How to raise menu prices
on your website
without losing reservations

Leer este artículo en español →

Read this with your menu open in front of you, because that is the page this article is really about. A price raise isn't one decision — it's one decision repeated down the menu, dish by dish, and each dish gets exactly one of four verdicts: raise it, hold it, re-engineer it, or sunset it. Run that call per line item instead of running a flat percentage across the whole sheet, and the numbers can go up without your reservation count coming down. That is the entire move, and the rest of this is how to make the call.

You don't want to raise prices. Your costs have already gone up, your suppliers' costs have already gone up, and somewhere in the back of your head you know the math stopped working about four months ago. You've been holding the line on menu prices because you remember what happened the last time the restaurant across the street tried a live increase in one shot, and you'd rather eat the margin than lose a regular.

I understand the fear. I also think the fear is usually solving for the wrong risk. The question isn't should I raise prices — at some point, if you keep eating every cost increase, you close. The question is which items, by how much, and in what order, so that the numbers on the menu can go up without your reservation count going down. Treat each dish as a branch, not a row in a spreadsheet, and the four-verdict spine does most of the deciding for you.

This is the playbook the company behind this library uses to reprice a menu and its website in the same week. The math is restaurant-specific, the execution is website-specific, and the two halves only work together. These numbers are illustrative, drawn from patterns across independent restaurants — not from a published dataset. Your own menu will vary. The shape of the move is the part that generalizes.

The anxious math nobody wants to run

Before anything else, do this one calculation. Pull the last ninety days of sales. Take your food cost percentage — what you're paying suppliers divided by what you charge guests. If it used to be 30% and it's now 34%, that four-point drift is your profit margin walking out the back door. For an independent restaurant doing $1.2M a year, that four points is around $48,000 — more than a reliable sous chef's salary, evaporating quietly.

Now do the uncomfortable one. If you raise your menu by an average of 6% across the board, and you lose 3% of your covers as a result, your revenue barely moves — but your margin recovers almost entirely, because the covers you lose at the margin were the lowest-spending ones anyway. That's the whole game. The trick is getting the 6% without the 3% becoming 15%.

Getting it wrong looks like this: a flat percentage bump on every item, a new PDF menu uploaded over the weekend, and a Monday morning where regulars open their phones, see the same burger that was $17 last week now at $19, and quietly book somewhere else. The numbers went up, the reservation count went down, and now it's your fault — not inflation's.

Getting it right looks like three tiers.

The logic is simple: guests don't notice price changes evenly across the menu. They notice them most on the items they've compared to other restaurants. They notice them least on the items that only you make. The playbook is to take your raise in the places the guest is least likely to feel it, and to leave the anchor items as-is so the menu still feels like "yours" when they open it.

The tiers, up close

Before the tiers up close, sit with the math that makes the whole move worth the nerve. The reason a few points of weighted lift matters so much is that food cost comes straight off contribution: on $100 of menu sales, every point the food-cost share drops is a dollar that stays in the room. The widget below takes the article’s own drift — a menu costed at 30% that crept to 34% — and lets you drag the food-cost share back. Watch contribution recover dollar-for-dollar. These are illustrative figures, but the trade is exact.

Tier 1 — Anchors you never touch

Every menu has three to five items that are load-bearing in the guest's memory. The burger, the Caesar, the house chicken, the signature cocktail at $14. These aren't just menu items — they're reference prices. When a guest opens your menu, they scan for those numbers first, usually unconsciously, to decide whether you're "the same price you were" or whether something has changed.

Leave the anchors alone. Eat the margin on them. This is counterintuitive — your instinct is to raise every item a little — but the anchor items are where any price movement gets talked about. A $1 bump on the burger will show up in a Yelp review within a week. A $2 bump on a pasta dish that's on the menu alongside twelve other pastas, nobody notices.

There's a secondary benefit. Holding anchor prices gives you a true line in future conversation: "We've held our burger at $17 since 2024." That's a sentence you can say out loud. Guests who care about prices respect it. Guests who don't care about prices never think about it. Everybody wins a small thing.

Tier 2 — Core items, raised 4 to 6 percent

The core of your menu is where the weighted-average raise actually happens. Most of your entrées. Most of your pastas. Most of your pizzas, if you're a pizza place. These are the items guests order regularly but don't have an exact-dollar memory of. They remember "it was around twenty bucks." A move from $21 to $22.25 is inside that memory tolerance. A move from $21 to $25 is not.

Four to six percent is the sweet spot. Below 4%, you barely recover margin and you've spent political capital for nothing. Above 6%, guests start to notice the menu "feels" more expensive, even if they can't point to a specific item. Stay in that band on the core and you'll recover roughly two thirds of the margin you've lost to cost inflation, silently.

One tactical note: when you raise core items, don't round to the dollar. A jump from $21 to $22 feels bigger than a jump from $21 to $21.75. The brain anchors on the leading digit. Going from twenty-one-something to twenty-two-something is a different price bracket in the guest's head. Twenty-one-something to twenty-one-something-bigger is the same bracket.

Tier 3 — Signature items, raised 8 to 12 percent

Your signature items are the ones guests can't compare to anywhere else. The tasting menu. The chef's special. The one dish you're known for that doesn't exist in the same form down the street. These are price-inelastic in a way the rest of your menu isn't — guests aren't choosing between your version and someone else's, they're choosing whether to order your version at all.

Raise these 8 to 12 percent without flinching. If the chef's tasting menu was $85, take it to $92 or $95. The guest who's there for the tasting menu is already past the price-sensitivity threshold — they walked in committed. What they're paying for is the experience, and that experience is worth more than you usually charge for it anyway.

The common mistake here is the opposite direction: treating signature items with more caution than core items because "they're the expensive ones." Wrong. The expensive items are the ones with the least price-comparison pressure. They're the easiest place to recover margin.

The three tiers tell you where the raise lives. But you still have to make the call one dish at a time, and not every dish gets a number — some get held, some need recipe work before any price moves, and a few should come off the menu entirely. Walk this spine per line item. The first verdict that fits is the verdict; you don't keep going down the branches once one lands.

The fear of a price raise is mostly fear of the wrong reservation showing up at the wrong table. The solution lives on the menu page, not on the spreadsheet.

The two verdicts operators mix up are raise and re-engineer, because both are answers to "this dish isn't carrying its margin." The difference is whether the guest has somewhere cheaper to go on your own menu. If the dish stands alone in its category, a 4–6% bump is inside the bracket and the right move is to raise. If a near-twin sits next to it at a lower price, the bump just shifts orders down the menu, and the right move is to re-engineer the plate cost so the same number holds. Same symptom, opposite fix — here they are side by side.

The website moves that make a price raise stick

The menu is half the work. The other half is the website the guest looks at before they decide to book. Most restaurants raise prices and then change nothing else online, which is how a 5% raise becomes a 12% perceived raise in the guest's head. Three website moves do a lot of the heavy lifting.

1. Don't change the numbers — change what's around them

If your menu page just shows a list of dishes with prices, a price bump reads as a naked increase. Guests scan, they see a bigger number, they flinch.

If your menu page shows a dish with a proper description — the sourcing, the cooking method, the one detail that makes this version yours — the price reads in context. It's not "$24 for a chicken sandwich," it's "$24 for a pan-roasted half chicken with brown-butter jus, from a farm you can name, served in a way you can describe." The price hasn't changed, but the frame the guest evaluates it in has.

When you raise a price, rewrite the description at the same time. Two sentences. Something specific. The guest who opens your site this week and sees "Chicken $24 — pan-roasted half-bird, brown-butter jus, two-hour brine" reads that as a restaurant investing in its product. The guest who sees "Chicken $24" reads a restaurant that just got more expensive.

2. Update at least one photograph

This sounds superficial. It isn't. Menu items that have visually moved — a new plating, a new garnish, a new photograph — get priced against the new image, not the old price memory. A guest scrolling your menu page on a phone will evaluate "the new version of this dish at $24" very differently from "the same dish I had last year, but $24 now."

You don't need to re-shoot the whole menu. Pick the three items that moved the most in price and have your kitchen plate a fresh version for an hour on a Tuesday afternoon. Natural light, clean table, phone camera is fine if you're handy with it. Ninety minutes, three new photos. Those three photos neutralize the price conversation on those three items for the next six months.

3. Remove stale discounts before you publish the new prices

Nearly every restaurant site has at least one stale "promo" somewhere — a happy-hour box that's still pointing at the old pricing, a banner about a summer special that ended in October, a PDF menu titled "Spring 2024" still linked from a sidebar. These are death on a price raise, because they tell the guest "the real price is lower than what's on this page."

Spend an hour before you publish the new menu walking your own site as if you were a first-time visitor. Every price that shows up anywhere. Every banner. Every PDF. If it doesn't match the new menu, kill it. A guest who finds a $17 burger on the homepage but a $19 burger on the menu page has just been trained to believe your restaurant is dishonest, even though you were planning to fix the homepage next week.

Every stale discount on your site is an argument against the real price. You're not saving the guest money, you're giving them a reason not to trust the new one.

What to do this week

If you've been putting off a price raise for three months and this post has you finally willing, here's the honest seven-day sequence. It's not glamorous and it doesn't need to be.

A note on adjacent decisions: if the price raise is part of a bigger compensation rework — DC’s Initiative 82, post-pandemic wage pressure, the question of whether to move to a service charge — there’s a paired piece on service charges vs tipping that walks the math on the same $200 check across three compensation models. The price-raise sequence below is model-agnostic, but the choice of model changes how the new prices read on the check.

1. Monday. Pull your last ninety days of sales by item. Tag every item as anchor, core, or signature. You'll disagree with yourself on a few — that’s the point, it’s a decision, not an algorithm.
2. Tuesday. Draft the new prices. Anchors held. Core items up 4–6%, avoiding round-dollar jumps. Signatures up 8–12%. Weighted-average it against volume — you’re aiming for a weighted 5–6%, not a flat 5–6% on every item.
3. Wednesday. Rewrite the descriptions on any item you raised more than a dollar. Two sentences, specific. Sourcing, method, the detail only you know.
4. Thursday. Re-shoot three dishes with the biggest price moves. Phone + natural light + hour of kitchen time.
5. Friday. Audit your own website cold. Every page that shows a price, every old promo banner, every PDF menu. Remove anything that contradicts the new pricing.
6. Saturday. Publish. Menu page first, homepage blocks second, PDF menu last. Tell your host and servers the three items that moved most so they’re ready for the one guest who asks.
7. Sunday. Watch your reservation count for the next two weeks. If it drops more than 5%, you took too much. If it holds or moves less than 2%, you didn’t take enough — and you can come back for another 3% in ninety days.

That's the whole playbook. Three tiers, three website moves, seven days.

You're not trying to extract more from the guest — you're trying to stop losing money on the guest you already have. The distinction matters. A restaurant that's quietly going broke because it won't raise prices closes the same as a restaurant that raised prices the wrong way. One of those outcomes, the operator at least got to choose.

Price it to survive. Describe it to land. Photograph it to reset. Then let the regulars show back up, the same way they always have.