Key Takeaways
TVL tracks capital size, not real usage, so protocols can look successful while activity is mostly passive or incentive-driven
Active users show participation but are easily inflated by bots, airdrops, and low-value transactions that don’t reflect real demand
Real growth appears only when both metrics align with volume, fees, and retention signals that prove users return and capital is actually used
Why DeFi Teams Obsess Over TVL
Total value locked became the default measure of DeFi success almost by accident. In the early years of the ecosystem, TVL was a reasonable proxy: if capital was being deposited into a protocol, something useful must be happening. Aggregators like DeFi Pulse and DefiLlama made TVL easy to track and compare across protocols, and the number stuck as the headline metric for the industry.
The problem is that TVL measures something real but narrow. It tells you how much capital has been deposited. It does not tell you whether that capital is being used, whether the users behind the deposits are genuinely engaged, or whether the protocol would survive the moment incentives change. Teams that optimise for TVL often find themselves with impressive dashboards and hollow underlying growth.
Active users offer a different angle on the same question. But user counts have their own distortions: bots, airdrop hunters, and incentive farmers can inflate wallet numbers without representing any durable engagement. Neither metric is sufficient alone, which is why protocols that navigate growth well tend to use them together, in combination with economic activity signals, rather than treating either as a standalone verdict.
This article unpacks what each metric actually measures, where each misleads, and how to build a combined signal that is harder to fake and more predictive of real protocol health. For the broader framework this sits within, see the onchain growth guide.
What TVL Actually Measures (and What It Does Not)
TVL is the total dollar value of assets deposited into a protocol's smart contracts at a given point in time. That is the entire definition. It is a balance sheet snapshot, not a measure of activity, engagement, or product-market fit.
What TVL does measure well:
Capital availability for the protocol's core function. A lending protocol with more TVL can support larger loans. A DEX with more TVL has deeper liquidity and lower slippage. In this narrow sense, TVL reflects protocol capacity.
Relative scale within a category. When comparing two lending protocols of similar age and design, the one with higher TVL has attracted more capital commitment, which is a real signal even if imperfect.
Market confidence in protocol security. Capital that has sat in a contract for 12 months without being exploited reflects a degree of market trust that is not nothing.
What TVL does not measure:
Whether deposited capital is being actively used. A pool can hold $200m in TVL with near-zero trading volume. The capital is parked, not working.
The number of participants. A small number of large wallets can account for the majority of TVL. A protocol with $500m TVL from 12 whales is structurally different from one with $50m TVL spread across 8,000 active depositors.
Protocol stickiness. TVL driven by token incentives will leave when those incentives change. TVL driven by genuine yield demand tends to be more durable. The number looks identical in both cases.
User intent. A depositor who is yield farming has entirely different intent from one who is actively using the protocol as a financial tool. TVL does not distinguish between them.
Key Limitation: TVL is a snapshot of capital, not a measure of usage. A protocol can have rising TVL and falling engagement simultaneously, and the TVL number will not tell you this is happening.
What Active Users Measure (and What They Do Not)
Active user counts in DeFi typically refer to unique wallet addresses that have executed at least one transaction within a defined period, usually 7 or 30 days. This is a meaningful improvement over TVL in one respect: it requires action, not just a balance. A wallet that appears in an active user count has done something, even if that something was minimal.
What active users measure well:
Transaction-level engagement. Active user counts confirm that people are interacting with the protocol, not just depositing and forgetting.
Breadth of participation. Unlike TVL, which can be concentrated in a few wallets, active user counts spread across many participants signal genuine product reach.
Activation and retention trends. When tracked as cohorts over time, active user data reveals whether the protocol is converting new wallets into repeat users, which is the core question of onchain retention.
What active users do not measure:
Quality of activity. A wallet that executes one $5 swap per month to qualify as an active user is categorically different from one executing daily lending position management. Both count identically in a 30-day active wallet metric.
Economic contribution. High active user counts without proportional volume or fee generation indicate low-value activity. This is common in protocols with airdrop expectations.
Authenticity. Bot activity, Sybil behaviour, and wash trading all generate wallet activity. A protocol with 50,000 monthly active wallets may have a significant fraction that are not human users at all.
Retention quality. A user who transacted once six months ago and has not returned is not the same as one who transacts weekly. Raw active user counts often collapse this distinction. The user lifecycle analysis guide covers how to segment users by engagement depth rather than treating all activity equally.
Key Limitation: Active user counts tell you something happened, not whether it mattered. Quantity of wallets is not a substitute for quality of engagement.
Scenarios Where TVL Is Misleading
Incentive-Driven Capital Inflation
A protocol launches a liquidity mining campaign with high token emissions. TVL rises sharply as mercenary capital flows in to capture rewards. The team celebrates the milestone. Three months later, the incentive budget runs low and the emissions schedule changes. When the emissions schedule changes, TVL drops sharply in the weeks that follow. The capital was never sticky. The users were never engaged. TVL was a measure of the incentive programme, not of the protocol. This pattern is well-documented in the analysis of how DeFi incentive programs shape onchain growth.
Whale Concentration
A lending protocol shows significant TVL. Closer inspection reveals that a small number of wallets account for the large majority of deposits, with the remainder spread across many smaller depositors. The underlying structure implies fragility. If any of the five large wallets withdraws, TVL collapses. The protocol is not broad-based. It is dependent on a small number of actors whose continued presence is driven by factors outside the team's control.
Market Price Inflation
TVL is denominated in dollars. When the price of deposited assets rises, TVL rises with it, even if nothing has changed about the protocol's actual usage or user base. A protocol holding 10,000 ETH has very different TVL numbers at $2,000 per ETH versus $4,000 per ETH, despite being identical in every other respect. Teams that do not control for asset price movements in their TVL analysis are measuring market conditions as much as protocol growth.
Circular Liquidity
Some DeFi ecosystems allow capital to be used as collateral to borrow more capital, which is then deposited elsewhere in the same ecosystem. The same underlying dollars can be counted multiple times across different protocol TVL figures. Circular or recursive liquidity creates TVL inflation that does not correspond to real capital or real users.
Scenarios Where Active Users Are Misleading
Airdrop and Farming Behaviour
Protocols that have signalled or rumoured an upcoming airdrop attract wallets specifically created to accumulate eligibility. These wallets execute the minimum transaction required to qualify, then go dormant until the airdrop. They register as active users during the eligibility window. They disappear immediately after the airdrop. User counts during the farming period look like growth. Cohort retention after the event reveals the truth. The onchain user segmentation guide explains how to identify airdrop farming behaviour in your wallet data before it distorts your metrics.
Bot and Sybil Activity
Automated trading, arbitrage bots, and Sybil attack wallets all generate transaction activity that counts as active user events. A protocol with significant arbitrage activity may show high active wallet counts that are primarily bots executing pricing corrections. This is not necessarily bad for the protocol economically, but it is not evidence of user growth. Treating bot-generated activity as human engagement leads to badly calibrated product and growth decisions.
Low-Value Transaction Inflation
Some protocols, particularly those with very low gas fees, see high active user counts driven by tiny transactions that carry no economic significance. A user executing $1 swaps repeatedly for reasons unrelated to genuine product use still generates active wallet events. Volume-per-active-wallet is a useful corrective here: protocols with high user counts but very low average transaction size should investigate whether the activity reflects genuine engagement.
Single-Action Activation Without Retention
A successful marketing campaign or integration drives a large cohort of new wallets to the protocol. Each wallet executes one transaction. Active user counts spike. But if none of those wallets return in the following 30 days, the spike is not growth, it is a one-time event. The activation guide covers the distinction between wallet connects, first transactions, and genuine activation, and why collapsing all three into a single user count obscures what is actually happening.
How to Combine TVL and User Metrics Into a Growth Signal
No single metric is sufficient. The protocols that read their growth most accurately use TVL and active user data together, alongside economic activity signals, to build a composite picture. Here is the framework, drawing on the DeFi KPIs guide for the full metric reference.
Signal | What to Track | What It Tells You That Neither TVL Nor Active Users Can Alone |
TVL per Active Wallet | Total TVL divided by unique active wallets in the same period | Whether capital is concentrated or distributed. Rising TVL per wallet alongside flat active wallets signals increasing whale concentration. Falling TVL per wallet alongside rising active wallets signals real user growth with smaller average deposits. |
Fee Revenue per Active Wallet | Protocol fee revenue divided by active wallet count | Whether active users are generating real economic value. High user counts with low fee revenue indicates shallow activity. Low user counts with high fee revenue indicates high-value users and potential for growth. |
Volume to TVL Ratio | Trading or transaction volume as a percentage of TVL | Whether deposited capital is being actively used. A high ratio means capital is working hard. A low ratio means capital is parked and the protocol is a storage vehicle, not a usage venue. |
D30 Return Rate of Active Wallets | Percentage of active wallets in one month that transact again in the next | Whether active users are sticky or one-time. This is the single clearest signal of whether user growth is compounding or leaking. TVL cannot tell you this. Raw user counts cannot tell you this. |
Organic TVL Growth Rate | TVL growth during periods with no active incentive programmes | Whether capital is genuinely attracted to the protocol or whether it is rented by incentives. Organic TVL growth is rarer and more valuable than incentive-driven TVL growth. |
Framework: Use TVL to measure capital capacity. Use active users to measure participation breadth. Use volume-to-TVL ratio, fee revenue per wallet, and D30 return rates to measure whether the combination is generating real, durable growth.
Stage-Based Metric Prioritisation
The right metric emphasis changes depending on where a protocol sits in its lifecycle. Applying the same metric framework to an early-stage protocol and a mature one produces different insights and different error modes.
Early-Stage Protocols
At this stage, TVL numbers are small and easily distorted by one or two large depositors. Active user counts are also small and prone to noise. Neither is statistically reliable yet. The most informative signal at early stage is the behaviour of a small number of actual users: are they returning? Are they increasing their transaction frequency? Are they adding to their positions over time?
Early-stage teams should prioritise repeat transaction rate over raw user count, and depth of engagement over TVL headline. A protocol with 200 wallets where 60% return monthly is in better shape than one with 2,000 wallets where 3% return. Focus on activation quality and early retention signals rather than headline numbers that will be too small to be meaningful to anyone outside the team.
Scaling Protocols
At scale, both TVL and active user counts become more reliable as noise-to-signal ratios improve. The priority at this stage is correlation: are TVL and active users growing together? If TVL is growing faster than active wallets, concentration risk is increasing. If active wallets are growing faster than TVL, the protocol may be attracting low-capital users without building the liquidity depth to serve them well.
Scaling protocols should add volume-to-TVL ratio and fee revenue per wallet to their core dashboard. These two metrics reveal whether growth is translating into economic activity or whether the protocol is accumulating balance sheet without usage. The DeFi growth funnel guide covers how to map each stage of user progression so scaling teams can identify where the funnel is leaking.
Mature Protocols
Mature protocols have enough history to track all metrics over meaningful time periods. At this stage, the most valuable signals are trend-based rather than level-based: is the volume-to-TVL ratio improving or deteriorating? Is the D30 wallet return rate stable or declining? Is organic TVL growing even when incentive programmes are paused?
Mature protocols also face a specific risk: metric complacency. A protocol with high absolute TVL and large active user numbers can be declining in relative terms without the headline numbers making this visible. Tracking growth rates rather than levels, and comparing against relevant benchmarks, is how mature teams stay honest about their trajectory. The onchain growth loops guide covers how mature protocols build the compounding mechanisms that sustain growth without constant acquisition spend.
What Investors vs Operators Care About
TVL and active user data serve different audiences with different questions. Understanding who is asking and why determines which metrics to surface and how to frame them.
Audience | What They Use TVL For | What They Use Active Users For |
Investors | TVL is a proxy for protocol importance and market position. Investors use it to compare protocols within a category and assess whether a protocol has achieved meaningful scale. They are aware of its limitations but use it as a starting filter before going deeper. | Investors want active user trends more than levels. A protocol growing active wallets month-over-month at a healthy rate tells a better story than one with a large but flat user base. Retention of active wallets is particularly important: it signals product stickiness that TVL cannot show. |
Protocol Operators | Operators use TVL to make capacity decisions: whether to expand supported assets, adjust fee tiers, or deploy additional liquidity programmes. TVL concentration analysis (whale vs broad-based) is particularly important for operators managing fragility risk. | Operators care about active user quality more than count. The breakdown by user type, transaction frequency, and acquisition source tells them where to invest in product improvement and where growth spend is producing durable results versus temporary spikes. |
Growth Teams | Growth teams track TVL as a downstream outcome of their work, not as a primary optimisation target. They care whether campaign spend is generating TVL that stays or TVL that leaves when rewards change. | Growth teams use active user data to optimise acquisition and retention spend. Activation rates by acquisition source, D30 return rates by cohort, and segment-level engagement patterns are the operational metrics that drive day-to-day decisions. |
External Analysts | External analysts and aggregators use TVL as a category benchmark and ecosystem health signal. They are less likely to dig into wallet concentration or incentive structure, so headline TVL carries more weight in external analysis than it deserves. | External analysts use active user data to assess community engagement and product adoption. They typically do not adjust for bot activity or airdrop farming, so active user counts in external reports often overstate genuine engagement. |
The Bottom Line
TVL and active users are both real metrics measuring real things. The error is treating either as sufficient. TVL tells you capital is present. Active users tell you wallets are transacting. Neither tells you whether the protocol is genuinely useful, economically healthy, or building the compounding growth that survives incentive changes and market cycles.
The protocols that grow sustainably track both, combine them with economic activity signals, and adjust their metric emphasis by lifecycle stage. They know which numbers are for external reporting, which are for internal diagnosis, and which are the leading indicators of problems that will not show up in the headline figures until much later. For the full picture of how to structure your growth measurement, the DeFi marketing analytics guide and the DeFi KPIs reference cover the complete measurement stack.
Track TVL, Users, and Economic Activity Together with Formo
The composite signals that actually predict DeFi success, volume-to-TVL ratio, fee revenue per wallet, D30 return rates, organic versus incentive-driven capital, require connecting onchain data with wallet-level analytics. Most analytics tools give you one or the other. Formo is the analytics and growth platform built for DeFi apps that combines both in a single view.
For DeFi teams moving beyond TVL and raw user counts, Formo provides:
Wallet-level cohort analysis to separate genuine retained users from mercenary capital and airdrop farmers — via retention analytics
Acquisition source attribution so you can see which channels produce wallets that deposit, transact, and return — via onchain attribution
Wallet profiles connecting onchain history, DeFi activity, and token holdings so you can segment by user quality, not just user count — see wallet profiles
Protocol-level economic activity tracking combining TVL, volume, fee revenue, and user data into a single growth dashboard — powered by Formo's analytics
Ask AI to surface the relationships between your TVL, user, and economic metrics directly, without SQL or a data team
DeFi teams including Kyberswap and WalletConnect use Formo to drive growth onchain.
Explore the Onchain Growth Series
This article is part of Formo's onchain series, a collection of practical guides for DeFi founders and growth teams covering the full post-launch lifecycle. Each guide goes deep on a single growth challenge with frameworks you can apply directly to your protocol.
FAQs About TVL vs Active Users in DeFi
Is TVL actually a good measure of growth?
TVL is a weak growth metric on its own. TVL mainly shows how much capital is parked in a protocol, not how much it is being used. Large deposits from a few wallets can inflate TVL without real user growth. TVL can go up even when product adoption is flat.
Are active users a better metric than TVL?
Active users are a better signal of product usage, but they are still easy to misread. A protocol can have many active wallets doing tiny or meaningless actions. High user counts do not guarantee meaningful volume or revenue. Active users without economic activity can be vanity metrics.
Why does our TVL look great but usage feels dead?
This happens when a small number of wallets provide most of the capital. TVL can stay high even if no one is trading, borrowing, or rebalancing. Locked capital does not mean the product is sticky or useful. Usage metrics often tell a very different story than TVL.
Can active users be misleading too?
Yes, active users can be misleading when activity is driven by bots or low-value actions. Farming, airdrop hunting, and spam transactions inflate user counts. If most wallets never return, user growth is shallow. User numbers matter only when tied to repeat usage and volume.
Which metric should we care about at different stages?
Early-stage teams should care more about repeat users and usage patterns. Scaling protocols should watch both TVL and active users together. Mature protocols need to track TVL, users, and economic activity as a system. No single metric predicts success on its own.
