Raising prices without losing guests

Price increases fail for a predictable reason: they are deferred until they are large, then applied evenly across the menu, then announced by a reprint that changes everything at once. Each of those three choices makes the increase more noticeable than it needed to be.

## Why deferral is the expensive option

Every month a price is held below its cost base, the eventual correction gets bigger. A business that adjusts a small number of items twice a year is invisible. A business that adjusts every item once every three years is a story guests tell each other.

The instinct behind deferral is understandable — nobody wants to be the venue that raised prices. But the alternative is not holding prices; it is quietly reducing what is on the plate, which regulars notice faster than a price.

## Where the increase should land

Not everywhere. A menu has items where price is closely watched and items where it is not.

**Anchor items** — the dishes a guest orders every visit, the ones they would notice on the bill — should move last and least. These are the reference points on which the perception of value rests.

**Category-typical items** in the middle of a list absorb a change more easily, particularly when the change lands on the price point rather than dragging it across a threshold.

**Add-ons, sides and beverages** are usually the most under-priced part of a menu and the least closely watched. A small correction here often produces more contribution than a larger one on mains, because the volume is high and the reference price is weak.

**Delivery prices** are a separate decision entirely, because the cost base is different.

## Thresholds are real

Guests read prices in bands. Moving an item from 38 to 42 crosses a threshold; moving it from 42 to 45 does not. Where an increase would cross a band, it is often better to make a larger single move later than a small one now, or to change the item — portion, garnish, accompaniment — so that it is honestly a different offer at the new price.

## The specification alternative

Sometimes the right answer is not a price change but a specification change: a different cut, a different accompaniment, a smaller portion with a better presentation. This is legitimate when it is a real reformulation and dishonest when it is a quiet reduction of the same dish. Regulars can tell the difference, and they will attribute a quiet reduction to the brand, not to the supplier.

## A staged approach

1. **Establish the current cost base.** You cannot price from a costing sheet that is a year old.
2. **Identify the anchors.** Ask the floor team which three or four items a guest would notice on the bill. They will know.
3. **Move the least-watched items first.** Sides, add-ons and beverages, in a small number of steps.
4. **Correct the middle of the menu next**, avoiding threshold crossings where possible.
5. **Leave the anchors until last**, and when they must move, consider whether the item should also change.
6. **Re-measure after each stage.** Volume by item, not just total revenue — a price change that holds revenue while losing covers is a warning, not a success.

## What to watch afterwards

Watch mix, not just revenue. The most common consequence of a price change is not that guests leave; it is that they substitute — moving from a high-contribution item to a lower one. That shows up as flat revenue and reduced contribution, and it is invisible unless you are reading the mix.

## In short

Price increases are not a single decision but a sequence. Taken in small, deliberate steps against a current cost base, and measured on mix rather than revenue alone, they are one of the least disruptive levers a venue has. Deferred until they are unavoidable, they become the most disruptive.