Most growth dashboards still lead with a single retention curve. It starts at 100% on day zero, falls hard in the first week and flattens somewhere uncomfortable. Everyone in the room agrees the curve should be higher. Almost nobody can say which users to call tomorrow.
That gap costs real money. An HBR article on customer retention cites research by Frederick Reichheld of Bain & Company showing that increasing customer retention rates by 5% increases profits by 25% to 95%. The upside is large, but a curve cannot tell a team where to act.
What a single curve hides
A day-30 retention rate of 20% sounds like one number, but it blends very different people: daily regulars, users who came back after a month away, and users who have gone quiet this week after weeks of steady use. That last group is the one most worth saving.
Averages flatten those groups into one line. A growth team that only sees the line ends up running broad campaigns to everyone, which annoys loyal users and arrives too late for the ones already gone.
Where growth teams lose users they paid for
The alternative is to sort users into stages based on recent behavior, then treat each stage as its own audience.
Asked where growth teams lose users they already paid to acquire, the team at Trophy, which sells gamification software to consumer apps, pointed to the gaps a single retention curve hides: “We track users through seven retention stages, so a team can spot someone who has gone quiet for a week after 23 active days and reach them before they go dormant.” That view comes from a company whose Gamification Intelligence Report 2026 draws on more than 1.5 million users and 250 million interactions.
The example in that answer is the important part. A user with 23 active days in a month has shown real intent. A week of silence from that person is a much stronger signal than a week of silence from someone who signed up and never returned. Stage-based models make that difference visible.
A practical set of stages
Teams can build their own version of this model with data they already collect. A workable set looks like this:
- New: signed up in the last few days and still forming a first impression.
- Current: active on at least two of the last seven days.
- At risk: a regular user who has gone quiet for about a week.
- Reactivated: back after one to four weeks away.
- Resurrected: back after more than a month away.
- Dormant: silent for more than a month.
The exact thresholds matter less than applying them consistently. Pick definitions that match the natural rhythm of the product, publish them to the team, and stop changing them every quarter.
A different play for each stage
Once users sit in stages, the right action for each becomes much clearer.
At-risk users respond best to something personal and specific. A reminder of their own progress, such as a streak about to end or a milestone they are close to, gives them a reason to return that a generic “we miss you” email lacks.
Reactivated users need an easy win in their first session back. If the product greets them with a backlog of unread items or a broken streak, many leave the same day again.
Resurrected users often need a short refresher. The product may have changed while they were away, and a quick tour of what is new helps them rebuild the habit.
Dormant users deserve one well-timed attempt. Persistent messages to people who have left mostly teach email providers to filter the brand.
Timing matters as much as the message
A well-written message sent at the wrong hour still fails. The simplest improvement most growth teams can make is to send stage-based messages in each user’s local time, close to the hour they usually open the product.
A user who logs a workout every morning at 7 am is most likely to act on a reminder at 6.45 am. The same message at 3 pm competes with a working day. Sending by local habit instead of by a single campaign time takes some setup, but it turns a batch send into something that feels personal.
Common mistakes with stage-based campaigns
Three errors show up again and again. The first is moving thresholds too often, which makes week-on-week comparisons meaningless. The second is sending every stage the same offer, usually a discount, which teaches loyal users to wait for one.
The third is ignoring the stage a user moves into after a campaign. A reactivated user who slips back to dormant a week later is a failed campaign, whatever the open rate says.
Measure how users move between stages
The most useful weekly report for a growth team is a movement table. How many users moved from current to at risk? How many at-risk users returned to current? How many reactivated users stayed active a second week?
Those numbers show whether each campaign works on the stage it targets. They also reveal problems a curve hides, such as a product change that pushes loyal users into the at-risk stage while overall retention looks flat.
For context, Mixpanel’s State of Digital Analytics benchmarks give teams a way to compare their engagement metrics with peers in the same industry, which helps set realistic targets for each stage.
Where gamification fits
Streaks, points, and milestones work best as tools for specific stages. A streak gives current users a reason to keep going. A milestone close at hand gives an at-risk user a reason to return. A welcome-back reward gives a reactivated user the easy win they need.
Used this way, gamification stops being a feature request and becomes a set of answers to a specific retention problem. The teams that get the most from it start with the stages, find where users leak, and pick the mechanic that plugs that exact gap.
