Chakra Campaign — Revenue-Weighted Incentive Design for a Token Launch

Product

Kriya was a multi-product DeFi platform on Sui - a spot AMM with both uniform (v2) and concentrated-liquidity (v3) pools, plus three vault families: leverage lending, LP auto-rebalancing, and automated market-making on Sui's shared order book. Chakra Season 3 was the final points season before the $KDX token launch - 15% of token supply, unvested, allocated to season participants.

Problem

Every point emitted was a claim on real token value, not a marketing cost. Pool fee tiers ranged from 0.01% to 1% - the same $1M of volume earned the protocol $100 in the cheapest pool and $3,000 in the most valuable one, a 30x difference for identical-looking activity. Reward volume flatly and rational users farm the cheapest-to-churn pools, buying a leaderboard full of wash volume with real token supply. Compounding the problem: most token pairs had no direct pool, so trades routed through 1-3 hops across 15 pools, producing 80+ distinct reward-bearing paths - and the platform's three vault families each generated revenue through a structurally different mechanism, so no single pricing rule could cover all of them.

My role

One of a three-person core strategy team. Owned reward-rate modeling and campaign mechanics end to end - the math, the UI/UX and copy that explained it to users, and coordination with engineering to ship it.

Scope

Campaign economics, reward modeling, and execution across all reward-bearing products, over roughly two months leading into the token launch.

Approach

Reframed the objective from "maximize activity" to maximize revenue per point emitted, subject to enough breadth to keep acquisition healthy. Every downstream decision followed from that single reframing.

• Priced by fee capture, not volume: Each pool's reward rate was set roughly proportional to its fee revenue, then adjusted for strategic weight. Published rates ranged from 0.19 MAYA/$/week on a low-fee stable pair to 4.25 MAYA/$/week on a volatile pair carrying real revenue and real risk. When rewards track fees paid, gaming the system just means paying the protocol.
• Modeled by route-leg, not by trade: A swap with no direct pool fills across multiple hops, each landing in a different pool at a different fee tier, so revenue and rewards were tracked per leg. Hub pools (the two assets sitting on most routes) were rewarded slightly above their raw fee capture, priced as infrastructure that improves execution across the entire routing graph.
• Reduced three different reward logics to one unit: Leverage lending, LP rebalancing, and market-making vaults each earn revenue differently, but every incentive reduced to the same measure: MAYA per dollar per week, making cross-product budget allocation an explicit, comparable decision. Published weekly budgets: leverage lending 6.6M, LP rebalancing 5.7M, alt-asset pools 4.0M, stable pools 3.5M MAYA.
• Ran the model on a spreadsheet, deliberately: Publish rates, watch capital migrate over the two-week epoch, recompute realized revenue-per-point by category, retune. Every rate needed a human-legible justification and same-day iteration speed.
• Layered behavior design on top of the economics: A season NFT that doubled as a referral loop, points burned one-way into the NFT to update rank (killing farm-and-dump dynamics), a manual review layer over the leaderboard metric most exposed to gaming, and a hard campaign deadline ten days before token launch.

The season lifecycle

1. 1

   Mint the Chakra NFT

   Free for S1/S2 users · 2 SUI referred · 5 SUI new user — a referral loop with a price on it

2. 2

   Earn $MAYA

   LP, swap volume, vault deposits, lending — priced 0.19 → 4.25 MAYA/$/week by pool, not by raw volume

3. 3

   Burn points into rank

   One-way, non-transferable — commitment, not a balance. No farm-and-dump.

4. 4

   Redeem for $KDX

   15% of token supply, unvested — token launched 10 days after campaign close

Weekly $MAYA emission budgets

One currency across four product families, so cross-product allocation was an explicit budget call, not a guess.

Pricing a multi-hop trade

Most token pairs had no direct pool. A trade like vSUI → USDC filled across two legs, each landing in a different pool at a different fee tier - so rewards were priced per leg, not per trade.

Outcome

• 70,000+ unique accounts acquired
• $30M to $117M AUM ($34M in strategy vaults)
• 5× transaction volume growth
• ~$700K protocol revenue over the campaign
• Token launched into an active, revenue-generating user base ten days after campaign close