Progression systems for campaigns that keep moving
A LinkedIn user who reaches Top Voice keeps writing. A Reddit user who hits 1,000 karma keeps posting. A Sephora Insider who sees 'two more visits to Rouge' keeps shopping. None of those campaigns ended in a week; they kept moving because the user always had a next step. This guide is how to build that next-step muscle into yours.
Key takeaways
- Progression is mostly visibility. Show the user where they are and what is next; most of the lift is paid for before any reward issues.
- Five primitives cover almost every program. Points, levels, milestones, streaks, badges. That's it.
- Pick one as the spine. Stack one or two as supports. More than three primitives in one program is usually a feature dump in disguise.
- Pacing is the design discipline you'll keep tuning. Too steep, users quit. Too shallow, the system feels decorative.
- Silent resets and unannounced rule changes are the fastest way to break trust. Communicate everything in advance, every time.
Definition
What progression actually means
Strip the framework. Progression is the small set of signals that tells a user 'you are here, this is what is next, and here is what changes when you get there.' Strava does it with segment ranks. Duolingo does it with leagues and streaks. Sephora does it with tier progress bars. The mechanic is universal; the dressing changes by category.
Plain definition
A progression system is a set of rules and visible signals that make ongoing engagement feel like an unfolding journey. Points accumulate, levels unlock, milestones get hit, and the user can see their distance to the next reward at any moment.
Who runs this
Lifecycle, growth, and product marketing teams who own retention or activation. Engineering supports the surface, finance signs off on the rewards, marketing operates the program.
How it differs from adjacent mechanics
- vs loyalty programs. Loyalty is the long-running business framework. Progression is the user experience inside it. Many loyalty programs are weak because the progression layer is invisible.
- vs single rewards. A coupon ends with one redemption. Progression turns that same reward into a destination on a longer arc.
- vs achievements alone. Achievements are honors. Progression is direction. Achievements without a path between them feel arbitrary.
- vs leaderboards. Leaderboards rank users against each other. Progression measures the user against themselves. Pair them carefully; one can crowd the other out.
The five primitives
The pieces every progression system is built from
Five primitives. Headspace's spine is streaks. Sephora's spine is tiers. Reddit's spine is karma points. Duolingo's spine is XP per league with streaks bolted on. Pick one as the load-bearing wall, layer one or two more as decoration, ship.
Points
The continuous fuel of progression. Earn for any action that matters. Points are flexible and easy to communicate, and they double as the input to levels and milestones.
Levels and tiers
Discrete status steps the user climbs (Bronze, Silver, Gold). Levels turn an open-ended point total into a clear next step with a meaningful unlock at the top.
Milestones
Specific accomplishments along the path: completed onboarding, finished 5 quizzes, made 3 referrals. Each milestone carries its own micro-reward and recognition.
Streaks
Consecutive days or visits. The habit primitive. Best for daily and weekly behaviors. Always include a forgiveness rule so a missed day does not erase weeks of work.
Badges
Earned, displayable status markers. Best as recognition of unusual or milestone behavior, not for every login. A few rare badges beat dozens of common ones.
The ladder
Status as the spine of the program
Tier ladder
Status climbs as lifetime value compounds
The ladder is the canonical progression UI. The current tier glows; the next tier is one step up; the lifetime reach of each tier is visible at a glance.
Combinations
Which primitives work together
Not every primitive needs to be in the program. Most of the time fewer is better. The table below is the working playbook.
| Goal | Spine | Supports | Skip |
|---|---|---|---|
| Activate new users in 7 days | Milestones (3 to 5 onboarding steps) | Points for each step, badge for completion | Streaks, leaderboards |
| Build a daily habit | Streaks | Points multiplier on long streaks, badge at 7/30/100 days | Levels (too slow to feel) |
| Drive repeat purchase frequency | Levels (3 to 4 tiers) | Points, milestones at each tier change | Badges (decorative, not load-bearing) |
| Run a 4-week engagement push | Milestones (weekly chapters) | Points, leaderboard for top performers | Levels (campaign too short) |
| Build a community of contributors | Badges | Points, levels for top contributors | Streaks (contribution is irregular) |
Pacing
Set the curve so it always feels reachable
The shape of the curve is the most underrated decision in the program. Get this wrong and no amount of creative recovers it. Most working programs use a J-curve: easy at the bottom, harder at the top, never impossible.
Front-load the first reward
The first reward should arrive in the first session or within 1 to 3 days. The first redemption is the strongest predictor of long-term engagement.
Use a J-curve, not a straight line
Early levels go fast. Later levels take 3 to 5 times longer. Mirrors how meaningful progress works and keeps committed users invested.
Always show the next step
Every member-facing surface should answer: how close am I to the next reward. A progress bar with a target is more motivating than a richer reward at the end.
Decay the curve, not the user
If a user is grinding and not progressing, the curve is too steep, not their effort. Adjust earn rates quarterly and communicate every change clearly.
Best practices
Seven rules that hold across categories
- 1
Every visible surface answers 'where am I'
The home screen, account page, email header, push notification: every touchpoint should make the next step obvious. Users do not memorize point totals; the program does.
- 2
Make the first milestone a 1-session reward
If users have to wait until visit 5 for any payoff, most never get there. Drop a small win at signup or in session one. The cost is tiny; the lift in week-1 retention is usually 15 to 25 percent.
- 3
Tie rewards to the goal, not the action
If you reward time spent, users will pad time. If you reward completion of a meaningful step, users will complete the step. Pay for the outcome you want.
- 4
Communicate every rule change
Earn rates, tier thresholds, and reward catalog changes are major events for power users. Email and in-app notice 14 to 30 days before changes go live. Silent changes are how trust dies.
- 5
Cap the maximum point total
Unbounded balances produce balance-sheet anxiety and user disengagement (the next reward never feels closer). Cap or auto-redeem at a clear ceiling.
- 6
Reset windows, not status
Annual point resets feel punitive and break trust. Reset competitive windows (weekly leaderboards, monthly streaks) but let lifetime status carry over.
- 7
Treat dormancy as a signal, not a failure
Inactive users are an opportunity, not a write-off. Trigger a winback after 30, 60, 90 days with progressively softer asks. A small re-entry reward usually outperforms an aggressive expiry email.
Use cases
When progression is the right call
Onboarding
Five-step progression bar across day-1, day-3, day-7 milestones with small rewards at each.
Day-7 activation typically lifts 15 to 30 percent. The bar does most of the work because users hate leaving it incomplete.
Habit-driven products
Streak with one freeze per week, points multiplier on long streaks, badge at 7/30/100 days.
Daily active users lift in habit-friendly categories like learning, news, fitness, and finance. Compounds over months.
Premium upsell
Tier-based progression with each tier unlocking a more meaningful perk: free shipping, early access, exclusive drops.
Top-tier members carry disproportionate revenue. The aspirational tier sells the program.
Community and contribution
Earned badges plus community-only levels that unlock visibility (verified flair, top-contributor surface).
Active contributor share rises and contribution quality improves because status is at stake.
When to skip
When progression backfires
The user does not need to come back
If the product is purely transactional and one-shot, a progression system pretends to add value but adds friction. Use clean rewards instead.
There is no real reward at the end
Climbing a ladder to nowhere produces short-term participation and long-term resentment. If the unlocks are not meaningful, remove the system.
The brand context is sensitive
Healthcare claims, debt repayment, grief services. Progression framing reads as flippant. Use careful language or skip the mechanic.
Progress can be reset for reasons the user does not control
If outages, refunds, or partner failures can wipe progress, the system creates more support tickets than goodwill. Fix the data layer first.
Common mistakes
Where progression systems quietly fail
Mistake
The first reward is too far away. Users sign up, see the goal is 5 visits out, and never come back.
Fix
Drop a small reward in session 1. Make the first milestone reachable in one to three actions. Move richer rewards later in the curve where committed users are.
Mistake
Too many primitives at once: points, levels, badges, streaks, and leaderboards in the same campaign.
Fix
Pick one primitive as the spine and at most two supports. More primitives is more rules to communicate, not more motivation.
Mistake
Streaks with no forgiveness. One missed day erases everything and most users do not come back.
Fix
Build in a freeze, grace period, or recovery action from day one. The mechanic is loss aversion. The brand still has to feel humane.
Mistake
Hidden balance and progress. The user has no idea how close they are to the next reward.
Fix
Surface the next milestone on every relevant screen and email. Progress visibility is the cheapest, highest-leverage thing in the system.
Mistake
Point values communicated in absolute terms (1,000 points) without translation to value (worth a 50 rupee discount).
Fix
Always show what the points are worth in money or perk language. Keep the abstract token; explain it in concrete terms next to it.
Measurement
The KPIs that prove progression is working
| KPI | What it measures | Healthy range |
|---|---|---|
| Activation milestone hit rate | Share of new users who reach the first paid-for milestone within the activation window. | 55 to 75% |
| Average time to first reward | Hours or sessions between signup and first reward earned. The strongest leading indicator of long-term engagement. | Within 1 to 2 sessions |
| Tier upgrade rate | Members who moved up at least one tier in the period. Signals the curve is reachable. | 10 to 25% per quarter |
| Streak resumption rate | Users who resume a streak after using a freeze. The cleanest signal that forgiveness is doing its job. | 55 to 75% |
| Repeat-visit lift on members vs non-members | Visit frequency for users in the program vs a matched control group. | +15 to +35% |
In the wild
Three patterns that work
Language learning
Daily streak with one weekly freeze, points per lesson, league-style weekly leaderboard, rare achievement badges.
Outcome. Streak is the spine. Daily active users compound for months because users will not let the streak break.
Beauty retail
Three-tier program (Insider, VIB, Rouge), points per dollar, milestone unlocks at each tier (birthday gifts, samples, early access).
Outcome. Aspirational top tier drives behavior of the middle tier. Top tier delivers 20-plus percent of revenue from 1 to 3 percent of customers.
Fitness app
Weekly milestone (3 workouts), monthly milestone (12 workouts), badge at 100 workouts, streak as a softer side metric.
Outcome. Weekly milestones are the engine. Monthly milestones keep the medium-frequency users invested. Badge at 100 sells the program in social posts.
Implementation
Build this with Bricqs
Bricqs handles the points engine, tier rules, milestone evaluators, streak periods, and badge issuance. Plug it into your existing app via the SDK or run member-facing surfaces straight from the builder.
Frequently asked
What teams ask before they design a progression
How many tiers should a program have?
Three to four is the working default. Two feels under-baked, five-plus dilutes meaning. Set the thresholds at the 80th, 95th, and 99th percentile of customer value and re-qualify annually.
How do we balance individual progression and competitive leaderboards?
Lead with personal progression. Add leaderboards as an optional layer for users who self-select into competition (a separate tab, segmented by cohort). Forcing every user into a leaderboard demoralizes the bottom 80 percent.
Should we let users see other users' tiers?
Yes for community-driven products (verified flair, top-contributor lists). Optional or hidden for private categories like banking or insurance. Status that is invisible to peers does less work than status that is visible.
How often should we adjust the progression curve?
Review quarterly. Adjust no more than twice a year. Communicate every change in advance. Frequent silent tuning erodes trust faster than rare deliberate tuning.
What is the difference between a milestone and a tier?
Milestones are specific accomplishments (complete onboarding, finish 5 quizzes). Tiers are status thresholds based on accumulated value (Silver, Gold). A program can have both: milestones drive activation, tiers drive long-term retention.
Ready to ship?
Try Bricqs free, or talk to a strategist
Plan a campaign, configure the engine, and ship in days. No credit card required to start.