Social Comparison
People judge their own ability and standing against others, especially when no objective measure exists.
Festinger proposed that people have a drive to evaluate their opinions and abilities, and use similar others as the yardstick when objective standards are missing. Upward comparison to better performers can motivate or demoralise; downward comparison usually comforts. Which way it lands depends on whether the gap looks closable.
How it shows up in software
Leaderboards, percentile rankings, follower counts, and neighbour comparisons all install a yardstick where none existed. Feeds do it passively, by presenting a stream of other people's best moments as the ambient standard. The design question is which comparison group the product picks, because the product chooses the yardstick even when it claims to be neutral.
Using it well
- Compare users to their own past by default, and offer peer comparison as a choice.
- Pick a comparison cohort: A group of users bucketed by when they joined, tracked over time rather than blended together. where the gap looks closable, such as people who started in the same month.
- Show the path as well as the gap, so an upward comparison comes with a next action.
- Let users hide counts and rankings, since the same feature motivates one person and corrodes another.
Where it turns manipulative
- Ranking users publicly on outcomes they cannot control, such as luck-weighted sales or algorithmic reach, produces distress without producing performance.
- Selecting comparison targets to create inadequacy, then selling the remedy, is manufacturing a problem to price it.
- Pushing appearance and lifestyle comparison to young users because it drives engagement causes documented harm, and internal awareness of that harm makes it a choice rather than a side effect.
Where you have seen it
Strava
Ranks segment efforts against other athletes and against the user's own previous attempts, letting either be the reference.
Instagram
Tested hiding public like counts and later offered it as a per-account setting, an acknowledgement that the visible number is a comparison device.
Apple Fitness
Builds the default loop around closing the user's own rings, with competitions against friends as an opt-in layer.
What the research says
- Festinger, 1954Mixed evidence
Set out social comparison theory, including the claim that comparison targets are chosen for similarity and that comparison pressure drives group uniformity.
A theoretical paper with supporting studies of its era. The framework endures; specific propositions have been revised repeatedly since.
- Gerber, Wheeler and Suls, 2018Mixed evidence
Meta-analysis of sixty years of research found comparison is common and mostly upward, but found weaker support than expected for several classic propositions including the similarity hypothesis.
The most careful available audit of the field, and it is deflationary about parts of the standard story.
- Lockwood and Kunda, 1997Mixed evidence
Exposure to an outstanding peer inspired participants when the peer's success seemed attainable and deflated them when it did not.
Lab studies with student samples. The attainability moderator is widely cited but rests on a modest evidence base.
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.