Anchoring Effect

Why this matters for your decision

For most products, people have no fixed internal price. They judge relatively: against the price they saw last, the first number in a conversation, the comparison product next to it. Ariely, Loewenstein and Prelec had participants first write down the last two digits of their social security number and then place bids for everyday products. Those with a number above the median bid 57 to 107 percent more.

This is central to pricing decisions. Many pricing surveys show several price points in succession or ask about price thresholds. The first number then acts as an anchor for all subsequent answers. A higher starting price can raise measured willingness to pay, a lower one can reduce it. The result then also describes the order of the questions, not just the target group.

Example

A telecommunications provider wants to set the price for a new mobile tariff with data abroad. In a survey, one group starts at €39.99 and asks downwards, another starts at €19.99 and asks upwards. The accepted prices of the first group are on average four euros higher.

In the behavioural test, people see the tariff on a realistic offer page with exactly one price: €24.99, €29.99 or €34.99, randomly assigned. Measured sign-up intent per price level shows where demand drops off, without a previously seen test price acting as an anchor.

Distinction

The framing effect describes how the wording of the same information changes the choice. The anchoring effect describes the effect of a number seen beforehand. A reference price, such as a struck-through price, is a deliberately set anchor and therefore part of price design. A price threshold is a point at which demand drops markedly; it can be shifted by anchors.

How Horizon deals with it

In a Painted Door Test, each person sees only one price. The price levels are distributed across separate groups, so that no test price influences another. The anchors that exist in the market anyway remain in place: competitor prices, familiar price levels, a reference price on the page. These are exactly what belongs to the decision that will later be made in the market. What is measured is purchase intent or sign-up intent per price level; nothing is sold.

Limits

Anchors can never be avoided entirely, not even in a behavioural test. Anyone who sees the test has experience with similar products and prices. The result therefore applies to the market environment at the time of the test. A behavioural test with a few price levels also only shows demand at these points, not a complete price-demand function. Survey methods such as Van Westendorp remain useful for narrowing down the search space for the price levels.