Here is the full breakdown at a glance:

The rest of this article explains why each row is different, what the practical consequences are if you get it wrong, and what to track instead.

DeFi protocols are not interchangeable products. A DEX competes on liquidity depth and price execution. A lending protocol competes on risk parameters and collateral flexibility. A vaults protocol competes on strategy diversity and APY consistency. Each model attracts different users, measures different success, and breaks down at different points in the user journey.

Yet most DeFi growth teams instrument their analytics the same way regardless of protocol type. They track wallet connections, measure TVL, and call retention a 30-day active user count. None of these metrics are wrong in isolation. All of them are incomplete in context.

This guide maps the specific analytics requirements for DEX, lending, and vaults protocols, covering activation events, retention signals, churn indicators, LP and depositor segmentation, revenue metrics, and attribution focus. The goal is to give growth teams a clear picture of what to measure, and why the answer differs by protocol type.

Why Protocol Type Changes Your Analytics Requirements

DEX users come to transact. Lending users come to open positions. Vaults users come to park capital. Each requires a different measurement approach.

The difference between protocol types determines what a user does in your product, how long a session lasts, what brings them back, and what constitutes value creation for the protocol.

• In a DEX, a user's core action is a swap or a liquidity provision. Both are short, discrete events. Users may interact with a DEX dozens of times in a week or not at all for months. The analytics question is frequency and volume, not depth of engagement in a single session.

• In a lending protocol, a user's core action is opening a position, whether as a borrower or a lender. Positions persist. A user who borrowed six months ago is still an active user if their position is open. The analytics question is position health and lifecycle, not session frequency.

• In a vaults protocol, a user's core action is selecting and depositing into a strategy. The user relationship is largely passive after initial deposit. Retention looks like a position that was never closed, or a user who rebalanced into a new strategy. What you are measuring is capital stickiness and strategy migration, not session activity.

These differences cascade through every analytics decision: what event you call an activation, what you call churn, which wallet segments matter most, and which attribution model reflects real acquisition.

Analytics Requirements by Protocol Type

DEX Analytics: What to Track and What to Ignore

The activation event

Wallet connection is not activation for a DEX. A user who connects their wallet and does not swap has not experienced your product. The activation event is the first completed swap above a minimum meaningful threshold, typically above gas cost, and ideally above a protocol-set floor that filters out test transactions.

Time to first swap is the metric that matters most in your onboarding funnel. If users connect wallets but do not swap within 24 hours, identify where in the flow they exit:

• Before they reach the swap interface: likely a landing page or onboarding flow problem

• Before they confirm the transaction: likely a UX or slippage settings problem

• After a failed transaction: likely a gas or wallet configuration problem

Retention and frequency

For DEXs, retention is trading frequency. A user who swapped once three months ago is not a retained user. The signals to track:

• Swaps per week per wallet

• Return visit rate within 7 days of first swap

• Average session value measured in swap volume

LP retention is a separate segment with different signals. LPs are retained as long as their position is open and not experiencing high impermanent loss relative to fees earned. LP churn looks like full position withdrawal. LP health analytics track the ratio of fees earned to impermanent loss exposure across your pools.

Revenue metric

Fee revenue per active wallet is the correct unit for DEX monetisation analytics. TVL in a DEX is a proxy for liquidity depth, not a revenue or growth indicator. A protocol with lower TVL but higher fee revenue per wallet is healthier than one with high TVL and thin volume.

Track fee revenue attribution by wallet cohort. Users acquired through different channels often generate materially different fee revenue. This is the signal that tells you which acquisition channels are worth investing in.

Wallet intelligence priorities

For DEX growth teams, the wallet segments that matter most are cross-protocol active traders and existing DEX users on competing protocols. A wallet with a history of active DEX usage on Uniswap or Curve is a high-probability converter. A wallet that has never executed a swap anywhere is a higher-friction acquisition.

Formo's wallet intelligence layer surfaces this data automatically. When a user connects their wallet, you can see their trading history, current positions, and protocol usage across chains without any additional instrumentation.

Lending Protocol Analytics: What to Track and What to Ignore

The activation event

For lending protocols, activation has two modes depending on user type:

• Borrowers: first borrow above a minimum threshold that signals real intent rather than a dust transaction.

• Lenders and depositors: first supply above a meaningful floor.

The distinction matters because borrowers and lenders have different retention curves, different churn risks, and different value to the protocol. Segment them from the first interaction.

Position health as a retention signal

A borrower who has not repaid or borrowed again is still an active user if their position is healthy. Traditional session-based retention metrics misread lending protocols entirely. A user who opened a position in January and has not logged in since is not churned if the position is open and the collateral ratio is healthy.

The retention metrics that matter for lending protocols:

• Active position count across your user base

• Collateral ratio distribution: what percentage of positions are at liquidation risk

• Percentage of positions at liquidation risk

These tell you the health of your retained user base, not just whether they visited recently. See DeFi onchain retention for how to build this view.

Churn definition

Churn for a lending protocol is position closure without re-entry within a defined window. A borrower who repaid and did not open a new position within 30 days is a churned user. A lender who withdrew and did not re-deposit is churned.

The analytics question is: 

• what precedes position closure? 

Track wallet behaviour in the 7 days before a position closes. 

• Are users withdrawing because of market conditions, liquidation risk, or competitive rates elsewhere? 

The answer informs retention interventions.

Revenue metric

Interest spread per wallet is the correct revenue unit for lending protocols. A borrower paying 8% APR while lenders receive 5% APR generates a 3% spread. Track this by wallet cohort to understand which user segments are most profitable, and which acquisition channels produce the highest-spread positions.

Vaults Protocol Analytics: What to Track and What to Ignore

The activation event

For vaults protocols, activation is a first deposit into a strategy above a meaningful threshold. The threshold should be set to filter out users testing the product without real capital commitment. An appropriate floor varies by protocol but is typically at least 10x the gas cost of entry.

Strategy selection depth is an early retention predictor. Users who evaluate multiple strategies before depositing tend to have higher retention rates than those who deposit into the top-listed option. Track strategy page views before first deposit as an engagement signal.

Capital stickiness as retention

For vaults protocols, a user whose capital is still deployed is retained — regardless of whether they have logged in.

For vaults protocols, retention is capital stickiness. A user who deposited six months ago and has not withdrawn is a retained user, regardless of login frequency. The signal to track is vault position age distribution across your user base. A protocol where most positions are under 30 days old has a capital stickiness problem.

Reinvestment behaviour is the strongest retained-user signal. A user who compounds rewards back into a vault, or who moves capital from one vault to another within your protocol, is a deeply retained user. Track this as an intra-protocol migration event. See cohort analysis for DeFi growth for how to track this by acquisition cohort.

Churn definition

Churn for a vaults protocol is full withdrawal without re-entry. Partial withdrawal is not churn. A user who reduces position size but maintains a deposit is still engaged. The analytics distinction matters because it changes your churn rate calculation significantly.

Track what happens after full withdrawal. 

• Do users come back within 30 days? 

• Within 90 days? 

• Do they migrate to a competitor protocol? 

If wallet intelligence data shows your churned users are active in a competing vaults protocol, that is a product signal, not just a marketing signal.

Revenue metric

For vaults protocols with management fees or performance fees, the correct revenue metric is fee revenue per deposited dollar per month. This normalises for position size and tells you which strategies and which user segments generate the most protocol revenue relative to capital committed. See how to measure LTV and CAC in DeFi for how to build this by cohort.

Wallet intelligence priorities

For vaults protocol growth teams, the wallet segments that matter most are wallets with existing stablecoin holdings, prior vaults protocol activity, and a history of compounding rather than withdrawing rewards. A wallet that has participated in vaults strategies on Yearn, Aura, or similar protocols is a high-probability converter because the behaviour pattern is already established.

• Risk-on wallets with concentrated exposure to volatile assets are more likely to engage with higher-APY, higher-risk strategies.

• Stable-preferring wallets with primarily stablecoin holdings are more likely to activate on conservative vault options. Segment on this dimension before targeting acquisition channels.

Attribution: Where Protocol Type Changes the Model

Attribution logic differs by protocol type because the conversion events and time horizons differ.

• DEX: For DEXs, first-touch attribution works well for measuring which channels introduced users who went on to become frequent traders. The conversion window is short. A user who clicks a campaign and swaps within 48 hours is a clear attribution signal. See  onchain attribution  for how to configure this.

• Lending: For lending protocols, last-touch attribution often misleads. A user might have encountered your protocol through content six months before opening a position. A  time-decay attribution model  that weights recent touchpoints but does not discard earlier ones produces a more accurate picture of which channels build long-term borrower and lender acquisition.

• Vaults: For vaults protocols, the attribution question is which channel drove strategy adoption, not just deposit. A user who deposited into the default vault from a referral link is a weaker attribution signal than a user who researched three strategies, read your documentation, and deposited into a specialised vault. Track strategy adoption depth as part of the attribution outcome, not just the initial deposit event.

How Kairos Swap Applied Protocol-Specific Analytics

Kairos Swap, a DEX on Base, was tracking wallet connections and TVL as their primary growth indicators. Both metrics were growing, but the team had no visibility into repeat trading behaviour or which acquisition channels were producing users who swapped more than once.

After switching to a DEX-specific analytics setup with Formo, the team reconfigured their activation event to first completed swap above a protocol-set threshold and began tracking fee revenue per active wallet by acquisition cohort. Within the first month, they identified that one acquisition channel was generating 3x the fee revenue per wallet compared to their highest-volume channel. They reallocated spend accordingly. Read the full Kairos Swap customer story for details on their setup and results.

How Formo Supports Protocol-Specific Analytics

The analytics requirements described in this guide are specific because DeFi protocol types are specific. A generic event model forces you to approximate every metric that matters. Formo is built around the assumption that protocol type determines what you measure, not the other way around.

• DEX teams: define your activation threshold, track trading frequency and fee revenue by wallet cohort, and surface cross-protocol DEX activity for incoming wallets so you know which users arrive with relevant trading history.

• Lending teams: capture position open, health change, and close events natively, without a custom data pipeline, and support time-decay attribution to give you an accurate picture of which channels build long-term borrower acquisition.

• Vaults teams: surface wallet risk profiles and historical strategy behaviour before a user deposits, enabling activation messaging that matches their existing vaults preferences.

Across all three protocol types, cohort analysis by acquisition channel lets you compare fee revenue, position value, and retention curves to identify which channels produce the highest-value users for your specific product, not just the highest volume.

Get started with Formo to configure analytics around your protocol type.

More in This Series

Evaluating how DeFi analytics requirements differ across use cases? Read the other articles in this series:

• Cohort Analysis for DeFi Growth: How to Use Cohort Data to Improve Protocol Retention

• DeFi Protocol Benchmarks: What Good Retention, Activation, and CAC Numbers Look Like

• Token Incentive Analytics: How to Measure Whether Your Rewards Program Is Working

• DeFi Liquidity Provider Analytics: How to Track and Understand LP Behaviour

Frequently Asked Questions

What is the most important analytics difference between a DEX and a lending protocol?

The core difference is in what constitutes an active user. For a DEX, activity is session-based: a user who has not swapped recently is not active. For a lending protocol, activity is position-based: a user with an open position is active regardless of login frequency. This changes how you measure retention, calculate churn, and design re-engagement campaigns.

Why does TVL mislead as a growth metric for DEX protocols?

TVL measures capital deposited, not capital active. A DEX with high TVL but low trading volume has liquidity that is not generating fee revenue. Fee revenue per active wallet and trading frequency are the metrics that reflect real protocol health. See the Formo article on TVL vs active users for a full breakdown.

How should a vaults protocol define churn?

Full withdrawal without re-entry within a defined window, typically 30 to 90 days depending on your average position duration. Partial withdrawals are not churn. A user who reduces position size but maintains a deposit is still engaged. Using full withdrawal as the churn trigger produces a more accurate retention picture than session-based definitions.

What attribution model works best for lending protocols?

Time-decay attribution works better than last-touch for lending protocols because the conversion window is long. Users often encounter a lending protocol through content or referral months before opening a position. A model that weights recent touchpoints but preserves earlier ones produces a more accurate picture of which channels build long-term borrower acquisition. See onchain attribution models explained for a full comparison.

Can Formo handle analytics across multiple DeFi protocol types in one dashboard?

Yes. Formo supports custom event definitions, which means you can configure activation events, retention signals, and revenue metrics to match your specific protocol type. Teams running multi-product DeFi applications can segment analytics by product type within a single workspace. See Formo Analytics for more.

What wallet intelligence data does Formo surface for DeFi growth teams?

Formo surfaces trading history, current positions, cross-protocol activity, wallet age, token holdings, and DeFi participation across EVM chains for every connected wallet. Growth teams can use this to segment users by protocol type experience, identify high-probability converters, and understand which wallet profiles are most likely to activate and retain on each product type. See wallet profiles for details.