A real analysis, anonymized: what we found, and what we can’t claim.
A coffee chain with hundreds of units in two cities, analyzed from the outside. Data through September 2026, covering the last twelve, six, and three months. It’s a sample — not the full analysis, and not everything we measure — and it’s anonymized, with volumes shown as ranges. Not everything shown here is part of the free first analysis, which covers two units, each against its local competition, and the chain as a whole in those two cities.
- Pricing unit by unit, not the brand as a whole. None of the units with enough evidence reaches its city’s median, but not all are equally far from it: the ones furthest behind can be carved out or made conditional, and the rest don’t warrant the same penalty.
- Not paying — or charging — for a recovery that isn’t visible. Flat for four quarters while rival chains improve.
- Treating service separately from product. Service is where the chain trails its rivals most, and that’s addressed with staffing and process; the smaller product gap calls for work on the menu.
- Using sales data where the public evidence falls short. In the city shown in section 02, three in four units have too little evidence to judge one by one: buyers can request those units’ sales figures in the data room; sellers can prepare them before being asked.
That’s what the evidence shows and how we read it. The decision, and the price, are yours.
Scope · a single unit and the coffee shops around it
It trails all eight coffee shops around it, and almost no one talks about it: fewer than forty people in twelve months. More than a thousand talk about the coffee shop a few blocks away.
- Position shows the actual distance from the unit.
- Size is how many people talked about each one in the last twelve months. That’s the difference the headline number doesn’t show.
- The number is where its customers place it today.
- Orientation on the radar is illustrative. No chain is identified here; in the analysis of your chain, every location is shown with its name and address.
| Location | Current standing | People talking · 12 mo | Most praised for | Complaint share · main complaint |
|---|---|---|---|---|
| The unit · headline number: 3.7 | 3.1 | under 40 | the staff and the atmosphere | more than half · the staff |
| Under 100 m | 4.4 | 200–250 | the staff and the coffee | 1 in 25 · no dominant theme |
| 100–200 m | 4.7 | 50–100 · just opened | the staff and the coffee | 1 in 10 · no dominant theme |
| 100–200 m | 3.8 | 50–100 | the staff and the speed | 1 in 3 · no dominant theme |
| 200–300 m | 3.8 | under 40 | the staff and the coffee | 1 in 3 · no dominant theme |
| 200–300 m | 4.9 | over 450 | the staff and the food | 1 in 15 · the wait |
| 300–500 m | 4.7 | under 50 | the staff and the coffee | 1 in 12 · no dominant theme |
| 300–500 m | 4.7 | over 1,000 | the staff and the coffee | 1 in 29 · the wait |
| 500–600 m | 4.5 | under 50 | the staff and the atmosphere | 1 in 7 · no dominant theme |
Swipe sideways to see all columns →
Recovering here doesn’t mean gaining half a point: it means becoming a real option again for customers choosing between it and its neighbors. That recovery takes money and time, and someone will pay for both — better to settle who before signing.
Eight nearby coffee shops, all eight ahead of it. The two that dominate the conversation — over 450 and over 1,000 people in twelve months — sit at 4.9 and 4.7. Fewer than forty people talked about the unit in twelve months: enough evidence to know where it stands — around 3.1, even though its headline number says 3.7 — but not to pin it down to the nearest tenth.
In a coffee shop, service is what earns praise: at all nine, it’s the first thing satisfied customers mention. What separates the nine is the share of complaints — more than half of what’s said about the unit, one in twenty-five at the neighbor across the street. And at the unit, both the best and the worst of what customers report come down to the same thing: who’s behind the counter. At six of the neighbors, complaints have no dominant theme; at the two with the most customer comments, it’s the wait — a complaint that comes with success.
And what we tested before accepting it.
Was it the neighborhood? The neighbors aren’t affected. The format? The one less than a hundred meters away sells the same thing. Who’s doing the talking? Both sides of the street are measured with the same yardstick — and across the whole chain, over four hundred customers weighed in on both the chain and its competition in the last twelve months (see 05 below). And we don’t hide the trend: over twelve months, the unit is still below every one of its neighbors. Over the last six, the direction is down; with this little evidence, we can say which way, not how far.
Strength of evidenceStrong on the unit’s current standing; on its trend, only an indication — direction, not magnitude.
Where you use itYour site-visit plan for due diligence, and what you request on that unit in the data room.
Scope · the same chain in a single city: its units with enough evidence against every coffee shop in that city
None of the chain’s units with enough evidence reaches the median for coffee shops in its city. Not one.
THE CHAIN
Its local market
Local market median · 4.4
How customers judge each unit today
NO UNIT WITH ENOUGH EVIDENCE REACHES THE MEDIAN
A brand-wide discount penalizes every unit equally, including those closest to their local market. With the distribution in front of you, the conversation shifts from how much to take off the price to which units go in, which come with conditions, and which stay out.
The chain sits in the bottom half of its local market. Between its best and worst units with enough evidence — same brand, same city, same twelve months — there’s a spread of almost two points. And none of them reaches the local market median.
Everything points to the units, not the market: if it were the city, the neighbors would score low too. They don’t. Problems like these aren’t priced with one across-the-board discount — they’re priced unit by unit.
And what we tested before accepting it.
Was it the city? The neighbors don’t score lower. The type of business? Against chains of the same format, the finding narrows but holds. And what weakens it, stated plainly: of the chain’s units in this city, three in four don’t have enough evidence over twelve months to be measured individually. They aren’t plotted; they’re counted and disclosed. The ones that can be measured are enough to see the pattern; we don’t guess at the rest.
The free first analysis shows two units of the chain you’re looking at, each against its local competition, and the chain as a whole in those two cities. Request it →
Strength of evidenceStrong for the units we can measure; the rest are counted and disclosed.
Where you use itThe deal perimeter and the purchase agreement: what’s in, what comes with conditions, and what’s out.
Scope · the entire chain: hundreds of units in two cities, against the rival chains and independents in its local market
No. It’s been flat for four quarters — with independents nearly a full point higher and rival chains six-tenths higher.
| Group | Units | Customer feedback · 12 mo | Oct–Dec 2025 | Jan–Mar 2026 | Apr–Jun 2026 | Jul–Sep 2026 |
|---|---|---|---|---|---|---|
| The chain | hundreds | about 10,000 | 3.65 | 3.65 | 3.56 | 3.59 |
| Rival chains | over 30 | about 1,500 | 4.04 | 4.01 | 4.13 | 4.20 |
| Independents | about 250 | about 38,000 | 4.62 | 4.59 | 4.54 | 4.54 |
An ongoing decline gets negotiated as a risk. A full year without a single quarter of recovery gets priced as a fact: it’s what you’re buying — and what you’re selling.
The chain holds at around 3.6 in all four quarters. The independents in its local market sit between 4.5 and 4.6; the rival chains between 4.0 and 4.2, and rising. No quarter shows a confirmed recovery.
This isn’t a dip a hundred-day plan will fix: it’s a level that holds quarter after quarter. At the unit level, over twelve months: among those with enough evidence to track quarter by quarter, none shows a confirmed decline, and three improve. The rest are flat.
The caveats, stated plainly.
The rival chains improve over these quarters and the chain doesn’t, so the gap widens: from about four-tenths of a point to about six-tenths. Each quarter adds more than two thousand customer comments on the chain, so the chain-level trend is solid even where individual units’ trends aren’t.
Strength of evidenceStrong for the chain; unit by unit, only an indication.
Where you use itThe growth assumptions in your investment memo, and the 100-day plan.
The gap between its service, product, and atmosphere results is nearly twice the rival chains’. Customers think more of what it sells than of how it’s delivered.
Service, product, and atmosphere, taken separately, reveal something the overall number hides: whether the problem lies in what’s sold or in how it’s delivered.
What this means for any improvement plan. The independents in its local market deliver a consistent experience: their results on the three dimensions are within 0.03 of each other. Rival chains, 0.16. The chain, 0.30. Against rival chains, it trails on everything, most of all on service: −0.54 on service, −0.44 on atmosphere, and −0.40 on product. A delivery problem is fixed with staffing, shifts, and process; a product problem, with the menu and the recipes. Here it takes both, with service the bigger lift: two plans, two price tags — for buyer and seller alike. Results on individual dimensions are measured separately from the overall number and run higher here, which is why we compare them across chains rather than against the 3.6. This measures how customers judge delivery, not the operation itself: we haven’t set foot in any location, and the analysis tells you where to look, not what’s happening inside.
Strength of evidenceStrong: more than six thousand ratings of the chain on each dimension.
Where you use itThe 100-day plan: what staffing and process fix, and what changes to the product offering fix.
Every competitor comparison meets the same objection: “those aren’t the same customers.” Here, for more than four hundred people, they are.
The same people judged both the chain and its competition: same yardstick, same season. They gave the chain 4.07; the competitor, 4.41. These customers are more generous than most — which is why both figures sit above the chain’s 3.6 — so what counts is the gap between them.
Why this test carries weight. Because the same people judge both sides, differences in audience, season, and personal standards cancel out — and still, more than one in four clearly prefers the competitor. The remaining 57% show no clear preference either way. Only the aggregate goes into the document; never individuals, or groups small enough to identify anyone.
Strength of evidenceStrong: more than four hundred customers judged both the chain and its competitor.
Where you use itThe competitive risk in your memo — and, if you’re lending, what to monitor.
“Very dissatisfied customers are twice as common at this chain as in its local market.”
It was striking and, on its own, true. We could have made it the headline.
Against businesses of the same type, it disappeared. The share turned out to be essentially the same. What we’d found wasn’t specific to this chain but to its entire category, and as a headline it would have been misleading.
A report where everything confirms the thesis isn’t a report — it’s a sales pitch.
Name the chain. We’ll show you what its price doesn’t reflect.
There’s no charge for the first one, and no obligation. You’ll hear back from a real person within two business days.
Request your first analysis Or ask first for an anonymized example from your sector → Or email us directly: contact@acumaris.com