Framing Effect

Why this matters for your decision

In 1981, Tversky and Kahneman used the well-known Asian disease problem to show how strongly presentation works. With the same starting situation, 72 percent chose the safe option when lives saved were mentioned. When the same option was described as the number of deaths, only 22 percent chose it.

In a consumer context, the effect shows up everywhere: a tariff for €1 a day or €30 a month, 20 percent off or save €50, an insurance policy that protects or one that avoids costs. For decisions about price, claim and offer, two things follow. First, the result of a survey depends on how the offer is described in it. Second, it pays to test not only the offer but the wording that will later appear in the market.

Example

A home appliance manufacturer offers an extended warranty for a cordless kitchen machine. Variant A describes it as five years of security for €49, variant B as less than €1 a month. Both variants are identical in content.

In the behavioural test, each person sees only one of the two wordings on a realistic offer page. Variant B achieves a higher measured purchase intent for the warranty. The decision is then not only whether the warranty is offered, but also how it is described.

How it differs

The anchoring effect describes the influence of a number seen beforehand, the framing effect the influence of how the same information is presented. The two often occur together with prices. Acquiescence bias concerns the form of the question in a survey, not the presentation of the offer. A claims test is an application: it checks which wording of a promise changes the choice.

How Horizon deals with it

In a survey, framing is a source of interference, because the description in the questionnaire rarely matches later market communication. In the Painted Door Test, the wording itself becomes the variable. Horizon tests up to six variants that differ only in presentation and measures the purchase intent or sign-up intent of real people. This way, the wording that will later appear in the market is what gets tested. Nothing is sold.

Limitations

A framing result applies to the tested wording in the tested environment. A wording that works in an ad can work differently in a sales consultation or in retail. Framing tests also show which presentation is chosen, not whether customers perceive it as fair in the long term. And where wording is regulated by law, for example for price statements or environmental claims, the legal department decides first which variants are eligible at all.