Restaurant Menu Design Tips That Boost Sales

Most menu redesigns start with a template and end with something that looks better and sells about the same. That’s because the parts people enjoy choosing (the paper stock, the script font, the little illustrated radish) are not the parts that change what guests order.

The decisions that move money are duller and more specific: which items sit in the first third of a section, how prices are typeset, how many dishes a category holds, and whether the thing a guest reads on their phone at 7pm says the same as the thing on the table. This guide works through those decisions in the order you’d make them. If you’d rather start from a concrete layout and work backwards, our restaurant menu design ideas collection has specific layouts, covers and section structures you can copy outright.

The Bottom Line: Design your menu around how guests really read it, top-left and downward, section by section, rather than the “golden triangle” eye-tracking research failed to find. Cap each category at roughly five to seven items, put your best-margin dish in the first two slots of its section, print prices as plain numerals without dollar signs, and write a short specific description for anything you want to sell. Then check your menu on a phone, because for most restaurants that screen now gets more traffic than the table.

What makes a restaurant menu design work

A menu works when a guest can find the kind of thing they came in for within a few seconds, and when the item you most want them to order is the easiest one to say yes to. Everything else is decoration.

That means a good menu design does three jobs at once. It organises dishes into categories a stranger would guess correctly. It directs attention toward the items that carry your margin. And it reads cleanly in the conditions it gets used in: dim light, a phone screen, a guest who is mid-conversation and not studying. If you’re starting further back than that, our breakdown of what a restaurant menu is and what it has to contain covers the components before you start arranging them.

For proportion: there are more than a million restaurants in the United States employing around 15 million people, according to figures compiled from the National Restaurant Association and others in this roundup of restaurant industry statistics. Your menu is competing for attention in that crowd, and it is one of the few things in the building you can change on a Tuesday afternoon for free.

The useful consequence of framing it that way: you can tell whether a design decision is working. “Does this look nicer?” has no answer. “Can someone find the vegetarian mains in three seconds?” does.

How guests actually read a restaurant menu

For years the standard advice was the “golden triangle”: guests supposedly look to the middle of the page first, then the top right, then the top left, so you should park your profit items in those spots. Eye-tracking research found no evidence for it.

In a study published in the International Journal of Hospitality Management, Sybil Yang used infrared retinal eye-tracking to record how diners actually scanned menus, and found no menu “sweet spots” at all (Yang, 2012, IJHM 31:1021–1029). Guests read menus much more like they read a book: starting at the top left, moving down and across in sequence. What the research did turn up was the opposite: a “sour spot”, the stretch of a menu holding general information about the restaurant and the salad list, which readers skimmed past (phys.org coverage).

Two practical consequences:

• Position within a section matters more than position on the page. Since guests read each category in order, the first couple of lines in a category get the most attention. That’s where your margin items go.
• Don’t spend your best real estate on prose about the restaurant. A paragraph about your supplier relationships at the top of page one is sitting in the sour spot. Put it on the back, or on your website.

Where to place high-margin items on a restaurant menu

Put the dish you most want to sell in the first or second slot of its category, and give it slightly more visual weight than its neighbours.

“Slightly” is the operative word. The standard moves (a box, a subtle background tint, a small “house favourite” tag, bold on the item name only) work because they break the pattern of the list. Apply them to six items and the pattern is the box, so nothing stands out. One highlighted item per section is plenty; two is the ceiling.

A few placement rules:

• One highlight per category. The point is contrast, and contrast doesn’t survive repetition.
• Anchor the section with a premium item. Listing a higher-priced dish near the top of a category makes the mid-priced options read as reasonable rather than cheap. This is the most reliable piece of menu psychology: guests judge a price against the other prices in front of them, not against an absolute sense of value.
• Don’t bury the thing you’re best at. If your short rib is the reason people come back, it should not be the ninth line of the mains.
• Leave the last slot alone too. Sequential reading means the final item in a short list also gets read properly. In a category of five, both ends are decent positions.

How to write menu item descriptions that sell

Write short, concrete, and specific. Four to twelve words naming the thing that makes the dish distinct beats a paragraph of adjectives.

The reason is simple: a description’s job is to answer the one question stopping a guest from ordering. For an unfamiliar dish, that question is “what actually is it?” For a familiar one, it’s “why yours?” A line like “slow-cooked overnight, served with the pickled slaw” answers both faster than “a delicious and mouthwatering house speciality prepared with love”.

There is also commercial evidence that descriptions pull their weight. DoorDash’s analysis of more than 15,000 small-business merchants on its platform between April and June 2022 found menus carrying item descriptions earned up to 18% more in monthly sales than menus without them (DoorDash Merchant Learning Center). That’s a platform reporting on its own marketplace rather than independent research, so treat the exact figure with some caution. But the direction is consistent with what operators see when they describe dishes instead of just listing them.

What to avoid: describing every single item. If all forty dishes carry three lines of copy, the menu becomes a wall and the descriptions stop signalling anything. Describe the items you want to sell and the items people won’t recognise. Let “Caesar salad” be Caesar salad.

How to present prices on a restaurant menu

Set prices as plain numerals, in the same size and weight as the item name, aligned close to the dish rather than in a far-right column.

Each of those three choices has a reason behind it.

Start by dropping the dollar sign. Cornell’s Center for Hospitality Research tested three price formats on 201 guests at the St. Andrew’s Café: numerals with a dollar sign ($14.00), numerals alone (14.), and prices written out in words. Guests who saw plain numerals spent around 8% more than guests seeing either of the other two formats, and the “$” and “dollars” versions produced statistically similar spending (Yang, Kimes & Sessarego, Cornell CHR, 2009). The currency symbol appears to act as a small reminder of spending money.

Avoid setting prices in a column. When every price lines up on the right edge, you have built a price list, and guests read down the column choosing the cheapest acceptable option. Setting the price a couple of spaces after the description, or on the same line in the same weight, makes the dish the unit of comparison rather than the number.

Skip the leader dots too. The row of dots between item and price exists to help the eye travel to a right-hand column, the thing you just decided not to build.

One caveat on all of this: a bare numeral only reads as a price when there’s enough surrounding context for it to be obvious. On a very sparse menu, 14 can momentarily read as a quantity. Test it on someone who hasn’t seen the menu.

How many items should a restaurant menu have?

Keep each category to roughly five to seven items, and let total length follow from the number of categories you need.

Choice overload is well documented outside restaurants. In Iyengar and Lepper’s much-replicated supermarket study, shoppers offered 24 varieties of jam bought in about 3% of cases; shoppers offered 6 varieties bought in about 30%. More options produced more browsing and fewer decisions. A guest facing eighteen pasta dishes behaves the same way: they take longer, they feel less sure, and they fall back on the safe familiar thing rather than the interesting higher-margin one.

Menu engineer Gregg Rapp, who has worked on menus for chains and independents for decades, puts the practical ceiling at seven items per category. As a rough shape by format:

• Fast casual: about 20 to 30 items total, because speed of decision is the whole proposition.
• Full service: roughly 40 to 60, spread across enough categories that no single list runs long.
• Fine dining: often under 30, sometimes far under, with the kitchen’s range expressed through changing dishes rather than a longer list.

Shortening a menu also has a kitchen benefit that shows up faster than the sales benefit: fewer SKUs, less prep, less waste, and fewer items sitting in the walk-in waiting for the two people a week who order them.

Menu typography and legibility rules

Set body text at a size you can read at arm’s length in your own dining room light, use at most two typefaces, and reserve the decorative one for headings.

The most common menu design failure is illegibility rather than ugliness: a script face at 9pt, grey ink on cream card, read at dusk by someone who left their glasses at home. Some specifics that hold up:

• Two typefaces maximum. One for item names and headings, one for descriptions. A third reads as indecision.
• Decorative type for headings only. Script and display faces work on a category header of two words. They fail across a line of ingredients.
• Mind the contrast, not just the size. Dark text on a light ground stays readable as light drops. Mid-grey on beige doesn’t, whatever the point size.
• Give it margins. White space is what makes a category scannable. A menu crammed to the trim edge forces guests to work, and working guests default to the familiar.
• Left-align the body. Centred paragraphs of description are slow to read because every line starts in a different place.

Should a restaurant menu include food photos?

Photos belong on digital and takeaway menus, and usually not on a printed in-house menu. And a bad photo is worse than none.

The split comes down to context. A guest at a table has their senses engaged already; a photo in that setting mostly signals a price point, which is why mid and upper-tier restaurants avoid them. A guest ordering from a phone has nothing to go on but your words, and that’s where imagery does real work. DoorDash’s same 2022 merchant analysis reported menus with a photo on every item earning up to 44% more in monthly sales than menus without them, with header images and logos adding further lift. Again, these are the platform’s own numbers on its own marketplace, so read them as directional.

The quality bar is the thing operators underrate. One poorly lit plate shot on a phone against a dark table does more damage than an unillustrated list, because it sets an expectation the dish then has to beat. If you can’t shoot the whole menu consistently, shoot your six best sellers properly and leave the rest as text.

Print menus vs digital and QR code menus

Most restaurants now need both: a printed menu for the table and a digital one for everywhere else. They’re different design problems, not the same file at two sizes.

The digital side is no longer a minority format. Around 52% of US restaurants have switched to QR code menus, and about 51% of consumers say a QR menu has a positive effect on their satisfaction, per this compilation of restaurant industry statistics. That makes the QR layout a primary design surface rather than a contingency.

The one mistake worth naming: publishing a print-layout PDF as your QR menu. A two-column A3 sheet opened on a phone forces pinching and panning, which is exactly the friction a QR menu was supposed to remove. If the digital menu is a PDF of the print menu, you have a scanning problem rather than a menu.

On formats, a digital menu means HTML: text that reflows to the screen, with categories a thumb can jump between. Most menu platforms generate this from the same item data that produces your print file, which is also what stops the two drifting apart.

Designing a restaurant menu for takeout and delivery

Build your off-premises menu as a shorter, separate menu, designed for a phone and for food that has to survive a journey.

This matters more than most menu guides admit, because off-premises is no longer the side business. The National Restaurant Association’s 2025 Off-Premises Restaurant Trends report found nearly 75% of all restaurant traffic now happens off-premises, with two-thirds of Gen Z and millennial consumers calling takeout essential to their lifestyle and close to 60% using takeout or a drive-thru at least weekly (National Restaurant Association, April 2025). Designing that menu as an afterthought means designing badly for most of your orders.

What changes when the menu is for travel:

• Cut the items that don’t travel. Anything that goes soggy, separates, or needs to hit the table in ninety seconds is a refund waiting to happen. Leaving it off the delivery menu is a design decision, and the cheapest one available.
• Lead with what packs well. The items that arrive exactly as intended should sit at the top of each section, where sequential reading puts them in front of guests first.
• Name the portion. A guest at a table can see the plate. A guest on a phone is guessing, and a guessed-wrong portion becomes a complaint.
• Flag the add-ons that survive. Sauces, dressings and sides packed separately are easy upsells and easy to describe in a line.
• Design for thumb width. One column, large tap targets, headings that stick. Test it on your own phone before anything else.

If your own drivers or a delivery partner handle the last leg, the menu is also the document that sets expectations about what arrives. Matching your delivery menu to what travels well is the single cheapest way to keep off-premises orders from generating the complaints that eat the margin.

How can I design a restaurant menu for free?

Yes, and for most independents a free tool is sufficient. The constraint is almost never the software.

Canva and Adobe Express both have free tiers that will produce a print-ready PDF, with template libraries you can start from rather than designing from a blank page. Free versions typically limit premium assets and brand controls rather than output quality, so a one-off reprint costs nothing but your afternoon.

The question to ask before you pick is how often your prices move. Design tools treat a menu as artwork: change a price and you re-edit the file, re-export and reprint. Menu platforms treat it as data, generating the print file, web menu and QR menu from one list of items, so one edit updates all three. If you reprint twice a year, artwork is fine. If you have market pricing or weekly specials, the data approach saves the slow drift where your printed menu, your website and your QR code all quote different prices.

Either way, budget for printing rather than software. Decent card stock, and a laminate or sleeve if the menu lives on a table, is where the money goes in menu design and printing.

Restaurant menu design trends for 2026

The direction of travel is menus that are shorter, cheaper to change, and built phone-first.

Four things to watch:

• Smaller menus, changed more often. Tighter categories with a rotating section has become the standard answer to both cost volatility and choice overload. It also makes the kitchen’s life easier.
• Phone-first as the default layout. The design that gets drawn first is increasingly the single-column mobile one, with the print menu derived from it rather than the other way round. Collected restaurant marketing statistics put the share of diners who research on their phone before visiting at around 89%, and roughly 60% of patrons now use QR code menus or the offers attached to them. For most restaurants the menu that gets read first is the one on a screen.
• Dynamic pricing and live availability. Menus held as data can hide a sold-out item or adjust a market price without a reprint, which is pushing operators away from static artwork.
• Plainer typography. The ornate, heavily illustrated menu has given way to clean type and real white space, partly for legibility and partly because it reproduces on a phone screen.

Underneath all four is the same shift: a menu is now a dataset with several outputs, rather than a printed artefact with a digital copy.

How to redesign your menu this month

A practical order of operations, roughly an afternoon each:

1. Pull your sales data and list every item by units sold and contribution margin. You cannot design placement without knowing what deserves it.
2. Cut. Anything in the bottom of both lists comes off. Get every category under seven items.
3. Re-sequence. Move your best-margin dish to slot one or two of its category, and put a premium anchor near the top of each section.
4. Rewrite the descriptions for the items you want to sell and the ones guests won’t recognise. Four to twelve words each.
5. Reset the prices as plain numerals, no dollar signs, no leader dots, no right-hand column.
6. Build the mobile version before the print version, then derive the print layout from it.
7. Test on a stranger. Hand it to someone who’s never seen it and ask them to find three things. Whatever takes them more than a few seconds is a layout problem.

Frequently asked questions

What is the best design for a restaurant menu?

One where each category holds five to seven items, the dish you most want to sell sits in the first two slots of its section, prices are plain numerals set beside the item rather than in a right-hand column, and the whole thing is legible at arm’s length in your own dining room’s light. Format (single page, folder, board, phone) follows from your service style rather than driving it.

What is the 30/30/30 rule for restaurants?

It’s a rough budgeting heuristic rather than a design rule: roughly 30% of revenue to food and beverage cost, 30% to labour, and 30% to overhead, leaving about 10% as profit. It matters to menu design indirectly, because it tells you which dishes can carry a highlight and which are quietly costing you the margin that box is supposed to protect.

Should I put my most expensive dish first?

Not necessarily first, but near the top of its category. A higher-priced item early in a section calibrates what guests read as expensive, which makes your mid-priced dishes, usually your best margin, look reasonable. What you shouldn’t do is sort each category from cheapest to most expensive, which turns the list into a price ladder.

How often should a restaurant redesign its menu?

Review prices and item mix quarterly, redesign the layout when the item list has changed enough that the structure no longer fits, which for most restaurants means every year or two. If you’re reprinting more than a few times a year, that’s an argument for holding the menu as data rather than artwork.

Do I need different menus for dine-in and delivery?

Yes, in almost every case. The dine-in menu is designed for a fixed space and a guest who can see the room; the delivery menu is designed for a phone and for food that has to survive transport. Shared items, different lists, different layouts.

Where the lift comes from

If you only change three things, change these: cap your categories at seven items, move your best-margin dish into the top two slots of its section, and set prices as plain numerals without a right-hand column. None of them cost anything, and all three are supported by better evidence than most of what passes for menu design advice.

The rest is craft: typography, paper, photography, how the digital version behaves on a phone. Do all of it properly, but it compounds on top of the structural decisions rather than substituting for them. Start with the structure, then make it look good.