Can I say it?
Can I say “works for 98% of patients” in a clinic ad?
Usually no. 5 rules fired across 3 sample sentences, and the checker's own text explains why.
The short answer
Usually no. The checker blocks the bare phrase in this setting. The first rule it hits is CG-STA-001. Its reason: A number next to an outcome word is an outcome statistic. It needs a source, the population, the time frame and a statement of what is typical.
Source the rule cites: FTC Act section 5; FTC Endorsement Guides, 16 CFR Part 255 (typical results); FTC Health Products Compliance Guidance
Checked as an ad promoting a clinical service. Context changes the result. The same samples hit these blocking rules: 4 in a drug or device ad, 4 in a supplement ad, 3 in a clinical service ad.
A clear result is not an all-clear. The checker flags wording for review and does not know your evidence, your state or your platform's review. Read the limitations.
Sample wording, run through the checker
These are sentences written to contain the phrase, not real claims. Each ran through the real engine and the shipped rule-set when this site was built.
Works for 98% of patients.
- BlockingCG-STA-001Outcome statistics. Matched: “98% of patients”
- BlockingCG-STA-008Outcome statistics. Matched: “Works for 98% of patients”
- BlockingCG-TRL-004Outcome statistics. Matched: “Works for 98% of patients”
- WarningCG-STA-005Outcome statistics. Matched: “98% of patients”
98% of our patients saw results in a month.
- BlockingCG-STA-001Outcome statistics. Matched: “98% of our patients”
- WarningCG-STA-005Outcome statistics. Matched: “98% of our patients”
Our shot works for 98% of patients, guaranteed.
- BlockingCG-STA-001Outcome statistics. Matched: “98% of patients”
- BlockingCG-STA-008Outcome statistics. Matched: “works for 98% of patients”
- BlockingCG-TRL-004Outcome statistics. Matched: “works for 98% of patients”
- WarningCG-STA-005Outcome statistics. Matched: “98% of patients”
- Warning
CG-ABS-001Absolute promises. Matched: “guaranteed”
What to say instead
This version came back with no rule fired:
Results vary from person to person. Book a consult to hear what to expect for your situation.
What the rules themselves suggest:
- CG-STA-001 Cite the study with population, time frame and sample size ("in a 2024 study of 212 adults, 62% ..."), or remove the number.
- CG-STA-008 Remove the figure, or cite the study and the group it covers, and keep the proof on file.
- CG-TRL-004 Cite a published study with sample size and method, or remove the number. Do not promise results are typical or permanent.
The rules behind this answer
Reason, source and citation for each rule that fired, in the checker's own words. Blocking rules come first.
CG-STA-001 · Blocks approval · Citation checked against stored text
A number next to an outcome word is an outcome statistic. It needs a source, the population, the time frame and a statement of what is typical.
Source: FTC Act section 5; FTC Endorsement Guides, 16 CFR Part 255 (typical results); FTC Health Products Compliance Guidance
Safer wording: Cite the study with population, time frame and sample size ("in a 2024 study of 212 adults, 62% ..."), or remove the number.
16 CFR Part 255 · Needs counsel review
CG-STA-008 · Blocks approval · Citation checked against stored text
A headline percentage with no study behind it, or one built from an informal poll, is an unsubstantiated statistic.
Source: FTC Act section 5; FTC Policy Statement Regarding Advertising Substantiation; FTC Health Products Compliance Guidance
Safer wording: Remove the figure, or cite the study and the group it covers, and keep the proof on file.
FTC Policy Statement Regarding Advertising Substantiation · Needs counsel review
CG-TRL-004 · Blocks approval · Citation checked against stored text
Survey tallies, average-result figures, lasting-result promises and percentage-of-patients claims with no stated study, sample or method are unsubstantiated outcome statistics.
Source: FTC Act section 5 and FTC Policy Statement on Substantiation (1984); FTC Health Products Compliance Guidance (2022); FTC Endorsement Guides, 16 CFR 255.2(c) (typicality of results)
Safer wording: Cite a published study with sample size and method, or remove the number. Do not promise results are typical or permanent.
FTC Policy Statement on Substantiation, 16 CFR 255.2(c) · Needs counsel review
CG-STA-005 · Blocks approval · Citation checked against stored text
A share or ratio of people who got a result is an outcome statistic. It needs a source, the population, the time frame and a statement of what is typical.
Source: FTC Endorsement Guides, 16 CFR Part 255; FTC Act section 5; FTC Health Products Compliance Guidance
Safer wording: Cite the study (population, size, time frame) or your tracked data, and say what results are typical.
16 CFR Part 255 · Needs counsel review
CG-ABS-001 · Warns · Citation checked against stored text
Outcome guarantees are unsubstantiated absolute claims.
Source: FTC Act section 5, 15 U.S.C. 45; FTC Health Products Compliance Guidance
Safer wording: Remove the guarantee. If you offer a refund policy, describe the policy ("30-day refund if you are not satisfied") instead of guaranteeing results.
15 U.S.C. 45 · Needs counsel review
Background: Typical results disclosure in the glossary.
Questions
Which rule covers percentage success claims?
Rule CG-STA-001. It cites 16 CFR Part 255. The rule's reason: A number next to an outcome word is an outcome statistic. It needs a source, the population, the time frame and a statement of what is typical.
What should I write instead?
The rule's own suggestion for rule CG-STA-001: Cite the study with population, time frame and sample size ("in a 2024 study of 212 adults, 62% ..."), or remove the number.
Does the answer change for a drug, a supplement or a service?
Context changes the result. The same samples hit these blocking rules: 4 in a drug or device ad, 4 in a supplement ad, 3 in a clinical service ad. Pick the type that matches what you promote when you run the checker.
Next step
Check your own wording
Paste your copy into the free checker. Every flag names its rule and its source.