How to Raise Menu Prices Without Losing Regulars

August 10, 2026 · 5 min read

Raise prices on high-volume, below-average-margin dishes first, by 5–8%, alongside a menu change rather than on their own. Leave your signature items and known price anchors alone. Most modest increases lose far less volume than the break-even threshold allows.

Nearly every operator raises prices too late and then too much at once. The reason is that a price increase feels like a risk taken on purpose, while cost creep feels like something that happened to you. The arithmetic says the opposite.

How much room you actually have

A price increase drops almost entirely to contribution, because plate cost doesn't move when you change a price. That means you can afford to lose meaningful volume and still come out ahead.

Break-even volume loss

1 − (Old contribution margin ÷ New contribution margin)

The share of volume you could lose before the increase stops paying. The price increase calculator returns it alongside the annual difference.

A $15.00 dish with a $4.20 plate cost
New price	New CM	Break-even volume loss
$15.50	$11.30	4.4%
$16.00	$11.80	8.5%
$16.50	$12.30	12.2%
$17.00	$12.80	15.6%

A dollar on a fifteen-dollar dish means you'd have to lose more than one sale in twelve before you were worse off. In practice, a modest increase on a dish people come for usually loses nothing measurable at all.

Which dishes to raise

In priority order
Priority	Which dishes	Why
1	Plowhorses — high volume, below-average contribution	Demand is proven and the margin gap is the reason to act
2	Dishes whose main ingredient genuinely rose	You have a real, explainable reason
3	Items with no obvious local comparison	Nothing for a guest to price-check against
4	Add-ons, sides and upgrades	Small absolute numbers, rarely noticed, very high margin

Which dishes to leave alone

Your signature dish. The thing people come for is the thing whose price they know. Raise everything around it instead.
Known price anchors. A pint, a house coffee, a lunch special. These are compared against the place next door, consciously.
Anything you raised in the last six months. Frequency is noticed more than magnitude.
Dishes already at the top of their local range. Check before you assume you have room.

When to do it

The timing rule is simple: a price change alongside something new reads as a new menu. A price change on its own reads as a price increase.

With a menu change or seasonal rotation. The best cover there is, and it's honest — you genuinely did change the menu.
With a visible improvement. A better bun, a bigger portion, a new plate. Guests trade money for value they can see.
Never during a bad week. Not after a rough review, not during a slow stretch you're worried about. You'll misread the result and panic.
Not more than twice a year on the same item, ideally.

How much, at once

Worked example · A targeted increase on a $1.25M restaurant

Dishes repriced: 14 of 42
Average increase: 6.2%
Share of covers affected: 48%
Additional revenue, annualised: $37,200
Additional plate cost: $0
Additional contribution: $37,200

Assuming no volume change, which for a 6% move on plowhorses is the realistic base case. Even at a 5% volume loss on every affected dish it clears $25,000.

How to communicate it

Mostly: don't. A new menu with new prices needs no announcement. Where communication genuinely helps is with your staff and with regulars who'll notice.

Brief the team first. They'll be asked. A server who says "I don't know, they just changed it" turns a non-event into a complaint. Give them one honest sentence — "beef is up a lot this year and we held the price as long as we could" — and they'll handle it fine.
Don't apologise on the menu. A printed note explaining your price increase draws attention to the exact thing you'd rather guests skim past.
Do tell regulars, if you have the relationship. "Heads up, prices went up a bit this month" from an owner lands very differently than a surprise on the check.
Never blame a vague force. "Inflation" is an excuse. "Our beef supplier is up 18% since April" is a reason, and it's true.

What to watch afterwards

Units of the repriced dishes, weekly, for four weeks. Compare against the same weeks last year, not against last month — seasonality will lie to you.
Total contribution, not units. Fewer units at a higher margin is the intended outcome, not a failure.
Mix shift. If guests moved from a repriced dish to a cheaper one, check whether the cheaper one actually contributes less. Sometimes it doesn't.
Check average, which should move by roughly what you'd predict. If it doesn't, something else changed too.

How much will I actually lose in volume?

For a modest increase on a dish people come for, usually nothing measurable. The risk is concentrated in price-sensitive, comparison-shopped items — a lunch special, a pint of the same beer the bar next door pours. Those are the ones to leave alone.

Should I raise prices or shrink portions?

Raise prices. Guests notice a smaller portion far faster than a slightly higher price, and shrinking a portion damages the thing they came for. If cost has to come out of the plate, take it from the components guests don't measure — see value engineering — and price any ingredient swap at its substitution cost rather than its invoice price, because a cheaper item with a worse yield often isn't.

What if a competitor hasn't raised theirs?

They will, or they'll close. Holding a price your costs no longer support isn't competing, it's subsidising. What matters is whether your dish is worth its price, not whether it matches the place down the street.

How do I know it's time?

When a dish's food cost percentage has drifted more than about three points above where you priced it, or when its contribution margin has fallen below your menu average. Both are visible the moment your recipes are costed against current invoices.

Know which prices to raise, and by how much.

Marji tracks every dish's margin against your real invoice costs and names the specific ones that crossed your target — with the cost evidence, so you know it's a real move and not a guess.

Start free →