Most liquidity mining campaigns fail quietly. Across 50+ campaigns tracked through Turtle's analytics (EVM L1s and L2s, Q2 2023 through Q4 2025), protocols typically lose 40-70% of campaign-attracted capital within 30 days of the final emission. In the sharpest cases, 42% exits within 24 hours and 70% is gone by day three.
Campaign design determines retention. The protocols that keep capital structure their campaigns for it from the outset.
This playbook covers seven steps to designing a campaign that retains capital. It applies to DEX pools, lending vaults, and pre-deposit campaigns.
Where Incentivized Liquidity Lives
DEX pools carry the highest mercenary risk: LPs can exit instantly, IL compounds in volatile pairs, and aggregator capital is structurally temporary.
Lending and yield vaults attract conservative, longer-horizon capital with no IL risk. Morpho operates through curators (management + performance fees). Euler is permissionless. Aave requires governance approval. Compound moved away from emissions entirely.
Pre-deposit campaigns open deposits before a product goes live. Turtle deployed over $50M into Decibel's pre-deposit vaults. Katana attracted $232M pre-launch. Pre-deposit LPs self-select for conviction, so retention outperforms both pools and vaults.
Points-based systems convert non-transferable points to tokens at TGE. Protocols that launched with ambiguous conversion terms consistently saw sharper capital exits than those with transparent mechanics.
Distribution infrastructure splits into two models: self-serve engines (Merkl: $1.5B+ distributed, 60+ chains) and full-stack platforms (Turtle: $5.5B deployed, 430K+ wallets, combining distribution with LP sourcing and advisory).
Step 1: Define Your Objective
The most common mistake is starting with the budget. "We have 5M tokens to distribute" is a constraint, not an objective.
Objective types shape everything downstream: bootstrapping new liquidity, deepening pools, defending against competitors, supporting a token launch, or growing ecosystem TVL. Each has different LP profiles and success metrics. A vault that attracts $100M in deposits but generates zero borrowing demand has a structural problem: organic yield won't sustain retention.
The target asset matters as much as the venue. Stablecoins are easiest and stickiest (no IL, deep LP base). ETH and BTC require clearing the native staking yield bar (3-4% for ETH). Exotic tokens are hardest: small LP base, high IL, highest mercenary ratio.
Define your objective as a measurable outcome with a timeframe. "$50M in stablecoin vault deposits within 60 days, with at least 40% retention at 90 days" is an objective. "Grow TVL" is not.
Step 2: Model Your Budget and Cost Per TVL
Work backwards from your objective to a budget.
Metric | Range | Notes |
Cost per $1 TVL at peak | $0.10-$0.30 | During active incentive period |
Cost per $1 TVL retained at 90 days | $0.25-$1.00 | True cost of sticky capital |
Median campaign duration | 8-12 weeks | Shorter campaigns skew mercenary |
Source: Cost-per-TVL Benchmarks, Turtle Research
These ranges shift by asset and venue. Stablecoin vaults: $0.15-$0.40 retained at 90 days. Volatile DEX pools: $0.50-$1.00+. Pre-deposits can be cheaper if the product delivers.
Budget in dollar terms across bull, base, and bear token price scenarios. A 50% token price decline mid-campaign halves your effective incentive value.
Factor in operational costs (15-25% on top of incentive spend): engineering, analytics, LP relations, legal review, platform fees, and curator fees if using managed vaults.
Minimum for meaningful retention: 8-12 weeks. Best 90-day retention comes from 12-16 week campaigns with structured taper.
Step 3: Choose Your Venue
Map your objective to the right venue type. Need trading liquidity? DEX pools. Need deposit depth for lending? Vaults. Launching a new product? Pre-deposit. Many campaigns span multiple types.
For vault campaigns, platform choice matters. Morpho gives the most control but requires a curator. Euler deploys faster but you handle risk parameters. Aave has the largest existing base but requires governance approval.
Concentrate incentives where your users already are. If 80% of your token's volume is on Uniswap, don't split across five DEXs. For pools, 2-3 primary venues is a reasonable starting point. For vaults, 1-2 platforms is often sufficient.
Match fee tiers to volatility: low fees (0.01%, 0.05%) attract sophisticated LPs, higher fees (0.30%, 1.00%) attract retail. For vaults, tight LTV ratios attract institutional capital that avoids aggressive-leverage markets.
Step 4: Design Your Emission Schedule
A flat emission rate followed by a cliff to zero is the default failure mode.
Tapered emissions front-load to attract attention, then step down. Pool campaigns typically use a 150/100/60/30% profile over four phases. Vault campaigns taper more gently (120/100/80/60%) since vault depositors are less rate-sensitive.
Cliff-and-vest adds a retention layer. LPs earn tokens but only claim a portion upfront; the rest vests over 30-90 days, with early withdrawal forfeiting unvested rewards. Campaigns using vesting saw 20-35 percentage points higher 90-day retention.
Lock-based incentives borrow from Curve's vote-escrowed model: boosted multipliers proportional to lock duration (1x/1.5x/2.5x for 30/90/180 days). 40%+ retention improvement over standard emissions, with lower participation but dramatically stickier LPs.
Conditional emissions tie rates to KPIs: base rate below TVL target, reduced above it, bonus at utilization thresholds. Effective for vaults where the goal is capital that gets lent out, not capital that sits idle.
For pre-deposits, the design question is how aggressively rates drop at launch. Decibel used a boosted pre-deposit allocation transitioning to competitive post-launch yields; capital largely stayed.
Step 5: LP Sourcing and Curation
A dollar from a yield aggregator that exits in 48 hours and a dollar from an institutional LP with a 6-month horizon look identical on a TVL dashboard. They behave very differently on day 31.
DEX pool campaigns skew broadcast. Vault campaigns can be targeted through access controls and deposit minimums. Pre-deposits are inherently targeted: Katana's $232M came through coordinated outreach, not open farming.
Institutional LPs and DAO treasuries are harder to attract but much stickier. Yield aggregators are inherently mercenary. Retail tends toward longer deposit durations than aggregator capital.
Anti-mercenary mechanisms work in combination: tapered emissions, vesting, locks, deposit minimums, and LP quality scoring. No single mechanism solves it. For vaults and pre-deposits, qualify LPs before they deposit.
Step 6: Build Your Measurement Framework
Track against the cost-per-TVL benchmarks from Step 2. Add retention rate targets (30d: >50%, 60d: >35%, 90d: >25%) and organic fee ratio (fee revenue / incentive spend, target >0.15 by campaign end).
Track LP concentration (HHI below 0.15; above 0.25 is danger) and utilization rate for vaults. If one wallet holds a third of your vault, your retention rate is that wallet's decision.
The 30 days after the final emission is the most important window. Track daily TVL, identify which segments stayed, calculate your true cost per retained dollar.
Step 7: Optimize Mid-Campaign
Attracting TVL faster than projected? Reduce emissions. Underperforming? Increase rates only after diagnosing why. If 60%+ of TVL is from known mercenary addresses, introduce curation mid-campaign.
For vaults, monitor utilization. If deposits grow but borrowing doesn't, subsidize borrowers to create the organic yield that sustains supply-side retention.
Run parallel variations across venues. Test different schedules, lock mechanics, and curation levels. Sometimes the right optimization is to end early: if retention signals are extremely poor (80%+ mercenary, cost per TVL 3-5x benchmark), cut losses and redesign.
Keeping Capital After Incentives End
If base APR without incentives is 2% and competitors offer 5%, no campaign design retains capital. This is the single most important variable.
Use a graduated wind-down: step from full emissions to 75% / 50% / 25% over 4-6 weeks beyond the core campaign. Publish the schedule at least two weeks in advance.
Loyalty mechanisms: a retroactive bonus (10-15% of campaign earnings) for LPs who maintain positions 30 days post-campaign, priority access in future campaigns, and governance power for long-term depositors.
Protocol-owned liquidity (POL) is the structural alternative: acquire your own position through treasury deployments or bond mechanisms. Many protocols are moving toward a hybrid model: campaign-bootstrap then gradual transition to POL.
How the Design Differs by Venue
Design Lever | DEX Pools | Lending / Yield Vaults | Pre-Deposit Campaigns |
What LPs deposit | Token pair (ETH/USDC) | Single asset (USDC, ETH) | Single asset, before launch |
IL exposure | Yes (volatile); minimal (stable) | None | None pre-launch; varies post |
Typical LP profile | Mixed: retail, aggregators, institutional | Conservative, longer-horizon, institutional-heavy | High-conviction, relationship-driven |
Retention behavior | Highest mercenary risk | 30-50% loss at 30d; stickier by default | 50-65% at 90d if product delivers |
Emission model | Aggressive taper | Gentle taper + curator fees | Front-loaded pre-launch, taper post-launch |
Distribution | Merkl (self-serve) or Turtle Streams (managed + sourcing) | Platform-native, Merkl, or Turtle Streams | Turtle (full-stack) or protocol-native |
Key risk | Mercenary capital + IL | Slow formation; curator misalignment | Product delays; trust risk |
Scenario Comparisons
Stablecoin vault on Ethereum. Morpho vault, 12-week gentle taper (120/100/80/60%), $50K minimum, 60-day lock bonus tier, borrower subsidies weeks 5-12. Cost: $0.20-$0.40/retained dollar. Retention: 40-50%. Risk: if borrowing demand doesn't materialize, organic yield stays near zero.
Volatile token pool on L2. GOV/ETH on Arbitrum, 16-week aggressive taper (150/100/60/30%), 60-day vesting on 50% of rewards, LP scoring. Cost: $0.55-$1.00/retained dollar. Retention: 20-30%. Risk: 50% token price drop halves effective incentive value.
Pre-deposit for new product. Base launch targeting $20M, 3-week pre-deposit at 200% boosted rate, 12-week post-launch taper. Cost: $0.15-$0.35/retained dollar. Retention: 50-65%. Risk: product delays or ambiguous conversion terms.
Stablecoin vaults cost least and retain best. Volatile pools cost most and retain worst. Pre-deposits outperform both if the product delivers.
Risks
Regulatory: incentive tokens with governance rights or revenue claims may trigger securities classification (SEC guidance, MiCA). Legal review before launch, not after. MEV: bots sandwich LP deposits, extract value from reward claiming, and game emission thresholds. Offchain computation (Merkl, Turtle Streams) reduces the attack surface vs. fully on-chain distribution.
Common Questions
What retention rate should a liquidity mining campaign target? Above 50% at 30 days, above 35% at 60 days, above 25% at 90 days. Lock-based structures show 40%+ improvement over standard emissions.
Should a protocol incentivize DEX pools, lending vaults, or both? Trading liquidity = pools. Deposit depth = vaults. Many run both. Vaults are cheaper per retained dollar and attract stickier capital.
How does the target asset change campaign design? Stablecoins: cheapest, stickiest, gentle taper. ETH/BTC: must clear staking yield bar (3-4% for ETH). Exotic tokens: mandatory vesting, aggressive curation, accept lower retention.
What is the cost per TVL for a DeFi liquidity mining campaign? $0.10-$0.30 per $1 TVL at peak. $0.25-$1.00 per $1 TVL retained at 90 days. Stablecoin vaults: $0.15-$0.40. Volatile DEX pools: $0.50-$1.00+.
What is the difference between Merkl and Turtle for incentive distribution? Merkl is self-serve distribution ($1.5B+ distributed, 60+ chains). Turtle is full-stack: LP sourcing, Streams distribution, and campaign advisory ($5.5B deployed, 430K+ wallets). Choose based on whether the gap is distribution mechanics or the full pipeline from sourcing through retention.
How long should a liquidity mining campaign run? Minimum 8-12 weeks. A 4-week campaign almost always produces mercenary outcomes. Best retention comes from 12-16 weeks with structured taper.
Internal Links
- Cost-per-TVL Benchmarks: What DeFi Protocols Actually Pay for Liquidity
- The Mercenary Capital Problem
- Case Study: Decibel Pre-Deposit Campaign
- Case Study: Katana Pre-Deposit Campaign
- Turtle Streams: Programmable Incentive Distribution
- How Turtle Sources and Qualifies LPs
- Q1 2026: Institutionalizing Liquidity Coordination
