Open Food Facts

Every grocery package is an argument. Marketing writes the front of the box and regulators write the back, and almost nobody standing in the aisle has time to reconcile the two. Open Food Facts tries to settle that argument with data. It has been at it since 2013, longer than the slicker commercial scanners that now get the press.

The app is the official mobile client for the Open Food Facts database, a non-profit project built by volunteers. People often call it the “Wikipedia of food,” and the comparison holds in both directions. Like Wikipedia, the database is free, open and enormous. Ordinary people build it by photographing packaging and filling in labels. Also like Wikipedia, its quality depends on whether someone has already edited the page you happen to need.

Point the camera at a barcode and you get a product page built around three ratings:

• The Nutri-Score grades nutritional quality from A to E.
• The Green-Score, the renamed Eco-Score, estimates environmental impact, including carbon footprint and packaging.
• The NOVA group flags how heavily a food is processed.

Below those sit allergens, additives, the ingredient list and the full nutrition table.

Where the app sits in the market is the interesting part. Its closest rivals, mainly Yuka and Fig, sell paid tiers or rely on their own scoring systems. Open Food Facts charges nothing, shows no ads, works without an account and does not sell your data. It doesn’t win on polish. It wins on trust, and on that measure it has a strong case. Users like it, with some reservations. The iOS version holds a 4.4-star rating, though from a small number of ratings.

• Three separate scores: Nutri-Score, Green-Score and NOVA appear together at the top of every product page. Nutrition, environmental impact and processing are judged separately instead of being averaged into one number you can’t inspect.
• Personal preferences: Tell the app what you care about, such as specific allergens or palm oil, and it highlights what matters to you on every scan. You no longer have to search ingredient lists yourself.
• Side-by-side comparison: Put products next to each other and see which is healthier or more sustainable, which is the decision you’re actually making at the shelf.
• Crowdsourced contributions: If a product is missing or incomplete, you photograph the packaging, ingredient list and nutrition label. That information then helps every other user.
• Sister databases: Coverage goes beyond food to cosmetics and pet food through companion projects, so the same ingredient transparency applies to shampoo and kibble.
• Privacy by default: No ads, no data sales and anonymous use.

The Scan Loop

The core interaction is simple: scan a barcode, get a verdict. The scanning half is built to be quick, and users regularly praise it — the camera is designed to lock on to a barcode almost immediately.

The verdict half is slower. After reading the code, the app has to fetch the product record, and users say that step is noticeably slower than the scan, and that the app feels slower than the website. That’s backwards, since a native app should be the faster option. When you’re in a store with a cart in one hand and a phone in the other, a second or two of loading spinner decides whether you keep using a tool or give up on it.

Three Scores, Not One

The app’s most important design choice is one casual users may never notice: it refuses to reduce food to a single number.

Yuka, its best-known competitor, gives you a score out of 100. That’s satisfying, but it’s also a simplification, because nutrition, processing and environmental impact are different questions with different answers. Open Food Facts keeps them separate. A flavored yogurt can earn a decent Nutri-Score and still land in NOVA group 4, the ultra-processed tier. A bag of plain lentils may score well on all three. When the scores disagree, that’s useful: it teaches you to read food yourself instead of just following an app.

Users pick up on this. Their most common compliment is that the NOVA, Nutri-Score and Eco-Score labels make ultra-processed foods easy to spot. For shoppers who have followed the recent research and coverage of ultra-processed diets, the NOVA badge alone is reason enough to install the app.

There’s a limit, though. These scores are summaries, not diagnoses. The Nutri-Score rates a product by formula, not against your overall diet. NOVA describes how a food was made, not whether it’s good for you in context. The app presents both clearly, but nothing stops you from reading a green “A” as permission to eat the whole bag.

Personalization and the Allergen Question

Preferences are where the app goes from reference book to assistant. Flag gluten, nuts or palm oil, and the product page puts what matters to you first. It’s a well-chosen feature that removes the most tedious part of reading labels.

It’s also where the stakes are highest. Allergen data is only as accurate as the last person who entered it, and manufacturers change recipes without changing barcodes. If you’re avoiding palm oil, an occasional error is an annoyance. If you have a severe allergy, treat the app as a first filter, never the final word, and read the actual package. A crowdsourced database makes that rule non-negotiable.

The Contribution Tax

This is the deal at the center of the app. When you scan something the database doesn’t know, or only partly knows, the app asks you to fill in the gaps by photographing the packaging, ingredient list and nutrition label. Your photos become the record for the next person.

The idea is admirable. The experience is not. The most common complaint is exactly this: products are missing and you have to add them by hand. Coverage is broad, and many users say the database is bigger than they expected. But it is deepest wherever volunteers are most active. The project started in France, and European shelves are mapped much more thoroughly than the long tail of regional American brands. A first-time user in a small-town US grocery store may hit enough blanks to wonder what the fuss is about.

Contributing also takes effort: several photos, framed properly, often under poor store lighting, for the benefit of strangers. Nobody pays to fill the gaps, so they get filled only as fast as volunteers are willing to do it.

Accounts and Rough Edges

You can use the app without an account, which is the right default. Contributing under your own name means creating one, and users have reported glitches there, including problems with special characters in usernames. It’s a small bug that hits at the worst moment: right when a motivated user has decided to give the project their time.

Beyond the Pantry

The sister databases for cosmetics and pet food apply the same approach to bathroom shelves and pet bowls. Expect thinner coverage than for food, since these are smaller projects, but ingredient transparency works the same way.

Open Food Facts looks like what it is: an app for browsing a very large database. The design relies on color-coded badges: the green-to-red Nutri-Score letter, the NOVA number and the Green-Score grade. They work well. You can understand a product at a glance, and the icons match the Nutri-Score labels already printed on packaging across much of Europe.

Further down the page, the layout gets dense. Ingredient lists, additive breakdowns and nutrition tables are thorough but visually flat, more reference page than consumer app. Nothing about it is ugly, and very little about it is enjoyable.

Performance is the weaker side, at least by users’ accounts. The wait between a successful scan and a fully loaded product page is the app’s biggest technical problem. A native app shouldn’t be slower than the website. For a tool meant to be used standing in an aisle, speed matters, and this design doesn’t prioritize it enough.

The current app is built from one cross-platform codebase, so iOS and Android offer almost exactly the same features: the same scanner, score card, preferences and contribution flow.

The user bases are very different in size. Most users are on Android, while the iOS rating of 4.4 comes from just 142 ratings — a decent score from a small sample, so it doesn’t mean the iPhone version is actually better.

On both platforms, the camera matters most. Barcode scanning is designed to work reliably on modern phones. Contributing is harder: photographing a wrinkled ingredient panel under fluorescent light rewards a good camera and steady hands, and older or budget devices will struggle more. Both versions are built for one-handed use in portrait mode and don’t try to be more than a pocket scanner, which suits the job.

Open Food Facts isn’t the slickest food scanner you can install, and it doesn’t try to be. It’s the most principled one. Competitors make money from your shopping habits or put features behind paid tiers. A free, ad-free, non-profit app that shows its reasoning, with three separate scores, open data and anonymous use, is uncommon in this category.

The downsides are real. The app is slower than it should be, the catalog has gaps you’re expected to fill yourself, and the interface feels like a database viewer. These are the predictable weaknesses of a volunteer project, and they’ll frustrate anyone expecting a polished consumer product.

The core promise holds up, though. Scan a barcode and you get an honest picture of how nutritious, how processed and how sustainable a product is. All it costs is some patience, and no competitor matches its trustworthiness. We recommend it to shoppers who value transparency over polish — and when you scan something that’s missing, take the three photos. The next shopper will benefit.