Number of food businesses with poor hygiene ratings in France: 14,700 in one year

How many poorly rated businesses, depending on your question?#

The answer depends on the question: 14,700 businesses received poor ratings in one year, but only 4,450 were still listed that way on November 9, 2026. Each figure in the table has its own denominator.

For market sizing, use 14,700: the number of businesses that needed to bring their operations into compliance at least once during the year. For a prospect list to call this week, use 4,450: those are the ratings still online. A business counts once, even if it has several activities rated separately, such as the departments of a supermarket.

Within the restaurant market, 7,600 restaurants received poor ratings in one year. France has 186,300 active restaurants: that is 4.1% of them. Of the 36,040 restaurants inspected during the year (19.3% of the country's restaurants), 21 per 100 received poor ratings. The two lists are built differently (the inspection file on one side, observed active restaurants on the other): of the 32,996 restaurants with a SIRET number (the French ID of a business location) in the official file of November 9, 2026, 19,847, or 6 in 10, can be matched directly to Scalon's count by that number. The 4.1% and 19.3% are therefore rough estimates. The full count is in the study on the number of restaurants in France, of which this study covers a submarket.

Why other sources give different numbers#

Other sources give different numbers because they count different things: ratings online, inspections, administrative measures or businesses to inspect. These figures are accurate; they do not measure the same thing.

The official dashboard and our file from the same day are close: the November 9 file contains 4,628 poor ratings, corresponding to the 4,450 businesses. The dashboard reports 4,610 ratings: 18 fewer, with no update date shown.

The gap with the DGAL is larger: 17,285 measures against our 14,700 businesses. The DGAL counts measures during 2025, published or not. We count businesses with a published rating, over twelve months ending in September 2026. It is the same order of magnitude, not the same measure.

Scalon's 186,300 restaurants and the national plan's 270,000 commercial restaurants do not use the same definition. The plan counts commercial restaurants that inspection services must inspect. Scalon counts active restaurants open to the public, excluding bars. Neither figure corrects the other.

Hygiene inspection terms in plain English#

Inspection terms have precise meanings, and many readers confuse them. Here are the terms used in this article.

An "à améliorer" rating is therefore not a final penalty. It starts a correction period, and that deadline creates a need: training, drafting the plan de maîtrise sanitaire, treating the premises.

Which types of business receive poor ratings?#

Restaurants make up a little over half of the businesses poorly rated in one year (7,600 out of 14,700), but bakeries receive poor ratings most often: 30 per 100 inspected.

The table totals 14,705; everywhere else, we round it to 14,700.

Each type of business forms a submarket, with its own suppliers and needs.

• Restaurants: about 7,600. 21 per 100 inspected during the year. The restaurant hygiene inspection guide is for operators.
• Bakeries and pastry shops: about 2,750. 30 per 100 inspected during the year, the highest rate. See the bakery hygiene inspection guide.
• Cafeterias and institutional kitchens: about 1,230. 17 per 100. The decision-maker is often a public body or nonprofit (see below).
• Butcher shops and charcuteries: about 970. 25 per 100.
• Supermarkets, grocery stores and departments: about 880. 17 per 100.
• Approved slaughterhouses, food-processing facilities and warehouses: about 710. 11 per 100, the lowest rate.
• Caterers, fish shops, cheese shops, chocolatiers: about 370. 14 per 100.
• Farm producers: about 200. 15 per 100.

Independents, chains, franchises: who receives poor ratings?#

The segment that matters to compliance service providers is independent businesses outside chains, franchises, shared storefront names and retailer groups, which make their own purchasing decisions. About 12,600 of the businesses poorly rated in one year fall into this segment (between 11,700 and 12,900), or about 85%. Chains, franchises, retailer groups and corporate groups account for about 1,160, and public bodies and nonprofits for 980.

Our list of 190 networks initially classifies 13,150 businesses as independent (89.4%). To check this, we reviewed 100 businesses drawn at random (see the method): 4 of the 90 classified as independent actually belong to a chain or group missing from our list. The table accounts for this: about 584 businesses move to an "unidentified" row. The actual share of independents is about 85%, between 79.7% and 87.9%.

A secondary detail: among poorly rated independents, 20% are operated by a company with several locations (790 out of 3,955 in the November 9 file). This is a detail about the decision-maker, not the market definition.

Do chains receive poor ratings less often? Yes, for the same type of business. Check the denominator: this table covers the businesses in the official November 9, 2026 file and their ratings online that day, rather than the year. That is why the rates are lower than the "per 100 inspected during the year" figures in the other tables.

Among restaurants in the November 9 file, 2,249 independents out of 30,918 are listed with poor ratings, compared with 35 chain restaurants out of 1,986. For ratings online, the ratio is 4.1. For new poor ratings that appeared between October 14 and November 9, it is 3.2: 2.11 independent restaurants per 100 compared with 0.65 chain restaurants per 100. Independent restaurants therefore receive poor ratings 3 to 4 times as often as chain restaurants. Franchised restaurants are at 2.1 per 100.

This comparison has a limitation that we measured. Chains leave the list faster: between October 14 and November 9, 33 per 100 poorly rated chain businesses left it, compared with 20 per 100 independents. The gap in ratings online is therefore somewhat inflated; that is why we also give the ratio for new poor ratings. The share shows where potential customers are, not who is cleaner.

Cafeterias have their own decision-makers. In the November 9 file, 6,765 cafeterias and institutional kitchens have been inspected and 394 are poorly rated. Half of the cafeterias inspected are public (3,385): 169 are poorly rated, or 5.0 per 100. Nonprofits account for 1,718, including 115 poorly rated (6.7 per 100). Cafeterias run by a private company outside the networks on our list account for 1,393, including 100 poorly rated (7.2 per 100): 26 of the 99 whose company is known belong to a company with 10 or more active locations (company cafeterias, kitchens of large groups), making the term "independent" unreliable for cafeterias, as it is for approved food-processing facilities. Institutional foodservice groups run 236, including 9 poorly rated (3.8 per 100). The remaining 33 inspected cafeterias, including 1 with a poor rating, belong to branded chains.

Which networks, and who makes purchasing decisions?#

In the November 9 file, 167 networks have at least one inspected business, and 54 of them have at least one poorly rated business. The network type tells you who makes purchasing decisions.

The ten largest network operators, ranked by businesses inspected, are listed below. We do not publish the rate of poor ratings for individual chains.

Overall, chain businesses (franchises, company-owned locations, retailer groups, unverified type) make up 7.5% of businesses inspected and 3.3% of poorly rated businesses: 147 out of 4,897, or 3.0 per 100.

For a supplier, this changes the sales process. A company-owned chain's headquarters can sign for its entire network. Selling to a retailer group or franchise generally means selling location by location, unless an agreement is reached with the network operator: check network by network. For public cafeterias, the contact is the municipality or public body.

How large are the accounts? Company revenue and profiles#

Half of poorly rated companies that publish their financial statements generate less than €961,000 in annual revenue. This is a narrow base: 500 companies out of 4,080.

We linked the poorly rated businesses of November 9 to their companies (counting each company once) and read financial statements published since 2023. Only 12.3% of poorly rated companies have published revenue since 2023. Published financial statements mainly cover larger organizations: revenue is published for 24.6% of locations with employees, compared with 7.8% of the others.

Half of inspected companies that publish their financial statements generate less than €1,451,000. Among poorly rated independents (445 companies), half generate less than €746,000. Note that this group also contains 6 large companies with more than one billion euros in revenue (€43 billion among the six), whose company cafeterias are classified as outside networks; without them, half of the remaining 439 generate less than €731,000, but their combined revenue is not meaningful for an independent business. Among poorly rated restaurants (208 companies), half generate less than €423,000, compared with €696,000 for inspected restaurants (4,177 companies).

Combined revenue is not useful for sizing an account: poorly rated companies that publish their financial statements total €107 billion, but the largest 10% account for 98.3% of it. These are large groups, often owning many locations. For training or audit services, the useful account size is the revenue threshold below which half of companies fall: less than €423,000 for restaurants, less than €961,000 across all business types.

Company profiles add to the picture of account size. Poorly rated businesses are slightly newer and slightly more often run as sole proprietorships (private businesses, excluding public bodies and nonprofits).

Specifically: 18.0% of poorly rated businesses are less than 2 years old, compared with 15.7% of those inspected; 13.7% are run as sole proprietorships, compared with 12.1%.

The gap is wider for restaurants: 23.2% of poorly rated restaurants are less than 2 years old, compared with 20.0% of inspected restaurants.

Two economic signals add to this. Among poorly rated companies matched in our business-event database (1,877), 2.5% have undergone insolvency proceedings, compared with 1.5% of inspected companies (25,939). And 11.0% have had a business asset acquisition since 2024, compared with 12.9% of those inspected. Insolvency proceedings do not explain why the rating is poor. They indicate a financially vulnerable customer: approach with a clear offer and a firm price.

Why did the inspection take place?#

Inspections triggered by a signal (complaint, foodborne illness, targeted inspection) result in nearly 1 poor rating in 5, compared with 7 per 100 for routine inspections. One in 5 inspections prompted by a complaint results in a poor rating. Ratings online on November 9, 2026 mostly come from routine inspections.

Inspections triggered by a complaint, foodborne illness or a targeted inspection together result in 19.2 poor ratings per 100 (593 out of 3,084), compared with 7.1 for routine inspections. They account for 12.8% of poor ratings.

For a service provider, this is a useful filter. A business rated poorly after a complaint or foodborne illness has an urgent need and strong motivation. A business rated poorly at a routine inspection also has a deadline to meet, but the need is less pressing.

Where are poorly rated businesses located?#

The départements (French county-level areas) of Paris, Nord and Seine-Saint-Denis have the most poorly rated businesses in one year: about 930, 690 and 550. These numbers cover all food businesses, not only restaurants.

The last column gives the number of businesses listed with poor ratings on the official site on November 9, 2026. The full table, with your département, is just below.

Taking the type of business into account, 10 départements have more poorly rated businesses during the year than the benchmark, and 14 have fewer, out of the 92 we were able to compare. The benchmark is the number of businesses that would be poorly rated if each type of business in the département had the same rate as nationwide. It is not a permissible quota of poor ratings. A département above the benchmark does not necessarily have less clean businesses: inspections may be organized differently there.

Here is an example above the benchmark. Of the 266 businesses inspected in Guadeloupe, about 120 were poorly rated during the year, or 45 per 100, when the benchmark is 20. And an example below: of the 1,276 businesses inspected in Loire-Atlantique, about 130, or 11 per 100, when the benchmark is 21.

For the other 68 départements compared, our study cannot conclude that there are more or fewer poorly rated businesses than the benchmark.

Regions provide another scale. This view uses ratings online on November 9, 2026, per 100 businesses in the file (rather than the annual count used for départements). Among the regions of metropolitan France (European France, including Corsica), the share of poorly rated businesses ranges from 4.8 per 100 in Nouvelle-Aquitaine to 11.1 in Normandie. In overseas territories, the share reaches 26.7 in Guadeloupe and 39.8 in Mayotte, based on small numbers of businesses.

Regions are ranked by number of poorly rated businesses, with the unspecified region last. The count shows where potential customers are; the share shows where inspections most often found a problem.

The market changes every month#

The number of businesses listed with poor ratings stays roughly stable, but the list changes constantly: 4,335 on September 25, 4,305 on October 14, 4,450 on November 9, 2026.

Between October 14 and November 9 (26 days), 1,186 businesses joined the list and 1,041 left it. Of those that left, 824 were reinspected without a poor rating and 217 were removed from the site. Another 3,264 remained, or 76 per 100.

That amounts to about 1,400 businesses newly listed with poor ratings per month (1,368 scaled to 30 days). This is consistent with the annual count: 14,700 over twelve months, or an average of 1,226 per month.

For prospecting, the implication is simple: work from a list of poorly rated businesses each week. A three-month-old file has lost entries, including businesses that have already corrected their deficiencies, and misses recent additions.

How do you find and understand a business's hygiene rating?#

Six steps are enough to find a business on Alim'Confiance.

The other food businesses have no rating online; among restaurants, more than 4 in 5 have none: 33,102 restaurants have a rating online on November 9, 2026, out of 186,300 active restaurants (a rough estimate, since the two lists are built differently). Did you see a problem on site? The restaurant hygiene inspection guide explains how to report it.

Poorly rated: what should the business do?#

A business rated "à améliorer" must correct its deficiencies within the deadline set by the authorities, then undergo a new inspection.

Each step leads to a purchase. Here is what the business needs and who supplies it.

And if nothing is corrected?#

If the corrections requested in the formal notice are not made, the fine can reach €1,500, or €3,000 for a repeat offense (a 5th-class contravention, Code rural, article R. 205-6). Failing to comply with a closure decision is punishable by 2 years in prison and a €30,000 fine (Code rural, article L. 237-2).

To challenge a decision, a recours gracieux (informal appeal) is a complaint sent to the authority that made it, within 2 months (3 months in overseas territories). The guide on what to do after an "à améliorer" rating gives the details. Our file does not publish the reasons for the ratings. For example, the préfecture de Seine-et-Marne (the local office of France's national government) cites pests, rodents and a loss of food traceability as reasons for closures. It also cites food kept past its use-by date, deteriorating premises and poor storage conditions.

How can you use these figures for market research?#

These figures serve four purposes: sizing the market, segmenting it, choosing a territory and tracking a changing market.

Size an addressable market#

An offering for all food businesses starts with 14,700 poorly rated businesses in one year. It breaks down as follows: about 12,600 independents to approach one by one; about 1,160 locations in chains, franchises, retailer groups and corporate groups, which can also be approached through their network operators; 980 public bodies and nonprofits, where the buyer is the municipality, public body or nonprofit.

Segment the market#

Break down business types by structure. The 7,600 poorly rated restaurants require a different sales approach from the 2,750 bakeries: among those in the November 9 file, independent restaurants are poorly rated at 7.3 per 100, independent bakeries at 9.6.

Choose a territory#

Counts (Paris 930, Nord 690) show the scale. To compare two départements, relate them to businesses inspected: 34 per 100 inspected during the year in Nord, 18 in Paris.

Track a changing market#

About 1,400 businesses join the list each month, and almost as many leave. A study needs a date and regular updates: the Observatory publishes its counts with their dates and methods.

How do you build a prospect list of poorly rated businesses?#

A good list starts with the official source, matches it to public listings and is updated every week. Five points determine its quality.

1. Start with the official file. The public Alim'Confiance file provides an address for 99.2% of businesses and a valid SIRET number for 99.7%. It is the most reliable starting point for identifying a business.
2. Match phone numbers and websites. The official file provides the address; phone numbers and websites come from other public listings and business directories. Among the 1,329 poorly rated restaurants matched to our public business listings, 44.3% have a website, compared with 47.4% of inspected restaurants. Limitation: we have not measured phone or email coverage, or websites for other business types; matching still needs to be done business by business.
3. Filter by structure. Independent, network operator or public body: the contact and sales pitch differ (see the networks).
4. Keep up with the pace. About 1,400 businesses join the list each month, almost as many leave: rebuild the list every week; a three-month-old file has lost a large share of its entries.
5. Comply with the GDPR. Use business contact addresses for outreach, explain where the contact information comes from and respect the right to object. A sole proprietor is still an individual: their listing requires extra care.

Scalon's market size calculator helps define this scope before placing an order.

Frequently asked questions#

How many food businesses have poor hygiene ratings in France?#

14,700 in one year, according to Scalon's estimate for inspections carried out from September 26, 2025 to September 24, 2026, or about 1 inspected business in 5. The official site showed only 4,450 on November 9, 2026.

How many restaurants receive poor ratings?#

About 7,600 in one year, out of 186,300 active restaurants in France (4.1%, a rough estimate: the two lists are built differently). Of 100 restaurants inspected during the year, 21 received poor ratings.

What share of poorly rated businesses are independent?#

About 85%, or about 12,600 businesses out of 14,700. Chains, franchises, retailer groups and corporate groups account for about 1,160, public bodies and nonprofits for 980.

How many closures?#

Only "à corriger de manière urgente" ratings come with a closure or a halt of activity: 322 times out of 333 among ratings online on September 25, 2026. The DGAL counts 1,685 closures in 2025, across all inspections.

How long does a rating stay online?#

A year from the inspection date. A follow-up visit replaces the previous rating.

Does a poor rating mean closure?#

No, not in general. An "à améliorer" rating is a formal notice: the business stays open and makes corrections.

Does the business have to display its rating?#

No. According to Service-public.fr, businesses do not have to display their hygiene rating on their storefronts. Businesses only have the option to display it.

Is hygiene training mandatory?#

Yes, in commercial foodservice: at least one person must have completed 14 hours of training, unless a manager can demonstrate 3 years of experience as a manager or operator in the food sector. Since February 1, 2026, only organizations authorized by the préfet de région may provide it.

Do you sell to food businesses?#

Method#

The ratings come from the official Alim'Confiance file, published by the DGAL, extracted on September 25, October 14 and November 9, 2026. The annual count reconstructs inspections from September 26, 2025 to September 24, 2026. The structural classification matches each business to the Sirene register and a dictionary of networks documented by public pages. To check this classification, Scalon randomly drew 100 of the 4,450 businesses listed with poor ratings on November 9, 2026 and reviewed them twice, independently. The two reviews agree on business type for 100 businesses out of 100 and on structure for 97 out of 100; the 3 disagreements were resolved. Our classification gets the business type wrong for 1 business out of 100 (at most 5 out of 100) and the structure for 5 out of 100 (at most 11 out of 100): 4 of the 90 classified as independent belong to a group or chain missing from our list, hence the corrected share of about 85%.

Key takeaways#

• 14,700 food businesses received poor hygiene ratings in one year, including 7,600 restaurants; 4,450 were still listed that way on November 9, 2026.
• About 85% are independents outside chains, franchises and retailer groups: about 12,600 owners to approach one by one.
• For the same type of business, independent restaurants receive poor ratings 3 to 4 times as often as chain restaurants (7.3 versus 1.8 per 100 restaurants in the November 9 file; 2.11 versus 0.65 for new poor ratings).
• Half of poorly rated companies that publish their financial statements generate less than €961,000: the appropriate sales prospect is a small organization.
• About 1,400 businesses join the list each month, almost as many leave: rebuild the list every week; a three-month-old file has lost a large share of its entries.

Cite this study#

Scalon Observatory, "Number of food businesses with poor hygiene ratings in France: 14,700 in one year," November 9, 2026: about 14,700 food businesses received at least one poor hygiene inspection rating between September 26, 2025 and September 24, 2026, or about 1 inspected business in 5; about 85% are independents outside chains; 4,450 were still listed with poor ratings on Alim'Confiance (direction générale de l'alimentation) on November 9, 2026. https://scalon.fr/en/observatory/food-businesses-haccp-compliance

Sources#

The official texts, studies and data used.