Improve store layouts using customer complaints
Shoppers often can't find what they came for, and they tell you — in reviews, in surveys, and at the service desk. NEXT reads that feedback across your stores and groups the complaints by where people actually get lost. What you get is a clear picture of which area confuses shoppers, how many stores show the same pattern, and what to change — routed to store design.
Most of this feedback already exists. It just sits in a thousand separate comments that no one has time to read store by store.
What the navigation complaint cluster looks like
When the same kind of "I couldn't find it" complaint repeats across enough stores, NEXT groups it into one readable picture and sends it to the people who fix the floor.
Store area
Rear-left quadrant — seasonal and promotional end caps
Where shoppers get lost
No clear sightline or signage from the main aisle to the seasonal zone; shoppers reach the back wall and double back
What shoppers say
"Came in for the garden stuff in your ad. Walked the whole store twice, gave up, asked someone at the till."
"Why is the sale section hidden behind the home aisle? You can't see it from anywhere."
Affected stores
14 of 38 large-format stores show the same pattern; concentrated in the post-2022 refit layout
Commercial exposure
These stores under-index on promotional sell-through versus the estate average — the feedback points at findability, not price or stock
Signal strength
Strong and consistent on the seasonal end caps; weaker and mixed on general aisle navigation
What it points to
Shoppers can't see the promotional zone from the main path. A sightline or overhead signage fix, templated across the 14 refit stores, is the candidate change to route to store design.
Example output based on grouped review, survey, and service-desk feedback. The grouped signal is ready before the review, not reconstructed during it.
Why layout problems surface late today
A store dashboard shows conversion dipped in a handful of locations; it doesn't tell you that shoppers can't find the seasonal range. You see the number move, then you go digging through comments yourself to guess why. Ask an assistant and you get the loudest recent review, not the repeating pattern across a region.
The feedback that explains the dip is scattered. A shopper writes a one-star review about getting lost. Another mentions it in a post-visit survey. A third tells a colleague at the service desk, who may or may not log it. By the time anything reaches a store ops review, the original wording is gone and only a vague sense of "wayfinding issues" survives — too soft to act on, too easy to deprioritise.
A dashboard reports the number; it doesn't tell you why it moved. NEXT brings the grouped shopper feedback to the people who fix the floor.
How this compares to the tools you already know
Approach | Where the evidence lives | What Retail Ops does at decision time |
|---|---|---|
Store conversion dashboards | In a metric that dropped | See conversion fell, then dig through comments to guess the cause |
AI assistant / chatbot | Wherever you think to ask | Get the loudest recent review, not the pattern across stores |
NEXT | In a continuously updated record of shopper feedback, grouped by location and routed to store design | Read the cluster, decide the fix, track whether the complaints fade |
How NEXT detects this
NEXT reads where shoppers already speak — reviews, post-visit surveys, and service-desk notes. It keeps a running record of that feedback and groups navigation and findability complaints by store and by area. When enough of the same complaint repeats across enough stores to clear the threshold you set, NEXT writes a short cluster: where shoppers get lost, which stores show it, the shopper wording, and a candidate layout or signage fix. That cluster is routed to store design and operations, and lands where the team already plans. You decide whether the fix ships, when, and across which stores.
What changes for the Retail Operations lead
Today, a wayfinding problem reaches you as a feeling. A regional manager mentions that two stores "seem confusing," or a quarterly survey flags signage in general terms. You can't tell whether it's two stores or twenty, or whether it's worth a refit line item.
With the cluster attached, you open a single picture: the seasonal end caps are unfindable in the 14 refit stores, here is what shoppers said, and here is the candidate fix. The problem that looked like two anecdotes turns out to be one repeatable layout fault across a third of the large-format estate — which means one templated signage change, not fourteen separate investigations.
The conversation in the store ops review changes. Instead of arguing whether the complaints are real, the team scopes the fix: overhead signage, an end-cap relocation, or a sightline change, applied to the stores that share the pattern. NEXT already supports retail and operations teams at companies like Action and Rituals in connecting customer feedback from reviews, surveys, and support to operational decisions.
The prioritisation call stays with you. NEXT supplies the grouped shopper feedback and the candidate fix; what ships, and where, is still your call.
Downstream effects
Store design starts from grouped shopper feedback rather than a single angry review or a hunch from one visit. The brief arrives with the wording attached, so design isn't reconstructing the problem from scratch.
Recurring faults across stores become visible, so a fix can be templated to the affected format instead of solved one store at a time. A signage change validated in three stores can roll to the eleven others that share the layout.
The change can be tracked. Once a fix ships, the same feedback stream shows whether the navigation complaints fade in those stores — closing the loop without a separate audit.
Where the human stays in control
NEXT does not change a floor plan. It groups the feedback and routes a candidate fix; people decide what happens next. You set how many stores and how many complaints a pattern needs before it routes, and you can require a human to review clusters before they reach store design. That is configuration of the thresholds and routing, not sign-off on every comment. The judgment — whether the fix is worth the capital and disruption — is still yours.
What to configure first
Source coverage. The cluster is only as good as the feedback feeding it. Make sure reviews, post-visit surveys, and service-desk notes are connected; if service-desk feedback isn't captured in writing, low-traffic stores will look quieter than they are.
Store grouping. Decide how stores are grouped — by format, refit generation, or region — so a layout fault tied to one floor plan doesn't get diluted across stores that don't share it.
Threshold. Set how many stores and complaints constitute a real pattern. Too low and one bad week routes noise; too high and a genuine fault waits behind the threshold.
Handoff and timing. Confirm who owns the store-design handoff and how often clusters land — aligned to your store ops or refit planning cycle, not a random drip.
Where this breaks down
Thin feedback in low-traffic stores
Small or quiet stores generate few reviews and surveys. A real navigation problem there may never clear the threshold, so coverage skews toward your busiest locations. Treat silence as missing data, not a clean store.
Complaints that aren't actually layout
"I couldn't find it" sometimes means the item was out of stock or staff weren't around to ask — not that the floor plan is wrong. Clusters that mix stock and staffing into layout will send store design chasing the wrong fix. Calibration and a human read reduce this; they don't remove it.
Vague location language
Shoppers say "the back somewhere" more often than "rear-left quadrant." When the wording is too loose to localise, the cluster points at a store but not a specific area, and design still has to walk the floor to confirm.
One-off versus pattern
A single store after a refit may generate a burst of complaints that settle as shoppers adjust. Routing a capital fix on a two-week spike wastes effort. Let the pattern persist across the window you set before acting.
FAQ
How is this different from our store conversion reports?
A conversion report tells you sales dipped in certain stores. It doesn't tell you why. NEXT groups what shoppers said about getting lost, shows how many stores share the pattern, and points at a candidate layout or signage fix. The report flags the symptom; the cluster explains the cause and routes it to the people who can fix the floor.
Does NEXT change the store layout automatically?
No. NEXT groups the shopper feedback and routes a candidate fix to store design and operations. People decide whether to change signage or layout, in which stores, and when. Nothing on the floor changes without a human acting on it.
How many complaints does it take before something routes?
You set that. NEXT routes a cluster when the same navigation complaint repeats across enough stores and enough shoppers to clear your threshold. Set it too low and one bad week creates noise; set it too high and a real fault waits. Most teams tune it against how many stores share a format.
What if shoppers don't describe the location precisely?
That's a real limit. When wording is vague, the cluster can point at a store but not a specific area, and store design may still need to walk the floor. NEXT groups what shoppers actually said — it can't invent a precision the feedback doesn't contain.
Can it tell layout problems apart from stock or staffing issues?
Partly. Calibration and a human read reduce the chance that out-of-stock or staffing complaints get grouped as layout faults, but the separation isn't perfect. That's why clusters can be held for review before they reach store design, so a person can confirm the problem is actually findability.