10 Menu Design Mistakes and What Smart Restaurateurs Do Instead

The costliest menu design mistakes are decisions made on folklore: a golden corner nobody has found, a dollar sign that probably does nothing, a number of items someone once heard was ideal. Ugly fonts come a distant second. Good menu design is mostly clarity, order and honesty about what the kitchen can do.

Some of the usual rules hold up. Some are weaker than the trade press suggests. Where a study is available, this post says what it found and how small it was. Where none could be checked, the advice stays and the number is left out.

Why is so much menu advice folklore?

Because menus are easy to talk about and hard to test. A restaurant has one menu, one room and one season, so a single result is rarely clean. Many "rules" travel from article to article without anyone returning to the original study.

That matters, because a rule you cannot check can cost you real effort. Before you redesign, and before you pay anyone to do it, ask of each tip: who tested it, where, and how many guests? If nobody can answer, treat it as an idea to try on one section, not a principle. That habit alone can save a reprint.

The ten mistakes

1. Designing around a sweet spot

The belief is that guests look first at the top-right of the page, so the high-margin dish goes there. A 2012 eye-tracking study led by Sybil Yang found readers moved through a mock menu sequentially, like a book, and that gaze did not linger in any one place. The trade report does not give a sample size (the published paper analyses 25 participants), and the menu was a mock one, so it is evidence, not proof.

Instead: order the menu the way a guest decides, from lighter to heavier dishes or by course, and give every section a clear first line. Use menu engineering to decide which dishes deserve attention, then use position, spacing and short copy, not a supposed hot corner.

One more point from the same study: the area that drew the least attention held restaurant information and a salad list, and the researcher suggested the mock diners may simply not have cared for salads. Interest, not geometry, may have driven it. If a section is skipped, ask whether the dishes in it are the problem before you move the section.

2. Obsessing over the dollar sign

You may have heard that dropping the currency symbol makes guests spend more. The Cornell field experiment behind that claim was run at one cafe at the Culinary Institute of America, with data collected in 2007 and 201 parties analysed. It found no significant difference between a "$XX.XX" format and prices spelled out in words. Only the numerals-only format showed higher checks (about 8 per cent per person), and the overall price-format effect was not significant. The paper's abstract also reports a significant reduction in spending with money cues, which the body of the paper does not clearly support: the abstract contradicts the body there.

Instead: choose one clean, consistent price format, set it in the same size as the dish name, and move on. Spend your effort on whether the price is right. Our guide to pricing a menu covers that.

3. Cutting the menu to an arbitrary number

The fashionable advice is that fewer choices always sell more. Research is less tidy. A 2009 paper running three experiments, including one on choosing restaurants, found the too-much-choice effect was less robust than previously thought. In its restaurant study, choice rates were about the same (30 per cent with 5 options, 35 per cent with 30).

That does not mean a large menu is wise. An RMIT summary of research on Australian cafes links shorter menus to easier stock control and less waste, though it gives no measured effect.

Instead: set the limit by your kitchen. Include what your team can cook well at peak, and cut the dish that is slow, waste-heavy and rarely ordered.

4. Describing every dish in the same way

A page of dishes, each with five ingredients in a comma list, reads like a spreadsheet. Guests cannot tell the signature from the filler.

Instead: give each dish a short name, a plain line on how it tastes or is cooked, and no more than a few ingredients that matter. Use the same voice as your floor team. If a server would not say it aloud, do not print it.

A quick test: read three dishes aloud to someone outside the business. If they cannot say what they would be eating, rewrite the line. Keep one sentence for the dish you most want guests to try, and let the others be shorter.

5. Hiding dietary and allergen information

When guests with allergies have to ask, the menu has shifted its work onto the floor, and a nervous guest orders less or leaves. Rules on allergen labelling differ by country, and you should check yours with the relevant authority.

Instead: mark dietary and allergen information clearly in one consistent place, keep a current allergen matrix in the kitchen, and make sure it is updated when a dish changes.

6. Ignoring how the menu is read

A menu is read in dim light, by tired people, often on a phone. Small grey type on a textured stock is a design choice that works against the guest.

Instead: test it. Print it, take it into the room at dinner, and read it at arm's length. If a person over fifty cannot read it comfortably, raise the type size and contrast.

Check the phone version too. Open the menu on a mid-range handset on a mobile connection, and see whether the first section loads before a guest loses patience. Note that a menu that needs pinching and zooming is a menu that is not being read.

7. Running a column of aligned prices

When prices sit in a neat column down the right edge, the eye runs down the numbers and not the dishes. This is a design judgement from experience, not a studied result, so treat it as a test.

Instead: place the price at the end of the description, in the same text line and weight, and see whether the questions your servers hear change.

8. Leaving the menu out of the sales data

Menus are often redesigned from taste alone. Then no one knows whether the new version helped.

Instead: record the sales mix before and after each change, and change one section at a time. The monthly review is the cheapest consultancy you will ever buy.

Keep the record plain: a sheet with the date, what changed, the section affected and the weekly sales of the dishes in it. After three or four changes you will know more about your own guests than any general article can tell you. Compare like with like, the same weekdays and a similar season, or the result will mislead.

9. Letting the website and the printed menu disagree

A PDF on the website with last season's dishes and old prices is a broken promise before the guest arrives. It also reads poorly on a phone.

Instead: keep one source of truth and publish it as text on a page, with the date. When a dish is sold out or a price changes, change the master first and let print and web follow, so the guest never meets two versions of the truth. Our checklist for a restaurant website covers menus, hours and booking.

10. Promising what the kitchen cannot deliver

The menu says "slow-cooked" and "to order", and the kitchen has a short rush and a small line. The mismatch shows up as long waits and sold-out dishes.

Instead: read the menu against a real Saturday night, with the actual booking count. For each dish ask which station makes it, how long it takes and what else is on that station. Remove or rework anything that breaks the rush.

How do you audit your own menu in an hour?

Take a printed copy, a pen and last month's sales mix, and run the checks below. Mark each item pass, fix or test. Anything marked fix goes on a short list; anything marked test is tried on one section only.

Do the audit with someone who works the floor. They know which dish needs explaining every night, and which section guests skip. Those two facts are worth more than any rule in this post.

A common fix is also the cheapest: remove the bottom five dishes on your sales mix, if they are also slow or wasteful, and see what the kitchen does with the time it gets back. You may find the rush runs quicker before you find sales are lower. Measure both, and keep the decision reversible.

How do you test a menu change without guessing?

Change one thing, record the date, and compare sales mix and guest comments across the same weekdays. That is the whole method. It is simple and slow and works.

A menu is a set of instructions the kitchen has promised to follow.

A practical monthly routine:

1. Pull the sales mix by dish and by section.
2. Note the dishes that sell well and cost well, those that sell but cost badly, and those that do neither.
3. Read the last month of guest feedback and server notes for menu confusion.
4. Pick one fix from the list above, make it, and write down the date.
5. Review it after four weeks, and keep or revert it.

If the menu changes with the seasons, see our seasonal menu planning post for a calmer way to rotate dishes.

Key takeaways

• Treat famous menu rules, sweet spots and dollar signs included, as hypotheses. The evidence that can be checked is thinner than the folklore.
• Order the menu by how guests decide, and use short, plain descriptions in your floor team's voice.
• Set menu length by what the kitchen can cook well at peak, not by a magic number.
• Make dietary information easy to find, and keep print, website and kitchen in agreement.
• Test legibility in the real room, and change one thing at a time against your sales mix.