Risky claims checker

• Paste the exact wording you are considering, up to the character limit.
• Run the check and read each highlighted phrase in context.
• For each one, decide whether to rewrite it, keep it with evidence, or cut it.
• Write down the evidence you hold for every claim you keep.
• Have a person with the right expertise review anything in a regulated area.

Take the line "Cures dry skin in one week, guaranteed." The screen would highlight the outcome and the guarantee. A safer rewrite describes what the product contains and what customers say, such as "A rich moisturiser many customers use daily", and keeps any measured result out unless you have a proper test to cite. The aim is not to dilute the message but to make every statement something you can stand behind.

Illustrative example, not a benchmark, customer result or prediction.

The risky claims checker reads the ad copy you paste and highlights phrases that commonly need evidence or careful wording. Examples are absolute promises, health or financial outcomes, comparisons with other products and statements about guaranteed results. Each highlight comes with a short reason, so you can decide whether to rewrite the line, add proof or leave it as it is.

It works on patterns of language. It does not know your product, your test results or the rules of the country where you advertise, so it cannot tell you that a claim is true or allowed. That judgement stays with you.

No highlights does not mean the copy is clear. A claim can be false or unlawful without using any phrase the screen looks for, and a flagged phrase can be perfectly fine when you hold the evidence. The result is a prompt for a conversation about substantiation, not a verdict.

This is not legal advice and it is not a check against any advertising platform policy. Platforms update their rules often, and review decisions are made by the platform. If your product is in a regulated area such as health, supplements, finance or children, take advice from someone qualified in that area before running ads.

Madly applies a default list of blocked-phrase patterns together with a smaller set of public editorial additions, each marked with its source. Your own workspace blocked phrases are never used on this public page. The check runs on our server because the pattern list lives there, so the text you paste is sent to be screened. Do not paste confidential or personal information.

Matches are shown inside your own text, at the exact position they were found, so you can read each one in context. Overlapping matches are merged so a long phrase is highlighted once.

Advertising rules in many places require that objective claims can be backed up before they are made. In the UK, the advertising codes expect advertisers to hold evidence for claims, and the same idea appears in rules elsewhere. What counts as enough evidence depends on the claim: a taste preference needs less than a statement about health.

A practical habit is to keep a short evidence note for each claim in a live ad: what the claim is, where the proof comes from, the date, and who approved it. If a customer, a platform or a regulator asks, you can answer quickly.

A good result is not an empty list. It is a short list of phrases you have each made a deliberate decision about. A long list is a sign the copy leans on promises, and it is worth rewriting around specifics you can show, such as ingredients, materials, dimensions or real customer words.

• The product category, since health, finance and children carry stricter rules.
• The country where the ad is shown.
• The platform and its current policy.
• Whether you hold documented evidence for each claim.
• Reading no flags as approval.
• Removing a flagged word but keeping the same promise in other words.
• Copying a competitor claim you cannot prove for your own product.
• Forgetting that images and video can make claims too.
• Not keeping a dated evidence note.