- Safety Gate is the EU's rapid alert system for dangerous non-food products — and cosmetics is its most-notified category, around 36% of all alerts.
- Reports drop weekly and do not reach you automatically — you either go look for them or automate the capture.
- When a competitor, a supplier or one of your portfolio ingredients shows up flagged, you want to know on day one. Not the week after, when a customer forwards you the news.
1. What is Safety Gate (and why is it not called RAPEX anymore)?
Until 2018, we all called it RAPEX. That year the European Commission rebranded it as Safety Gate, but the underlying system is identical: a cooperation network between national authorities that shares fast, public information about dangerous non-food products detected on the market.
31 countries participate: the 27 EU member states, the three EEA-EFTA countries (Iceland, Liechtenstein, Norway), plus the United Kingdom in respect of Northern Ireland. Whenever a national authority spots a product it considers unsafe for consumers, it notifies the system. That notification gets published and stays publicly accessible.
Safety Gate covers a wide range of product categories: toys, vehicles, electrical equipment, chemicals, clothing, jewellery — and yes, cosmetic products. And here is what most people get backwards: cosmetics is not a small slice of that volume. It is the most-notified category in the entire system — around 36% of all alerts, ahead of toys and motor vehicles, for three consecutive years.
2. How the alerts get published
Alerts are published as they come in and bundled into weekly reports, with the consolidated report landing on Fridays. Each report contains N alerts, and each alert documents:
- Product notified (name, brand, model where applicable)
- Category (Cosmetics, in our case)
- Notifying country and country of origin of the product
- Risk identified (banned substance present, over-concentration, contamination, false labelling, undeclared allergens)
- Chemical substance involved (if any) with its CAS number
- Action taken by the authorities (market withdrawal, import ban, voluntary manufacturer recall)
The real cadence is rock-steady: practically every week has a report, though the count of alerts per report varies. In cosmetics, the usual triggers are hydroquinone in unauthorised skin-lightening products, UV filters above the legal limit in sunscreens, banned preservatives in baby care lines, and microbiological contamination.
Access is public: there is the Safety Gate web portal, plus a machine-readable feed that lets you download the full reports. No credentials needed.
3. Why it matters (four concrete reasons)
Fair enough — interesting system. But why should your regulatory team be looking at Safety Gate every week? Four reasons worth your time.
3.1. Your competitor's recall = your next reformulation
When a competitor gets pulled for a specific substance, it is not just their problem. It is a signal of regulatory movement ahead. If that substance shows up repeatedly across notifications, the odds of it landing in Annex II (banned) or Annex III (restricted) of Regulation 1223/2009 go up materially. Reformulating BEFORE the regulatory change beats reformulating in an emergency recall later.
3.2. A flagged supplier = a supply-chain problem
If one of your raw-material suppliers shows up in an alert, audit your supply chain immediately. Your product probably is not the one flagged yet — but if you share a supplier with someone who got flagged, the next batch arriving at your door could be the contaminated one.
3.3. A recurring ingredient = a future Annex II candidate
When the same substance shows up across multiple notifications from multiple countries inside a short window, that pattern usually precedes an SCCS opinion and, eventually, an amending Commission Regulation. Tracking recurring ingredients gives you 12 to 18 months of lead time over competitors who only learn about the change when the regulation drops.
3.4. Reputation: finding out before the press does
If your own product gets flagged, the clock starts the moment it hits the system. The distance between proactive handling (your statement out the same day, recall plan running, customer support briefed) and reactive handling (you find out because a journalist calls) is the distance between a managed crisis and a viral one. Not an exaggeration.
4. How to monitor it without losing your mind
The Safety Gate portal is public and useful. The operational problem is everything that comes next: scanning manually every week is tedious and error-prone. And the volume is anything but small. Cosmetics is the single largest category in the system — around 36% of all alerts, the most-notified product type three years running. That is hundreds of cosmetic alerts a year to read, classify and cross-reference.
Stack on top the cross-referencing step: each alert mentions, say, a brand and a country. Is this a competitor that matters to you? Are any of the listed ingredients in any of your formulations? Is it one of your suppliers? Without a systematic process, this gets done by eyeball. And what gets done by eyeball gets missed, sooner or later.
The operational rule is straightforward: automate the capture and the cross-referencing, leave the regulatory team to analyse the alerts that actually touch your portfolio. That is the division of labour that scales.
5. How BD-API handles it
Here is how we automate this, without rehashing the technical detail already on our Regulatory Watch page:
- Weekly automated monitoring: BD-API polls the Safety Gate feed every week, with a retry tick every 5 minutes if a fetch fails.
- Automatic filtering: we only process
category === 'Cosmetics'. Everything else never reaches your team — and what does arrive comes structured, not as a wall of weekly bulletins. - Smart cross-referencing: every alert is compared against the CosIng INCI inventory tied to your portfolio. If a substance mentioned matches one of your ingredients, criticality is bumped automatically.
- AI enrichment: a source-specific analyser extracts substances, CAS numbers, affected Annexes, critical dates and proposed actions — all in structured JSON, ready for your pipeline.
- Signed delivery: email plus webhook with HMAC-SHA256 signature. Your system gets the alert ready to act on.
- Granular subscription: take Safety Gate on its own, or combine it with SCCS, DG SANTE and EUR-Lex. No service restarts required.
The philosophy is simple: BD-API delivers the structured data, your regulatory team decides what to do with it. We do not replace human judgement — we hand back the hours you currently spend on manual review.
6. Bottom line
Safety Gate is not a regulatory curiosity, it is a public observatory of your market that very few people are watching with the attention it deserves. Each weekly report is a small window into what is coming: what is about to be banned, who is in trouble, which supply chains are under pressure. And cosmetics is not a footnote in that observatory — it is the headline category.
The question is not whether it is worth watching. It is how to watch it without burning half a workday a week on manual work that automates in five minutes.
Premium plan. A human replies within 24 business hours. Already have a specific case — a supplier that appeared in an alert, a recurring substance worrying you? Tell us in the form and we will skip the small talk.