[2024 - 2025]
Increasing Streamer Activity Through Aligned Incentives and Guided Growth
Building a progression system that guides streamers from competition entry to goal-driven growth execution.
I transformed fragmented growth mechanics into a connected engine: rank progression, stage-based missions, and reward reinforcement loops.
ROLE
Product Designer
SCOPE
Ranking system, mission system, reward loop
KEY IMPACT
in target cohorts, streaming duration increased by 44.11% and active stream days increased by 18.18%.
01
Overview
A Two-Layer Growth System for Continuous Streamer Progression
This project built a streamer growth engine on a live platform where supply stability and retention are strongly tied to streamer motivation.
The first phase was Snack Rank, which established a visible progression framework and competitive movement through rank-up and rank-down mechanics.
The second phase was Star Boost, a mission-and-reward layer mounted on top of Snack Rank to help streamers who did not know how to plan their own growth path.
Together, these two systems turned ranking from a passive status display into an active, guided growth loop.
02
Product Context
The Limits of the Previous Incentive Model
The platform needed streamers to keep broadcasting consistently while improving content quality and audience value over time.
Before this work, the incentive model mainly relied on repeating monthly task cycles. Streamers could complete tasks individually, but the system did not create an ongoing competitive structure that encouraged healthy comparison and collective growth across the broader streamer group.
(must) [ image | Comparison board of the old task model versus the new growth engine. ]
Competitive pressure mostly appeared only during campaign periods, rather than as part of the platform's everyday growth model. As a result, streamers had limited visibility into whether they were continuously improving relative to their wider peer group.
Through VOC, we observed that many streamers also did not have a concrete strategy for growth planning. This created a gap between motivation, positioning, and execution.
03
Problem
04
Outcome
We delivered a two-layer growth system:
This created four new capabilities:
Stage-specific mission guidance for streamers at different maturity levels.
Visible short-term and long-term goals that connect daily actions to rank growth
Reward reinforcement at mission completion and full rank-group completion, feeding momentum back into the growth loop.
A reusable mission module that supported faster rollout of future mission sets.
05
Role & Scope
I was the product designer across the progression-system work, from Snack Rank through Rest Day and the later Star Boost extension.
OWNERSHIP
Progression visibility, competitive feedback, mission structure, and reward communication; co-shaped the Snack Rank progression model and translated it into the user-facing system.
WORKING WITH
PM and engineering — turning the mission framework into a reusable module for future growth initiatives.
06
Key Decisions
1
Two-Way Rank Movement
2
Rest Day for Sustainable Participation - Prospect Theory
3
Stage-Based Mission Guidance
(skip) [ image | Framework diagram of rank tiers, mission gradients, and reward checkpoints. ]
4
Two-Point Reward Structure
07
Design Execution
Rank Progression (Snack Rank)
I started from rank progression, because the whole system would stay passive if streamers could not immediately tell whether they were moving up, staying flat, or drifting toward demotion. So I designed the ladder and progression states to make status change legible at a glance. Promotion and demotion communication also had to feel fair. If it felt punitive or opaque, the system would lose trust fast. That is why I always showed current position and next-step targets together. Showing status alone would have created awareness without action.
Pressure Management ("Rest Day")
Rest Day came from a different kind of problem. Once the system started putting real pressure on streamers, I needed a way to reduce that pressure without collapsing the logic of competition. So I designed the rule communication and entry points to make the mechanism understandable without making people feel like they were gaming the system. The key was clarity. Streamers needed to know exactly what paused and what did not, otherwise the feature would create more anxiety than relief.
(nice-to-have) [ image | Flow diagram of backward-looking clip capture from replay moment to editable 90-second output. ]
Mission Guidance (Star Boost)
Star Boost was where I translated competition into guidance. I designed the mission architecture by rank level and by gradients within each level, because one flat mission set would have been noise for advanced streamers and overwhelming for early-stage ones. I mapped missions to stage goals such as accumulated comments, follower growth, broadcast duration sprints, paid subscription growth, and live-clip highlights publishing. The point was to make the mission set itself feel like stage guidance, not just more tasks.
(must) [ image | Mission matrix by rank tier and growth stage. ]
Then I built the feedback layer around it: mission status, completion signals, and reward triggers, so streamers could see how close they were to the next payoff instead of guessing. I also defined the module so the same mission logic and UI patterns could be reused later without rebuilding from scratch. The system only started to feel complete when rank progression, pacing relief, and mission guidance read as one connected loop instead of three separate mechanics. That was the point where the design stopped feeling like a set of features and started behaving like one incentive system.
(must) [ image | Before-and-after comparison of passive ranking versus guided mission flow. ]
08
Impact
Active Streamer Rate
+0.9%
Overall uplift, with the strongest gain of about +8.1% in early-stage tiers.
Streaming Duration
+44.11%
In target cohorts, alongside +18.18% active stream days and +44.64% progression gain.
Viewer Quality
+11%
Full snack completion rate, with +8% average contribution and +6% qualified viewing quality.
Livestream re-engagement
+3.3%
`Entry-tier mission completion`
Entry-tier mission completion
40%+
With early-to-mid tiers improving in participation (+2%), interaction depth (+4.12%), and rank contribution (+6.33%).
What This Indicates
Rank visibility alone was not enough; pairing competition with guided missions created a more reliable activity and progression loop.
The strongest gains appeared in early-to-mid tiers, showing that explicit goals and stage-based incentives were especially effective for streamers who lacked a clear growth plan.
09
Reflection
What worked well
The strongest bet was changing progression from individual accumulation into an always-on competitive system among streamers at the same level. That shift turned growth from a private, repetitive task cycle into a visible behavioral loop, and Star Boost later worked because it was mounted on top of that stronger competitive foundation rather than replacing it.
What was challenging
The system needed long-term consistency to work, but operational strategy and budget varied by region. Some regions adopted the new incentive model while others continued using the old one, which made the growth framework harder to roll out as a shared platform rule and reduced how fairly outcomes could be compared across markets.
What to improve next
I would leave more design room for pacing and pressure-relief mechanisms earlier in the system. "Rest Day" only became obvious after strong streamers reacted to the live pressure of the model, which showed that sustainability features may not fully surface until a competitive system meets real usage at scale.
✨
•
Made with love, Framer, and Codex