On or Off Track

In one paragraph

Jackie Silverman and Alixandra Barasch are marketing researchers. They define a streak as the same behaviour three or more times in a row, and ask how showing it in an app’s log changes what people do next. They use data from 980 university employees in a 30-day step challenge, and six experiments in which people did desk exercises, answered language questions or played word and number games. In most experiments everyone did the same things; only what the log showed was different. Their answer: people adopt keeping the streak as a goal in itself. An intact streak keeps them going. A broken one, shown on screen, makes them likelier to stop. The recommendations are written for companies that want more engagement.

Three takeaways

1. 1

   The display, not the miss, drives the drop.

   In a follow-up experiment, after the same single missed item, 37.61% carried on when a streak had come before the miss, against 64.22% when there had been no streak. The loss attaches to the broken record, not to the missed day.

2. 2

   Self-blame makes it worse.

   In a game study, 53.17% carried on with an intact streak, 42.00% when the app caused the break, and 28.87% when people broke it themselves by failing a hard game. The log looked the same in both broken conditions; only who was to blame differed.

3. 3

   Repair helps, but not fully.

   Offering to repair a broken streak lifted the share who carried on from 68.66% to 85.20%, still short of 93.14% with an intact streak. People also pay to protect streaks: 45% said they would watch an advert to keep one, and 43% to repair one.

What it means for design

Proposals for our patterns, drawn from the book. They are not yet part of the published pattern pages.

Progress and streaks

The spectrum already includes “No progress view” and “Today’s step”. This research favours views that never display a break over repairing one afterwards. If you keep a count, the authors’ own suggestion helps: count broadly and over longer units, such as weeks, so an irregular rhythm doesn’t read as failure.

Notifications and reminders

The authors themselves suggest not alerting people when a streak breaks. Never send a miss notice. After a gap, a reminder should carry the next reading, not the count.

Reading plans

Self-blame deepened the drop. Avoid copy that puts the gap on the person (“You broke your streak”, “You’re behind”). After a gap, show the next day, not the missed ones; the pattern’s guidance on returning after a break is under “Go further”.

Paid repair

Willingness to watch an advert or buy a “streak freeze” shows that a streak is also inventory. Charging for relief from a loss the design created is a line the Progress and streaks pattern already lists under what to avoid.

Make room for rest, and say when it is enough

A habit someone chose is worth supporting. A count that turns a day of rest into a loss works against the rest the principle asks you to make room for.

Where we differ

The paper treats carrying on with the app activity as the good outcome, and calls highlighting streaks a win-win for company and customer. It never asks whether the activity was worth continuing. Someone who stops a game after a broken streak may simply have got their evening back.

In my paper Digital Spiritual Formation in Post-Christian Europe I put it this way: a user who opens a devotional to preserve a 47-day streak is being managed through loss aversion; same frequency, different formation. This study shows the mechanism from the company side: the record itself becomes the goal.

The self-blame finding matters more in faith than in fitness. Many Christians already treat a missed quiet time as a personal failure. A broken streak on top of that reads as spiritual failure, which my paper also warns about.

Return is received, not earned back. God’s welcome doesn’t depend on an unbroken record, and the church has always known seasons of rest, dryness and return. A plan that simply shows the next day says that better than any repair shop.

Who should read it

• Designers and product managers working on habits, reading plans, prayer reminders or progress screens.
• Anyone deciding whether to offer streak freezes or repairs.
• Short on time? The abstract, Studies 6 and 7, and the general discussion (manuscript pages 38–52).

On or Off Track at Journal of Consumer Research (Oxford University Press). No affiliate link.

Evidence and limits

• Consumer research, not faith practice. Exercise, language learning and games. The only link to Bible apps is that 6% of a 100-person pilot named a Bible-verse app among the streak apps they used.
• Engagement is the outcome. The studies measure whether people kept doing the app activity in the moment, not whether their fitness, learning or wellbeing improved.
• Short streaks, single sessions. Most streaks were three or four items, and length was tested only up to 20. No experiment followed people over weeks or months.
• The field data are correlational. The step challenge can’t separate the effect of the streak from the kind of people who keep streaks; the authors say so.
• Mostly paid online samples. Participants were largely paid workers on Amazon Mechanical Turk, plus lab and field data, collected 2017–2021.
• We read the accepted manuscript. Figures could differ slightly in the typeset journal version, which is behind a paywall.

Cited studies we have not checked yet

What the book reports, as it reports it. We don’t cite these as findings until we have read the original.