Two restaurants, the same town, the same £34 average spend. One gets steady reviews calling it good value. The other gets a review a fortnight saying it is overpriced.
The food is comparably good in both. The portions are similar. Nobody is being overcharged.
The difference is that the first one looks like a £34 restaurant before anyone opens a menu, and the second looks like a £22 one. The complaint is not about the price. It is about the distance between the price a guest silently predicted on the way in and the price they actually paid.
The guess happens in about eight seconds, and it is mostly furniture
Diners estimate a price band from physical evidence, fast, before any food information is available. The inputs are depressingly mundane and remarkably consistent:
- The frontage. Signage quality, whether the glass is clean, whether there is a menu displayed outside, how the door opens.
- Seating. Hard chairs and bench seating signal cheap. Upholstery, space between tables and a chair with arms signal expensive. This one is close to deterministic.
- The table surface. Bare laminate against bare oak against linen is three different price bands and guests read them correctly nearly every time.
- Weight in the hand. Cutlery, glassware, the menu itself. A thin laminated card and a heavy paper menu set different expectations about the bill, with identical dishes printed on them.
- Lighting and sound. Bright and loud is cheap and fast. Dim and quiet is slow and expensive.
- Who greets you, and whether anyone does. Being met and seated sets a higher band than ordering at a counter, regardless of what arrives.
None of that is about your cooking. All of it is in place before the first drink.
Which is why these reviews read strangely
A genuine price complaint names the money and compares it to something:
£19 for a chicken burger when the place down the road does a better one for £13.
A signalling mismatch almost never does. It talks about the room, and the price turns up as a conclusion rather than a premise:
Food was actually really good but for these prices you expect a bit more than being squashed on a bench under a strip light. Felt overpriced for what it is.
Read that again: food was actually really good. They are not disputing the cooking. "For what it is" is the whole complaint — they placed you in a band, you charged outside it, and the gap got filed under price because that is the vocabulary available.
Other tells, in the same family:
Lovely meal but £140 for four with the plastic glasses and paper napkins, I don't know.
Can't fault the food. Just didn't feel like a £40 a head experience.
Tasted great, looked like a canteen.
When a review praises the food and condemns the price in the same breath, you are almost never looking at a pricing problem. You are looking at a presentation problem that has been reported in the only units guests have.
The cheap end of the fix is very cheap
The useful thing about signalling is that the inputs are weighted oddly. Guests respond strongly to a handful of tactile details that cost very little, and barely at all to some expensive ones.
Worth doing, roughly in order of return per pound:
Replace the napkins. Paper to cotton is the single highest-leverage change in this entire list. It is touched, it is in front of the guest for the whole meal, and it moves the perceived band by a tier.
Replace the glassware. Thin, light, mismatched glasses undo a lot of work. A decent wine glass changes what a £9 glass of wine feels like.
Weigh the cutlery. Lightweight cutlery reads as canteen. You can buy heavier for a few hundred pounds and it lasts years.
Reprint the menu on heavier stock, with fewer words. Dense descriptions on thin paper signal low price. Fewer items, more white space, heavier paper, no clip art, no exclamation marks, and — importantly — align the price column properly or drop the column entirely.
Turn the lights down and get the music right. Free. A room at a lower light level with music under the conversation moves the band more than most refurbishments.
Have someone greet and seat. Also free, if you have the bodies. Being walked to a table is a price signal.
Not worth doing first, despite being what people reach for: a full refurbishment, new flooring, a wine list rewrite, or — the expensive mistake — cutting prices. Dropping the price to meet the signal closes the gap from the wrong side and permanently lowers what you can charge for food you are already producing. Discounting into a value complaint usually makes it worse.
The other direction, which nobody talks about
Signalling above your price band is also a failure, and it generates a different review you may already be getting.
A venue that looks expensive and charges moderately gets guests who arrive braced, order defensively, drink tap water, skip the starter, and are then pleasantly surprised. Pleasantly surprised is nice. Ordering two courses instead of three is a revenue problem, and no review will ever tell you about it, because nobody writes a review about the thing they did not order.
If your average spend is well below what your room implies, the problem is not perception. It is that your menu is not asking for the money the room has already prepared the guest to spend.
How to tell which problem you have
Sort your last three months of reviews that mention price into two piles, using one test: does the review name a specific item and its price, or does it describe the room?
| The review says | You are looking at | First move |
|---|---|---|
| "£19 for a burger" | A real price-anchor problem | Compare that item, locally, honestly |
| "for these prices, on a bench" | A signalling gap | Napkins, glassware, light |
| "didn't expect that bill" | A surprise, not a price | Check the charges a guest discovers late |
| "fine but wouldn't rush back" | Possibly neither | Read what follows "though" |
Only the first pile is about money. The third is its own thing and has a different cause — every charge a guest can discover too late is worth auditing separately. The fourth is where retention actually leaks, and it hides in the caveat at the end of four-star reviews.
The limits of reading your own room
You are the worst-placed person to judge your own signalling. You have seen the room ten thousand times and you know what the food costs to make. The useful substitute is a photograph: take one of your dining room at seven in the evening, show it to three people who have never been, and ask them to guess the price of a main. Their answer is the band you are signalling, and if it is six pounds under your actual main, that is your gap in pounds.
Some mismatch is deliberate and correct. A tiny, scruffy place with extraordinary food and queues out the door is reading the signal wrong on purpose, and everybody understands the trade. That works when the food is genuinely exceptional and the scruffiness is clearly a choice. It does not work when the room is merely tired.
Changing signals changes your guests. Raise the band and some of your regulars were there precisely because it felt unpretentious. Move in increments and watch covers, not just reviews.
One change at a time, and give it six weeks. Price language in reviews is slow-moving and seasonal. Swapping napkins and glassware and the lighting in the same week leaves you unable to say which mattered, which matters because the next venue you open will need to know.
OMMU separates the price complaints that name a number from the ones that describe the room, which is the split that decides whether you are having a menu conversation or a napkin conversation. Three months of reviews and two piles on a table does it just as well — the trick is making the sort before you reach for the price list.