


A claims test checks which product statement, i.e. which advertising or performance promise, has the strongest effect on a target group; in the behavioural test, what is measured is which statement triggers purchase intent, not just approval.
On packaging, offer pages and ads, claims often decide within seconds. Which wording convinces can be checked before it is used, in a survey and through behaviour.
A claim is rarely decided in isolation. Packaging print, campaigns, shelf communication and often product development depend on it. A wording that seems credible and relevant in a survey need not be the one people choose at the moment of the offer. In a survey, participants read every claim attentively and rate it individually; in the market, it stands next to price, image and alternatives.
Claims tests do not only concern innovation: existing products, tariffs and brands are also reworded, for example during a repositioning or when framework conditions change.
In a survey, claims are typically rated for credibility, relevance, uniqueness and willingness to buy, often also by comparing several wordings. This provides good diagnostics: which statement is understood, which seems exaggerated?
In the behavioural test, the offer pages differ only in the claim; product, price, image and structure remain identical. Real people each see one variant, and what is measured is the last binding click before the reveal. This shows which statement triggers purchase intent at the moment of the offer. Horizon runs claims tests as Painted Door Tests with up to six variants.
For a cordless kitchen machine, three main statements are under consideration: "Dinner in 20 minutes", "One appliance instead of five" and "Cooks while you do something else". In the survey, "One appliance instead of five" performs best on relevance.
In the behavioural test, 3.1% of visitors click "Add to basket" for "Cooks while you do something else", 2.4% for "One appliance instead of five" and 2.6% for "Dinner in 20 minutes". The survey explained which statement is understood; the behavioural test shows which one triggers purchase intent.
A value proposition test changes the entire value proposition, often with several elements of the page; a claims test changes only the statement itself. An ad test measures which ad gets more clicks, i.e. advertising effectiveness; the claims test measures the effect on the decision on the offer page. Special legal requirements apply to environmental claims.
The behavioural test shows which wording works, not whether it is permissible. Which statements may legally be used is clarified by the legal department before the test; only approved wordings are tested.
The test also measures the effect at the moment of the decision. It does not reflect long-term effects on brand image or trust. And if claims differ only minimally, the difference in behaviour may be too small to establish with a reasonable sample; in that case, a pre-selection in a survey is worthwhile.
Chandon, Morwitz & Reinartz 2005: Among surveyed customers, the relationship between intention and purchase is 58% stronger than it would be without the survey: the act of asking itself inflates the measured validity. Study on self-generated validity effects in survey research, Journal of Marketing 69(2). Source
Up to six variants run in parallel. Each additional variant increases the number of visitors needed, so a pre-selection is often worthwhile.
Yes. Only wordings approved by your legal department are tested, not least because the offer pages look realistic.
No. The survey shows how claims are understood. The behavioural test adds which statement triggers purchase intent.
Bring your decision question, and we will outline a possible test design.
You will speak with Daniel Putsche
Founder & CEO, 30 minutes
Read more