any argument we have ever made in a meeting.
When to take it: lighting strategy at concept; lighting budget as an absolute figure, not a percentage; and any value engineering assessed against what it does to contrast ratio, not against what it saves per fitting.
5. Merchandising is not designed at all
This is the most common and the most damaging. The architecture is designed to a high standard, photographed on handover day, and then filled by a shop team working from a planogram and a promotional calendar that nobody showed the designer.
Within a month there are three sizes of promotional sign competing on the same wall, a temporary display blocking the sightline the whole plan was built around, and price tickets in four different formats. The store does not look designed any more, because the design only ever covered the empty state.
The fix is to design the system rather than the snapshot: a fixed graphic hierarchy with defined sizes and heights for each tier, designated hero positions where the store is permitted to shout, and protected sightlines marked as no-build zones. Then hand it over as a document the shop team can actually follow.
When to take it: during design development, as a deliverable — not as advice given verbally at handover.
6. The first store is treated as a one-off
If you intend to open more than about five stores, your first one is not a store. It is a prototype, and it should be resourced like one.
That means budgeting for the fact that you are debugging a format: expect to redraw things, expect the fixture family to change after you see it built, expect to discover that one zone does not work at all. It also means documenting the outcome properly — a maintained drawing register, typical details, a module standard, a finishes schedule — so that store two starts from an asset rather than from a photograph of store one.
Clients who skip this reach store six with six incompatible sets of drawings, no agreed standards, and a contractor in every state pricing from scratch. The design cost of the format is then paid again, in instalments, forever.
When to take it: at the point you decide to open more than one.
The pattern
Every item on this list is the same failure in a different costume: a decision taken by whoever happened to own that phase, without anyone asking what it does to the rest of the project.
That is an organisational problem more than a design one, which is why we structure our work to span the whole arc — strategy through construction documentation — rather than handing off at each boundary. The hand-offs are where the value goes.
A store is not a series of decisions. It is one decision, taken repeatedly, by people who need to be in the same conversation.
This piece sets out professional judgements formed across YOI Design's retail projects. It does not report a survey or a controlled study, and it is written as opinion rather than measurement.