Short answer. The research behind the statistic found that keeping more customers raised profit sharply at the companies it studied. But "5% more retention, up to 95% more profit" is not a rule. The 95% is the top of a range from a 2000 article about online businesses. The 1990 research reported results for particular firms, such as 85% more profit in one bank's branch system. Below: what the sources say, four things that bring app users back, and how to test it on your own numbers.
Where the numbers come from
The statistic traces back to two Harvard Business Review articles co-written by Frederick Reichheld. Bain & Company carries both on its site.
In 1990, Reichheld and W. Earl Sasser Jr. published "Zero Defections". Cutting the customer defection rate by 5% produced 85% more profit in one bank's branch system. It produced 50% more in an insurance brokerage and 30% more in an auto-service chain. The authors wrote that the size of the effect varies by company and industry (the article's abstract).
The 25% to 95% range is from a later article, by Reichheld and Phil Schefter in 2000. It was about online businesses (Bain's page for that article).
So "up to 95%" describes particular businesses, not your app. What the research does support is the direction: in the companies studied, cutting defections raised profit sharply.
Four things that bring app users back
- A useful first session. People keep an app that helps them the first time and delete one that doesn't. If the first session fails, there is rarely a second. On one client's ordering flow, we talked them out of adding many confirmation dialogs, because the extra steps would have made the app worse to use.
- A place on the home screen. An installed app sits one tap away. A user who stays signed in doesn't meet a login screen on every visit, and doesn't have to remember your web address.
- Notifications about the user's own business. A notification the user cares about brings them back without another ad. In the client app we built for Merezha Prava, a Ukrainian legal-services firm, 77% of users turned notifications on (the case study). The notifications follow the stages of each client's own case. Our write-up of that app covers the three product decisions behind the number.
- Retention compounds. Paid reach has to be bought again every month. A customer who stays keeps coming back with no new acquisition cost.
Put your own numbers on it
Our ROI calculator counts retention for service businesses. It adds 3, 7 or 12 points of retention, depending on the scenario, to the share of revenue that comes from returning clients. You can replace that default with your own figure.
If an app is how you plan to give customers a reason to come back, new app and MVP development shows how we scope a first version. Restaurant and coffee brands can see what loyalty features involve in loyalty features that move repeat purchases. To go through your own numbers with us, book a call.
The original post
As Oleksiy Kazavchynskyy posted it on LinkedIn on 16 June 2026, unchanged. The statistic in its first line is corrected above.
A 5% lift in retention can raise profit up to 95%. (that's Bain's research. Proven over decades). But most founders still POUR money into ads.
Here's the thing: People don't stay because you paid to reach them. They stay because your product solves their problem.
I've built mobile apps for 8 years. And retention rarely lives in the ad budget.
Here's what actually keeps people coming back:
- Day one decides everything. People keep what helps and delete what doesn't. Nail the first session - or you've lost them.
- The home screen is presence. One tap, no login. You're in their day, not waiting to be remembered.
- Push notifications are free return trips. That's the comeback you'd otherwise pay for through ads.
- It compounds. Acquisition resets to zero every month. Retention doesn't.
Most founders ask: how do I get more users? The better question: Do your customers have a reason to come back tomorrow?
An app can be THAT reason. We've built 13 of them → nexterasolutions.app