Visual menu engineering: using food photos to steer guests toward higher-margin dishes

Raising prices is the obvious way to earn more per table. The less obvious way is to change what people order. That's the idea behind menu engineering, and on a phone menu, photos are one of the strongest tools you have for it.

Menu engineering in five minutes

Menu engineering comes from a 1982 book by Michael Kasavana and Donald Smith of Michigan State University. It sorts every dish in a category by two measures:

• Popularity: how often it's ordered compared with the other dishes in the same category.
• Contribution margin: the menu price minus the ingredient cost. It's the dollars a dish leaves you, not a percentage.

That gives you four groups:

	High margin	Low margin
Popular	Stars: keep them visible	Plowhorses: sell well, earn little
Not popular	Puzzles: earn well, few people order them	Dogs: consider cutting them

A dish counts as popular if its share of orders is at least 70% of an equal share. With 10 mains, an equal share is 10%, so anything ordered 7% of the time or more is popular. A dish has a high margin if it's above the category's average margin, weighted by how many of each dish you sell.

You need two numbers per dish: how many you sold last month (your POS report has it) and what it costs to make. Our free food cost calculator works out the second.

Where photos come in

Once you know your Puzzles, the dishes that make good money but don't get ordered, the question is how to get more eyes on them. On a phone menu, a photo takes up more of the screen than anything else on it.

What does the evidence say?

• A peer-reviewed study by Hou, Yang and Sun in the International Journal of Hospitality Management (2017) found that for dishes with clear, descriptive names, adding a picture raised ratings, purchase intention and willingness to pay. For dishes with vague names the effect was mixed, and some diners rated them lower with a photo. It measured what diners said they would do, not real sales.
• Delivery apps report bigger numbers. DoorDash says menus with item photos saw "up to 44%" more monthly sales, and Grubhub says photos "can increase your food sales by up to 30%". Treat these as marketing: neither company publishes a full method, "up to" hides the typical result, and they measure delivery apps, not dine-in menus.

Photos usually make a clearly named dish more appealing and more likely to be picked. Give the dish a name and description guests understand, then add the photo. How much it helps depends on your menu, and nobody can promise you a percentage.

Which dishes to photograph

If photos pull attention, a photo on every item pulls attention to nothing in particular. Be selective:

1. Puzzles first. High margin, low orders. A good photo is the cheapest thing to try before you rewrite or cut the dish.
2. Stars second. They already sell. A photo keeps them selling and makes the menu look good.
3. Plowhorses: think twice. They sell without help. A photo can push even more orders toward a dish that earns little.
4. Dogs: no photo. If you're keeping them, don't promote them.

After a few weeks, compare. SpotMyMenu's insights show which items guests view most and how long they spend on them. If a Puzzle gets more views but not more orders on your POS report, the problem may be its price or description, not how visible it is.

Photos that work on a phone

Everyone wants a digital menu with high-res photos, but on a phone "high-res" means sharp at phone size and quick to load. A 6 MB photo straight from your camera looks no better on a 6-inch screen than a well-compressed one, and it takes far longer to load on a patio.

SpotMyMenu resizes every photo you upload to three widths (320, 640 and 1280 pixels), converts it to WebP, and sends each phone the size it needs. Photos load only as the guest scrolls to them. Your part:

• Shoot in daylight, near a window, not under kitchen lights.
• Fill the frame with the food. Guests see a small version first.
• Shoot the real plate, the way it leaves your kitchen. A photo that oversells leads to a disappointed guest.
• Keep one look across the menu: same surface, same angle.

Layout matters too

Photos are only part of a visual menu layout. A few things that help guests find what you want them to find:

• Specials at the top. SpotMyMenu shows specials above the first section, with their own photo and price.
• Labels people scan for. Popular, Chef's pick and New work like arrows. Use them sparingly or they stop meaning anything.
• Short sections, with a section bar that follows the guest as they scroll.
• Add-ons, which raise the check without raising prices: "Add grilled shrimp +$6" on a salad.

Menu myths you can skip

A lot of menu psychology advice rests on weak or discredited research:

• The "golden triangle" or "sweet spot". The idea is that eyes go first to the upper right of a menu. An eye-tracking study by Sybil Yang (2012) found diners read menus in order, like a book, and found no sweet spot.
• Descriptive names that "raise sales 27%". That figure comes from research by Brian Wansink, who resigned from Cornell in 2018 after a misconduct finding and has had many papers retracted. We don't rely on his work.
• Dropping the dollar sign. One Cornell field study (Yang, Kimes and Sessarego, 2009) found guests spent more when prices were written as plain numbers ("14") instead of "$14.00". It was one restaurant. Worth testing, not a law.

Start with one category

Pick your mains. Pull last month's sales, cost each dish in the food cost calculator, sort them into the four groups, and add photos to two Puzzles. Check again in a month.

Photos, labels, specials, add-ons and item-level insights are all in SpotMyMenu. Try it free for 7 days.