Compromise effect
An option gains share simply by sitting in the middle of the range offered.
When options line up along a dimension such as price or capability, the middle one draws more share than it would in a two-option set. The middle position reads as a defensible compromise, which matters when the chooser has to justify the decision to themselves or to someone else. Move the range and the middle moves with it.
How it shows up in software
Three-tier pricing is the whole story. The middle tier gets a 'most popular' badge, and the badge is doing two jobs: marking the compromise and adding social proof: People look at what others did to decide what is correct here.. The effect also appears in subscription lengths, seat bundles, and support levels, anywhere a product lays choices out on a single scale.
Using it well
- Design the middle tier to be the one you would honestly recommend to most customers, then let the position work for you.
- Scope the top tier around a real segment's needs so the range is genuine rather than a prop for the middle.
- Label tiers by the job they serve (solo, team, org) rather than by size words, so position is not the only signal available.
- If you badge a tier 'most popular', check that it is. Making that claim falsely is a straightforward misrepresentation.
Where it turns manipulative
- Inventing a top tier nobody buys purely to reposition the middle, which is the decoy move under a different name and carries the same replication problem.
- Placing a 'most popular' or 'recommended' label on the tier with the best margin rather than the best fit.
- Stripping essentials out of the entry tier so the middle becomes compulsory, then marketing the entry tier as a real option.
Where you have seen it
Notion
The pricing page presents Free, Plus, Business, and Enterprise, with a badge marking one of the paid tiers.
Adobe Creative Cloud
Single-app, all-apps, and bundle plans are laid out along one price scale on the same page.
Starbucks
Drink sizes are offered as three named steps, with the middle sitting between the smallest and largest.
What the research says
- Simonson, 1989Mixed evidence
Adding an extreme option to a choice set increased the share going to the option that thereby became intermediate, and the effect was stronger when participants expected to justify their choice.
- Simonson and Tversky, 1992Mixed evidence
Across several product categories, an option's share rose when it occupied the middle of the offered range, which the authors modelled as tradeoff contrast and extremeness aversion.
- Frederick, Lee and Baskin, 2014Mixed evidence
The same programme of replications that undermined the attraction effect found the compromise effect held up better, though still weaker with familiar real products than with abstract numeric options.
Sturdier than the decoy effect, weaker than framing or anchoring. It belongs to the same family of context effects, so a claim of a large, dependable compromise effect in your specific market needs your own test.
Grades are a judgement about the evidence, not about how useful the idea is. Plenty of contested effects are still worth knowing, as long as you do not cite them as settled.