Skip to content

HOME INSIGHTS Food

Only 39% of Menu Items Have a Photo: The Cheapest Conversion Fix NYC Restaurants Keep Skipping

QUICK ANSWER

New 2026 data found only 39 percent of menu items carry a photo, and that medium detail menus outperform exhaustive ones. Optimized profiles earn four to five times the menu clicks of basic ones. The fix is not more information; it is better structured information.

Guest browsing a restaurant menu with dish photos on a phone in New York City

Most restaurants approach their online menu the way they approach a filing cabinet: get everything in there, correctly labeled, and the job is done. The 2026 data says that instinct is actively costing you clicks. Marqii's industry report, published September 10 from location level engagement data and an operator survey, found something counterintuitive. Menu completeness does not improve engagement in a straight line. Locations with medium menu completeness averaged 205 daily search views and six menu clicks. Locations with high completeness averaged 154 views and three clicks. Low completeness performed worst at 59 views and two clicks.

What the numbers say

Menu completeness

Daily search views

Menu clicks

Low

59

2

Medium

205

6

High

154

3

The middle wins. That is unusual enough to be worth explaining rather than just repeating.

Why would a more complete menu perform worse?

Because a guest scanning a phone is making a fast decision, and every additional field competes for the same attention. A menu that lists eighty items with full attribute detail on each one is harder to act on than one that shows twenty five items with a clear description and a photo. The goal was never completeness. It was helping someone choose. Photos are where the gap is widest. Marqii found only 39 percent of menu items across the platform carry an image, and 19 percent of operators describe their Google photos as minimal or unmanaged. That is a majority of dishes asking a guest to imagine what they are ordering.

What does profile completeness actually earn?

This is the part that pays for the effort. The same report scored locations into four tiers from Starter to Optimized and found engagement rising at every step, with Optimized profiles averaging four to five times the menu clicks, direction requests, and website clicks of Starter profiles. Most of that gap can be closed with information you already have, which is the same compounding effect we described in our Google Maps guide. Four to five times is not a marginal improvement, and there is no media spend attached to it. The work is a few hours of organizing what already exists in your kitchen and on your printed menu.

What should you actually fix first?

In this order, because the order affects the return.

  • Photograph your top ten sellers properly. Not the whole menu. The dishes that already sell are the ones people are deciding between.
  • Write real descriptions for those ten. One clear line describing what arrives, not a paragraph of provenance.
  • Trim the online menu to what you actually serve consistently. Every item you removed last spring that is still listed is a guest disappointed on arrival.
  • Upload the menu as text rather than a PDF, since a PDF cannot be read the way structured text can.
  • Add current photos of the room as well as the food. People are deciding where to sit, not only what to eat.
  • Then, and only then, worry about completeness across the rest of the list.

Do the photos need to be professional?

They need to be honest and well lit. A clear phone photo of the actual plate as it leaves your kitchen outperforms both a stock image and a three year old shot of a dish you have since changed. The failure mode that costs you is not amateur photography; it is a photo that does not look like what arrives at the table. Consistency matters more than polish, and it is worth being deliberate about which dish carries the weight. Give people one thing to remember rather than a tour of everything, which is the same principle we applied to creator partnerships in our piece on micro-influencers versus food creators.

How does this connect to what you charge?

Directly. A guest deciding from a bare list of names and prices has nothing to judge value against except the number. A guest looking at a clear photo and a real description is evaluating what they get for that number, which is a completely different calculation. In a year when 82 percent of diners say prices rose and only 28 percent consider them fair, presentation is doing price defense work. That is the practical version of the argument in our piece on value against discounting: make what the guest receives legible, and the price stops being the only thing on screen.

Who does this work, and when?

Somebody named, within the next two weeks, before the October event weekends bring people who have never heard of you. A convention attendee near Javits or a festival visitor near the Seaport is choosing from photos and descriptions with no local knowledge to fall back on. The realistic scope is one shift. Ten dishes shot during a quiet service, ten descriptions written at the end of it, the online menu pruned the same afternoon. The event calendar that makes this urgent is in our October revenue map, and the listing maintenance rhythm it belongs to is covered alongside hours in the same Marqii data.

What does a good menu description actually look like?

One line, concrete, in the order a guest cares about. What it is, what it comes with, and one detail that makes it specific. Nothing about philosophy, nothing about the farm unless the farm is the reason it tastes different. A description that reads well on a phone tells someone what will arrive on the plate. Compare two versions of the same dish. The first says roast chicken, seasonal vegetables, pan sauce. The second says half a roast chicken with charred leeks and a black pepper pan sauce, enough for two to share. The second one answers a decision. The first one describes a category. Length is the trap. Anything past about fifteen words stops being read and starts being skimmed, which means your most persuasive detail gets skipped if it is buried at the end. Put the specific thing early.

Structure beats volume

The instinct to add more information is understandable and wrong. The restaurants earning the most engagement from their listings are not the ones with the most complete data; they are the ones whose essentials are structured well: real descriptions, current photos, accurate items, nothing listed that you no longer serve. My Chef Social builds and maintains the menu, photo, and listing structure that turns local search interest into bookings. Get in touch if you want your profile audited before the month's event traffic arrives.

FREQUENTLY ASKED QUESTIONS

Marqii's 2026 industry report found only 39 percent of menu items across its platform carry a photo, and 19 percent of operators describe their Google photos as minimal or unmanaged.

Put these fixes to work in your venue. Start with the free audit.

GET A FREE AUDIT