Price tiers are found in tariffs, subscriptions, insurance policies and devices. The question is rarely just what each tier costs, but how the tiers influence each other.
Why this matters for your decision
Tiers make it possible to cover different levels of willingness to pay with one offer. At the same time, they affect each other: an expensive premium tier can make the middle tier look more attractive, while a basic tier that is too strong can pull customers away from the higher tiers. Which tier gets chosen therefore depends not only on the price and content of the tier itself, but on the whole set.
Behavioural research showed this early on: an additional, clearly inferior option can increase the likelihood that the option dominating it gets chosen (Huber, Payne & Puto 1982). Effects like these are the reason why tiers should be tested as a whole, not each on its own.
How to test price tiers
What gets tested are complete sets of tiers that differ in exactly one respect: for example the price of the middle tier, the scope of one tier, or whether there is a third tier. Each person sees exactly one set.
In Horizon's behavioural test, a realistic offer page shows the respective set; what is measured is which tier gets chosen and how high the overall measured purchase intent is. This shows how the distribution across the tiers and the average revenue per visitor differ between the sets.
Example
A streaming service has two tiers, Basic for €7.99 and Standard for €12.99. The test looks at the effect of an additional Premium tier for €17.99. In the set with three tiers, fewer people choose Basic and more choose Standard, while Premium itself remains small.
Overall measured purchase intent stays almost the same, and the average revenue per interested person rises. The Premium tier works mainly as a reference point for the middle tier.
How it differs
A simple price test compares prices for a single offer. Tiered pricing is about the architecture of several offers and their interactions. A test of the payment model, for example one-off purchase versus subscription, changes the structure of the price, not the tiers.
A conjoint analysis can help to distribute features sensibly across tiers. Which finished set produces the desired distribution is shown by the comparison of behaviour.
Limitations
If you change several levers at the same time, the result can no longer be attributed to one cause. In each test, the sets should therefore differ in only one quantity; further questions follow in a test sequence.
Small tiers such as a premium variant chosen by few people need large samples to estimate their share precisely. And the test shows the choice at first contact, not later switching between tiers among existing customers.
Evidence
Huber, Payne & Puto 1982: An additional, clearly inferior option can increase the likelihood that the option dominating it gets chosen (decoy effect). Adding Asymmetrically Dominated Alternatives: Violations of Regularity and the Similarity Hypothesis, Journal of Consumer Research 9(1). Source
Frequently asked questions
How many tiers are right?
That depends on the category and the target group. Whether a third tier improves the distribution or only causes confusion can be measured by comparing two sets.
Can you test the price of each tier individually?
Yes, but always within the set. The price of one tier changes the choice of the others, so the whole set is shown and only one price is changed.
What is a decoy?
An option that mainly serves to make another option look more attractive. Whether such an effect occurs with your offer is shown by the comparison of behaviour.
Which tier do your customers choose, and why?
Bring your tariff or tier structure, and we will outline a possible test design.
You will speak with Daniel Putsche
Founder & CEO, 30 minutes
Read more