Everybody agrees value complaints should be sorted before they are acted on. Almost nobody can tell you how to sort a specific one, which is why the exercise tends to end with the reviews in a spreadsheet and no decision taken.
So here is the sorting done out loud on twelve lines, all of them the sort of thing that arrives in a normal month. Three buckets:
- Pricing — the number is genuinely wrong for what is on offer
- Execution — the experience did not reach the standard the price set
- Communication — the guest was surprised, or arrived with the wrong expectations
The point of the exercise is that the same word can land in different buckets, so keyword sorting does not work. You have to read for the mechanism.
The twelve
1. "£14 for a burger and no chips, genuinely take the mickey." → Communication. The number is not the complaint; the discovery is. A menu line reading "burger £14, add chips £4" produces this review. "Burger with chips £18" does not, at identical revenue.
2. "Lovely food. Three of us, £120, wouldn't rush back at that." → Pricing. No fault named, no surprise, no service failure. They enjoyed it and judged the total against the alternatives. This is the clean signal and it is rare.
3. "Waited 25 minutes for mains. For these prices I'd expect better." → Execution. The price appears only as the yardstick. Fix the wait and this review does not exist.
4. "Service charge added automatically for a table of two — cheeky." → Communication. Almost always. A policy stated on the menu and mentioned on seating generates a fraction of the complaints of the same policy discovered on the bill.
5. "Nice, but you can get the same for £8 less on the high street." → Pricing, tentatively. Note what to check: is the comparison to a genuine peer, or to a different category? If ten reviews make this comparison and all name the same kind of venue, it is a positioning finding.
6. "Four ravioli. Four. For £19." → Pricing. Countable food is judged by counting and presentation cannot rescue it. Either the count goes up or the price comes down.
7. "Beautiful plate, still hungry after." → Execution, in the plating sense — see portion complaints are usually about the plate. Beautiful is the tell. They approved of the presentation, so the food-to-plate ratio did the damage.
8. "Charged £3.50 for tap water in a jug." → Communication, and the amount is irrelevant. This review will be quoted back at you for years over £3.50.
9. "Second visit was nowhere near the first. Same price though." → Execution, and the most urgent of the twelve. A returning guest with an accurate memory is reporting a decline. Consistency complaints from regulars predict retention loss better than anything else in your review set.
10. "Pricey but honestly worth it for the occasion." → None of the three. A positive review that mentions price. It gets swept into value analysis by keyword searches and should not be. Occasion-framed spend is a different judgement entirely.
11. "The set menu is great value, à la carte is a joke." → Pricing, with a structural note: your own set menu has become the reference price and everything else now reads as inflated. Discussed in why discounting makes value complaints worse.
12. "£11 cocktails and they were fine, I suppose." → Execution. "Fine, I suppose" at that price is a quality complaint wearing a price complaint's clothes. Check the spec and the pour before you look at the price.
What the distribution told us
Pricing: 4 (2, 5, 6, 11). Execution: 5 (3, 7, 9, 12, and arguably 7 twice). Communication: 3 (1, 4, 8). Not applicable: 1 (10).
Two things follow, and they are the reason to do this at all.
Communication is a quarter of the complaints and costs nothing to fix. Three menu wording changes — chips included or clearly priced, service charge stated on seating, water policy printed — remove three of twelve complaints this week for the price of a reprint. No other intervention available to you has that ratio.
Only one of the four pricing complaints is actually about the menu being too expensive in general. Number 6 is a single dish. Number 11 is a self-inflicted anchoring problem. Number 5 needs corroboration. Number 2 is the only clean one. Which means a general price reduction would be answering roughly one twelfth of the incoming complaints.
How to run it on yours
Take thirty value-flavoured reviews. For each, ask in this order: was the guest surprised? (communication) — did something specific fall short? (execution) — did they judge the total against alternatives with nothing else named? (pricing). Stop at the first yes. Reviews that reach the end without a yes are usually not value complaints at all.
Then count. If communication is not your largest bucket, you are unusual.
Thirty reviews takes about forty minutes and is worth doing once by hand, because the reasoning is the part you keep. Keeping the count current across several months and several sites is the part that does not survive contact with a trading year, and it is what OMMU does instead — the buckets stay counted, and what arrives is which one is growing.