BricqsBricqs

Badge systems that actually drive behaviour

Badges look free. They are not. Every badge you ship has an ongoing cost: design, placement, the question of when to award it, the user who emails support because they think they earned one they did not. Done well, badges shape onboarding, retention, and feature discovery. Done badly, they become a row of grey icons nobody notices. This guide is the marketing team's playbook for picking what to badge, what not to badge, and how to measure it.

Last updatedMay 2026
Reading time10 min
AudienceGrowth, lifecycle, product

Key takeaways

Quick read
  • Badges are recognition, not currency. They mark accomplishment; they do not buy anything.
  • Five to seven badges at launch beats thirty. Rare beats common; specific beats generic.
  • If a user can earn a badge by doing nothing unusual, the badge has no signal value. Save badges for moments that deserve recognition.
  • Track earn rate, collection breadth, and completion path. Anything below 30% earn rate in 30 days means the program is invisible or the triggers are too hard.
  • Hidden badges add intrigue. Use sparingly for surprise moments (a first-anniversary badge, a hidden achievement) where the reveal is half the reward.
  • Pair badges with the underlying behaviour, not with arbitrary thresholds. A badge for 'completed 100 quizzes' has no soul; a badge for 'finished the security training' has clear meaning.

Definition

What a badge system actually is

Plain definition

A badge is a named, permanent recognition awarded to a user for doing something specific. The user collects it, sees it on their profile, and keeps it forever. Badges do not change the user's points balance, do not expire, and do not affect their tier.

Who runs this

Lifecycle and growth teams choosing what to recognise; product managers deciding which moments deserve a celebration; engineers wiring badges into product surfaces.

Why it works

What makes badges actually drive behaviour

Badges are not decoration. The mechanic taps four well-studied behavioural patterns, and when the program design lines up with those patterns, badges move retention, activation, and feature adoption in measurable ways.

Status signalling on profile

Humans seek visible markers of competence. A profile that shows three earned badges out of twenty available creates an unfinished story; users return to close the gaps. The signal works on the participant who earned the badge and on the people who see it.

Collection completion

Half-finished sets pull stronger than empty ones. Once a user has earned two badges in a category of five, the cost of returning to finish the set feels lower than the cost of starting one from scratch. This is why category-shaped badge programs out-retain flat lists.

Surprise reveals (variable reward)

Hidden badges that appear unprompted create one of the cleanest variable-reward loops in product design. The participant did not expect the recognition; the celebration moment registers as a gift rather than a transaction. These are the badges that get screenshotted.

Cheap visible progress

Compared to points or tiers, badges are visually compact and easy to render in any surface. A first-week activation flow with five tappable badges costs almost nothing to render but communicates progress at a glance. The economics favour badges for any surface where attention is short.

The business case follows from the behaviour. Programs that use badges as visible activation milestones report 8 to 15 percent lifts in first-week return rate; community programs that surface top-contributor badges see 15 to 30 percent more inbound posts in the recognised categories; onboarding flows pinned to five-badge checklists complete 20 to 40 percent more often than unstructured nudge sequences. The exact numbers depend on the audience and the underlying behaviour, but the direction is consistent: visible recognition tied to behaviour the participant already wants to be doing.

Core mechanics

The four parts of a badge program

Every badge program decomposes into the same four design decisions. Get them right at launch; revisit them once a quarter.

Rarity tiers

Five tiers (common, uncommon, rare, epic, legendary) are the standard ladder. Common badges drive activation; legendary badges drive long-tail mastery. Each tier carries a visual treatment (border, glow, animation) so users can scan their collection and see what is hard at a glance. Skip tiers that do not have a clear earn rate target.

Categories

Group badges by the behaviour family they recognise (onboarding, mastery, community, feature discovery, seasonal). Categories make the collection scannable and the program legible to a new participant. Five to seven categories is the sweet spot; more than ten and the collection page becomes overwhelming.

Visibility (visible vs hidden)

Most badges should be visible from day one so participants can see what is possible. Hidden badges sit in the catalog but reveal only on award; they are the surprise mechanic. A program with 80 percent visible and 20 percent hidden badges captures both planned engagement and delight moments.

Trigger model

Three patterns dominate. Server-evaluated rules (the platform watches state and awards automatically), engagement-completion actions (a quiz or game completion includes the award as a side effect), and direct API calls (your backend awards when a domain event fires). Most programs use all three; pick the simplest one per badge.

One decision worth getting right at launch: the badge code (the stable identifier you use in API calls and analytics) cannot be changed without re-issuing every existing award. Use snake_case and pick names that describe the behaviour, not the marketing copy. first_quiz is correct; quiz_starter_v2_2026 is a tax you will pay forever.

When badges fit

Five moments where badges earn their keep

If the moment fits one of these, badges are the right primitive. If it does not, consider points, tiers, streaks, or just a notification instead.

Onboarding

Give new users a visible to-do list disguised as recognition.

First-week badges (completed profile, finished tour, sent first message) make activation steps feel earned rather than nagged. Pair with a checklist UI and you have a guided onboarding without writing any walkthrough.

Feature discovery

Push users toward features they have not tried yet.

A badge for the first use of a power feature signals it exists, gives a small win for trying, and gives growth a measurable goal: how many users have the badge after 30 days.

Mastery recognition

Mark accomplishments that took real effort.

Rare-tier badges (completed all advanced courses, hit a personal milestone) carry signal because they are hard to earn. The badge becomes a status marker on the user's profile.

Community contribution

Recognise the behaviours your community depends on.

Top contributor, helpful answer, first to report a bug. The badge tells the user 'we see you' and tells everyone else 'this person carries the community'.

Surprise and delight

Anniversary moments, hidden achievements, easter eggs.

A first-year-with-us badge that appears unprompted gets screenshotted. A hidden badge for an obscure action gets shared. These create the social moments good loyalty programs depend on.

When to skip

When badges are the wrong tool

If your use case is in this list, pick a different primitive. Forcing badges into the wrong shape is how you end up with a graveyard of icons nobody cares about.

Do not use badges when...
  • The user should be able to spend the recognition
    That is points or a reward, not a badge. Badges have no spend value. A 'Free Coffee Badge' will confuse users when they discover it does not redeem to anything.
  • You want it to expire or recycle monthly
    Badges are permanent by design. For monthly status, use a tier (Bronze, Silver, Gold can shift) or a streak. Expiring badges trigger 'where did my badge go' support tickets every month.
  • The trigger fires constantly
    Awarding a badge per login devalues the system. If a behaviour fires more than once a session, model it as a counter or a streak, then award one badge at a meaningful threshold.
  • It is a ranking or leaderboard
    Ranking is for points or a score. A 'Top 10 of the Month' badge means stripping it the next month, which violates the permanence contract and creates angry users.
  • It is a short campaign that ends in two weeks
    Badges are forever, so a campaign badge sits in the user's profile forever as an awkward archaeology layer. Use a contest entry, a streak, or a time-limited tier instead.

Measure it

Three numbers that tell you if badges are working

Most teams launch badges, watch them get awarded, and never go back. Track these three KPIs in a monthly review and you will know when to add badges, retire badges, or rethink the whole program.

KPI 01
30-day earn rate
30% or higher
Percentage of monthly active users who have earned at least one badge in the past 30 days. The single best indicator that your badge program is actually visible.
Watch for: Below 30% means badges are invisible, the triggers are too hard, or the rewards do not match the work.
KPI 02
Average collection size
3 to 6 badges per engaged user
For users with at least one badge, the average number of badges they hold. Tells you whether users come back for more or earn one and forget.
Watch for: Below 3 means low retention on the program. Above 6 with only 10 badges total means you are giving them out too freely.
KPI 03
Onboarding completion path
50% or higher
For multi-badge onboarding journeys, the percentage of users who finish the full set. The most actionable engagement KPI.
Watch for: Below 50% means one of the steps is broken, hidden, or not worth the work to the user.

Best practices

Seven rules of a badge system that lasts

  1. 01
    Launch with five to seven badges, not thirty
    Programs that ship with thirty badges always have ten that nobody earns and fifteen that everybody earns by accident. Five well-chosen badges outperform thirty noisy ones.
  2. 02
    One badge per real behaviour
    A badge per real moment ('completed your first review', 'sent five referrals') has meaning. Generic badges ('user level 1') do not.
  3. 03
    Use rarity tiers to signal effort
    Common, rare, epic, legendary. Map each tier to a visual treatment (border, glow, animation). Users learn the system at a glance.
  4. 04
    Show locked badges, with hints
    Hidden badges feel mysterious; locked-but-visible badges feel motivating. A grayed-out 'Power User' tile with the criteria visible tells the user what to do next.
  5. 05
    Celebrate the moment of earning
    A modal, toast, or full-screen reveal at the moment a badge is awarded multiplies the impact. Surfacing it only on a profile page nobody visits wastes the work.
  6. 06
    Refresh the catalog seasonally, not monthly
    Annual or quarterly badge drops give long-time users something new to chase without making the existing collection feel stale. Monthly badges are a treadmill.
  7. 07
    Audit yearly: retire what nobody earns
    A badge that 1% of users earn in a year is not rare, it is broken. Either lower the bar, remove it, or replace the trigger. Dead badges devalue the whole catalog.

Common mistakes

Where badge programs go wrong

Each of these is something a real team has shipped and rolled back. Recognise the pattern before you ship it.

01Mistake

Awarding badges for trivial actions to make the system feel 'alive'. A badge for opening the app is a badge for breathing. Users see it for what it is and stop trusting the system.

Fix

Award badges for things the user worked for. Save the easy moments for points or a quick toast notification instead.

02Mistake

Using a badge to gate a feature ('unlock dark mode at 10 referrals'). The user feels punished, not rewarded, and the gating logic becomes brittle.

Fix

Gate features with a tier or feature flag. Badges are recognition, not access control. They have different lifetime semantics.

03Mistake

Shipping a badge for an event campaign ('Summer 2026 challenge'). Two weeks later the campaign is over but the badge lives in the user's profile forever, looking dated.

Fix

Use a contest entry or a time-bound streak for campaigns. Save the permanent profile space for badges that age well.

04Mistake

A profile page with 40 grey lockable badges and 2 earned. The page looks like a failing grade.

Fix

Group badges into categories (Onboarding, Mastery, Community). Show categories with progress bars. Hide the noisy 'all locked' grid.

05Mistake

Awarding a badge for purchase. The user can buy recognition, and the badge stops carrying signal.

Fix

Badges should mark behaviour, not transactions. For transactions, use points (which the user can compare and spend).

In the wild

Three badge programs that work

Language learning (Duolingo-style)
Streak loop
01
14day streak
Repeat useHabit

Achievement badges tied to streaks, lesson milestones, and feature discovery.

What it is buying

Badges become inseparable from the streak system. A 30-day streak badge is harder than 100 lessons because it requires consistency, not effort. The result: users protect their streak, and the badge becomes social status.

Community Q&A (Stack Overflow-style)
Campaign pattern
02
Capture
Engage
Reward
ParticipationReward

Reputation-aligned bronze/silver/gold badges for contribution behaviours.

What it is buying

Tiers map directly to the difficulty of the underlying behaviour. A 'Famous Question' gold badge (10k views) is rare and respected; the 'Editor' bronze badge (first edit) is easy and onboarding-y. The hierarchy is self-explanatory at a glance.

Fitness and athletics (Strava-style)
Campaign pattern
03
Capture
Engage
Reward
ParticipationReward

Activity-specific achievement badges scoped to monthly challenges.

What it is buying

Each monthly challenge awards a badge for completion. The collection becomes the user's record of what they have done, not a generic loyalty trophy case. Tightly tied to behaviour, never abstract.

Implementation

How to ship the program in Bricqs

Bricqs ships the badge engine, the trigger rules, the rarity and category model, the hidden-badge mechanic, and the React rendering helpers. Configure from the dashboard or wire through the API.

Frequently asked

Common questions before launch

Q01How many badges should we ship at launch?

Five to seven. Each one tied to a specific behaviour worth recognising. Programs that launch with thirty badges almost always have ten that nobody earns and fifteen that everybody earns by accident, and the remaining five do all the work. Start small, learn what your users actually engage with, then add.

Q02What is the difference between a badge and a tier?

A badge is a recognition the user collects and keeps forever. A user can have dozens of badges. A tier is a status level (Bronze, Silver, Gold) and a user is at exactly one tier at any time. Badges accumulate; tiers replace.

Q03Should badges expire?

No. Badges are by design permanent. If you need a recognition that disappears (e.g. monthly leader, weekly streak), use a tier or a streak. Both have built-in time semantics. Expiring badges create user confusion and support tickets every month when they vanish.

Q04Can a user re-earn a badge?

Most platforms (Bricqs included) treat badges as one-and-done by default. If the user has the badge, awarding it again is a no-op that returns the original earned_at. If you want repeatable recognition (e.g. monthly contributor), model it as a different mechanic: a streak, a contest, or a counter with milestones.

Q05Where should badges show up in the product?

Three places. (1) An always-on profile or progression page where the full collection is visible. (2) A celebration moment right when one is earned (modal, toast, or full-screen reveal). (3) A category-grouped onboarding checklist for the first-week badges, so new users see what to do next.

Q06How do we measure if badges are working?

Track three KPIs. (1) Earn rate: what percentage of active users have earned at least one badge in the past 30 days. (2) Collection breadth: average badges per user with at least one. (3) Completion path: for a multi-badge journey (e.g. onboarding), what percentage finish the whole set. Anything under 30% on the first metric means your badges are not visible enough or the triggers are too hard.

For developers

Bring it to life

Wire your first badges in under an hour

The developer guide pairs with this strategy doc and shows the API, the React hook, and three common award patterns. If your engineering team is implementing, send them there.

1 brief to align the room2 mechanics max in version one
What happens next
01
Pick the mechanic
Choose the smallest working system for the brief.
02
Launch without rebuilds
Configure rules and rewards in one place.