Shopper claim tests fielded in the USA, the UK and 12 more markets

Claims Testing for US Consumer Brands Before Launch

Your US launch has a long list of candidate claims and room on the pack for two, and a survey average will not tell you which one shoppers believe. Alchemic runs claims testing as AI-moderated interviews that show each claim on its own and ask why every rating landed where it did. We test concept claims, pack and label claims and ad claims, so you learn which claim is understood, which is believed and which makes people buy.

Book a Demo Contact Us

A US shopper in Columbus, Ohio takes part in an Alchemic AI-moderated claims test, with the moderator's question, his quote about which claim he believes and a key insight on screen

Our Claims Testing Services

Each service below uses the same AI moderator: it shows a claim, checks what the shopper thinks it means, then asks how believable and motivating it is and why, with hundreds of interviews running at once.

Claims on the Concept Card

Key claims surface one at a time on an interactive concept card, so shoppers read each claim before rating it, and the AI asks why 3, not 4. Alchemic's stated norm is 100 to 200 interviews per concept.

Explore Concept Testing

Pack and Label Claims

Find which front-of-pack claim shoppers see first and which never lands, from hair care to snacks. Alchemic's stated diagnostic base is 30 to 50 conversations per variant per segment, each theme cited to the shopper's voice note.

See Packaging Testing

Claims Inside Your Advertising

Test a claim in the creative that carries it. Camera-on interviews probe element by element, including the claim line and the call to action, with explicit recording consent and typical studies of 200 to 500 interviews per market.

Learn More

Purchase Intent Behind a Claim

A claim that lifts stated intent may not lift sales. Our sourced guide explains what a top-two-box score on a 5-point scale predicts, why stated intent overstates demand, and why it ranks claims better than it forecasts them.

Read the Intent Guide

Where Claims Fit in a Launch

Claims are best tested once the concept is settled and before the pack prints. Our launch guide puts 7 research stages in order, from need and opportunity to a tracking baseline, with the question each stage cannot answer.

See the Launch Sequence

Packaging Testing Platforms in the US

Comparing vendors for pack and label claims? Our US guide sets out shelf, monadic and sequential monadic methods, a common 50 to 200 shopper sample, and how to check sample quality before you trust an appeal score.

Compare Platforms

Alchemic claims test dashboard showing live US shopper interviews and emerging themes such as believable, too vague and would buy

Our 5-Step Product Claims Testing Process

Set the claim shortlist

Bring the claims you are choosing between, from a one-line benefit to a full front-of-pack hierarchy, and the US shoppers whose belief matters: category buyers, lapsed users or a competitor's loyal buyers. Our researchers brief the AI on your category vocabulary and prior research, so nobody writes an exhaustive script.

Build the stimulus and sample

Each claim becomes its own card, image or pack render, shown monadically or in rotated order across cells. Alchemic's stated norm is 100 to 200 interviews for one concept and 200 to 400 across cells. Recruitment is managed fieldwork or bring your own list, with live screening and quotas.

Check meaning, then probe the rating

Shoppers answer in the browser or by phone in 60+ languages, or on WhatsApp by text or voice note. The AI first asks what the claim means to them, then how believable and motivating it is, and follows every rating with why that number and not the next one.

Read belief and confusion live

Answers code into themes such as believability, clarity, relevance and confusion as each interview completes, and the dashboard is live before fieldwork closes. Automated flags catch speeding, straightlining and copy-pasted answers, with auto-reject and auto-approve rules your team can tune for every study.

Choose the claim on evidence

Your US brand team drills from any claim to the shoppers, quotes and voice clips behind its score, cuts by segment or cell, and exports to PDF or PPTX. The result is a ranked claim shortlist showing the share of shoppers who understood, believed and would buy on each claim.

Start Your Claims Test

Honest fit

Claims Testing, Substantiation or a Benchmark Score

Claims testing measures what shoppers make of a claim: whether they understand it, believe it and would act on it. Whether a claim is true and supportable is a different job, claims substantiation, settled through lab, clinical or instrumental testing and legal review, and a perception test never stands in for it. If you need claim scores against a normative benchmark database, the established survey vendors serve that well, and many teams run both. Alchemic's version is the perception read at scale: every claim shown on its own, every rating probed, every finding traced to the shopper and the quote. Pick by the output your decision needs.

Talk Through Your Claims

A benchmark believability score beside an Alchemic AI claims interview quote from a US shopper in Phoenix, showing the reason behind the rating

Testimonials

Discover what our clients have to say about our consumer research services

Alchemic is ridiculously fast, and getting both qual and quant insights at that speed is a game-changer.

Alok Mahajanex-CMO, Sleepwell · via thealchemic.com

Alchemic's interviews go far deeper than any survey we have ever run and I love being able to pull up user stories for every insight.

Mayank BanerjeeFounder of Even Health · via thealchemic.com

Why Alchemic

Why Choose Alchemic for Claims Testing Market Research?

Comprehension before belief

A claim nobody understood cannot be believed. The AI asks what a claim means in the shopper's own words before asking how credible it is, which separates a wording problem from a credibility problem.

American shoppers already interviewed

Alchemic has fielded studies with consumers in the United States and the United Kingdom, among 14 markets in all, so a US claims test is run by a team that already handles American fieldwork.

Each claim read on its own

Claims surface individually on the concept card and the AI tracks engagement before any question is asked, so a shopper who skimmed past a claim is never counted as having rated it.

Every score has a source

Drill from any claim's score to the shoppers behind it, their verbatim and the voice clip. Brand and design teams hear exactly what consumers took a claim to mean, quote by quote.

About Alchemic

An AI-native research company running end-to-end consumer research at scale, built by researchers.

Alchemic is a founder-led, AI-native research company focused on end-to-end consumer research at scale, with studies fielded in the United States, the United Kingdom and 12 other markets, 14 in all. Its human-briefed AI interviewer holds one-on-one conversations in the browser and by phone in 60+ languages including English, Spanish, French, Arabic and Hindi, and on WhatsApp by text or voice note. Alchemic has run 50,000+ interviews for brand and insights teams in consumer goods, fintech, healthcare, retail and technology, among them Unilever, Mars, Sleepwell and CaratLane. Claims work draws on its concept, packaging and ad testing: the same moderator shows a claim, checks comprehension and probes the rating, and every theme links back to the shopper and the quote. For a US brand team that means a claim chosen on stated reasons rather than a mean score. Studies are quoted per project, with no per-seat fees.

50,000+interviews run
60+languages, built with native speakers
14markets fielded, including the USA and UK

Claims Testing FAQ

What is claims testing?

Claims testing is market research that shows consumers the benefit statements a brand might make, such as "made with real fruit" or "dermatologist tested", and measures which ones are understood, believed and motivating. It ranks candidate claims before they go on a pack, an ad or a product page. It tests perception, which is separate from proving a claim is true.

How does claims testing work?

A shortlist of claims is shown to a target sample, each claim on its own or in rotated order, and every respondent rates comprehension, believability, relevance and motivation to buy. With Alchemic an AI moderator then asks why each rating landed where it did, so the output is a ranked list plus the reasons behind every rank.

Is claims testing the same as claims substantiation?

No. Claims substantiation establishes that a claim is true and supportable, through lab, clinical or instrumental testing and legal review. Claims testing measures how consumers read a claim: what they think it means, whether they believe it and whether it would make them buy. A brand usually needs both, and a perception test cannot stand in for the proof.

What does product claims testing measure?

Product claims testing usually measures four things per claim: comprehension, believability, relevance to the shopper and motivation to buy. Some studies add uniqueness against the claims already on shelf and fit with the brand. Alchemic asks each one as a rating and then probes the answer, so a low believability score arrives with the shopper's reason for doubting it.

Can claims testing cover cosmetic and hair product claims?

Yes, for how consumers read them. A cosmetic or hair product claims test shows shoppers lines such as "long-lasting frizz control" or "gentle enough for daily use" and asks what they expect, whether they believe it and whether it suits the product. Whether the product actually delivers is a separate question for clinical or instrumental testing.

How many respondents does claims testing need?

It depends on the design. Alchemic's stated norms are 100 to 200 interviews for a single-concept monadic test, 200 to 400 in total when claims are split across cells, and 50 to 80 per sub-group when segment cuts matter. Most studies test three to six concepts before respondent fatigue sets in.

How do you rank a long list of claims?

Show the list in rotated sets and ask which claim is most and least motivating in each set, a designed choice exercise such as MaxDiff, then take the leaders into monadic interviews to learn why they lead. Alchemic runs both the choice exercise and the probing interview, so the ranking and the reasons come from the same study partner.

How long does claims testing take, and what drives the cost?

Alchemic's stated timeline for a 200-interview, single-market study is the brief on day 1, the builder live within 48 hours and a theme-coded report within 5 days; multi-market studies take a little longer. Cost is driven by sample size, the number of claims and cells, and the recruitment source. Running on your own customer list costs less than panel recruitment.

Still deciding which claims to test?

Send us the claim list and the decision it feeds, and we will tell you which test design fits it and what Alchemic's version would produce.

Send the Claim List

Plan Your Claims Test With Alchemic

Send the claims you are weighing, the product they belong to and the US shoppers whose belief matters. We come back with a test design, a sample plan and a per-study quote, and you work directly with the founders rather than a sales desk.

hello@thealchemic.com

Tell us about your claims test

Tell us a bit about yourself and we'll be in touch

hello@thealchemic.com Book a Demo