Total value locked, or TVL, measures the value of assets counted inside a decentralised-finance protocol under a particular methodology. It is useful for understanding the scale of deposited capital, but it is not a profit figure, a safety certificate or a direct measure of how many people use an application.
The most important question is not simply whether TVL has risen. It is what is being counted, how it is valued, and whether the change came from new assets entering the protocol or from existing assets becoming more expensive.
Start with quantities and prices
A simplified TVL calculation adds up the quantity of each included asset multiplied by its price. If a hypothetical contract holds 1,000 ETH and 500,000 dollar-denominated stablecoin units, the dollar total depends on the ETH price and the price assigned to the stablecoin. This example is illustrative, not a quotation of any live protocol’s balances.
At an assumed ETH price of $2,000 and a stablecoin price of $1, that contract would show $2.5 million in assets. If ETH rose to $2,400 with both quantities unchanged, the total would become $2.9 million. TVL would rise by 16% even though no one deposited another token.
Reverse the price move and the same logic works in the other direction. A falling dollar TVL does not necessarily mean users withdrew. Separating quantities from valuation is essential before describing a chart as evidence of confidence, growth or flight.
Stock is not flow
TVL is a stock measured at a point in time. Deposits and withdrawals are flows over a period. A price-adjusted flow calculation asks how balances changed and values that change separately, rather than treating every movement in the dollar total as money entering or leaving.
For example, a pool could receive new ETH while the ETH price falls. Its token balance might increase even as its dollar TVL decreases. Another pool could lose deposits during a price rally yet show a higher dollar total. The net result alone does not identify either underlying path.
DefiLlama’s data definitions distinguish TVL from its USD inflows metric for this reason. Its treatment is a methodology to inspect, not a universal rule that every dashboard or protocol necessarily follows.
Even a flow estimate needs context. A token migration, contract replacement or change in classification can alter reported balances without representing a new customer allocating fresh capital. Before interpreting a sharp discontinuity, compare the methodology and the actual contracts included on both sides of it.
Different activities put assets to different uses
In a lending application, deposited assets may support loans. In an exchange pool, reserves support swaps. Other contracts hold collateral or assets allocated to strategies. The same dollar total can therefore describe very different economic functions and risk profiles.
Consider two hypothetical protocols, each showing $100 million. One might have substantial borrower demand and a diversified asset mix. Another might hold a concentrated position that earns little recurring revenue. Equal TVL does not make their businesses, liquidity or risks equivalent.
The distinction also applies within one platform. Assets in an idle contract and assets serving active users can produce the same headline valuation, depending on what the dashboard includes. Understanding the contract’s purpose is more informative than assuming every counted dollar is performing the same job.
Receipt tokens make aggregation difficult
DeFi applications can issue a token representing a claim on deposited assets. If that receipt token is then placed into another application, both applications may display value associated with the same underlying capital.
Imagine a hypothetical $1,000 deposit that produces a receipt. The owner places the receipt into a second protocol as collateral. Adding $1,000 from each application without considering the relationship would describe $2,000 of combined positions, even though the original external deposit was $1,000.
Neither view is automatically meaningless. The second application really has a position and an exposure. But a measure of gross protocol activity is not the same as a measure of unique external capital. An aggregate must explain which question it is answering.
DefiLlama flags certain receipt-token positions as double counted and offers a corresponding dashboard toggle. Readers comparing totals should use consistent settings. A chain total and a sum of individual protocol pages can differ because they apply different inclusion boundaries.
Borrowing is not another independent deposit
A lending system introduces a related accounting problem. A user can deposit an asset, borrow against it and deploy the borrowed asset elsewhere. Counting each step as wholly new external capital would overstate how much independent funding supports the combined system.
This does not mean borrowing is irrelevant. Outstanding loans, available liquidity and collateral are all economically important. It means they should be reported as the appropriate measures rather than folded into one number without explanation.
For a lender, utilisation can help explain the relationship between borrowed amounts and supplied funds. For an analyst examining an ecosystem, interconnected positions can reveal exposures that a simple aggregate hides. TVL is a starting point for those questions, not their answer.
TVL and trading liquidity are different
A large deposited balance does not guarantee that a trader can execute a large order near the displayed price. The relevant assets may not form the required trading pair, and the capital may not be positioned where the trade occurs.
Uniswap’s concentrated-liquidity documentation illustrates why placement matters: providers choose price ranges for their liquidity, and positions outside the active range do not support swaps in the same way as in-range positions. A headline total conceals that distribution.
Our guide to liquidity, spreads and slippage explains the broader execution distinction. An asset stock and the capacity to absorb a specific trade are related, but neither can be substituted mechanically for the other.
Withdrawal access is another separate question. A balance can be economically attributable to users while its movement is subject to contract rules, collateral obligations or other restrictions. The word “locked” should not be read as promising a single standard withdrawal timetable.
Do not confuse TVL with market capitalisation
Market capitalisation generally relates a token’s price to its circulating supply. TVL relates included assets to a protocol or aggregation boundary. A project’s governance token and the assets deposited by users are not the same thing.
A ratio between token market value and TVL may be used as a comparison, but it is not a conventional earnings multiple. Deposited customer assets do not automatically belong to token holders, and a large deposit base does not automatically create a distribution of income to them.
Suppose two hypothetical tokens have equal market capitalisations and their associated protocols have different TVL figures. The lower ratio alone does not prove that one token is cheap. Fee capture, expenses, incentives, token rights and risk can differ materially. The denominator cannot resolve those differences.
Revenue and incentives need their own examination
A protocol can attract deposits with token rewards. That may increase its asset base without demonstrating durable demand once rewards decline. Conversely, an application can process significant activity using a relatively small amount of capital efficiently.
To assess economic activity, examine appropriate measures alongside TVL: fees generated, revenue retained, borrowing demand, transaction activity and the cost of incentives. Do not assume any one metric replaces the others.
An illustrative comparison makes the limitation clear. A protocol paying $2 million in incentives to generate $1 million in recurring receipts faces a different economic question from one generating the same receipts without that subsidy. Their TVL charts might look similar even though the sources of activity differ.
The exact definitions of fees and revenue also vary. A payment to liquidity providers is not necessarily income retained by the protocol or distributed to governance-token owners. Comparisons should preserve those boundaries rather than presenting every fee as profit.
Large balances do not establish safety
TVL shows that value is counted inside a defined system. It does not prove that the contracts are free of vulnerabilities, that collateral prices are reliable or that users can exit under stressed conditions.
Concentration deserves particular attention. A large total supported mainly by one volatile asset or a small number of positions can behave differently from a diversified balance. The total alone does not show how a price shock would propagate through borrowing and collateral arrangements.
Stablecoin-heavy TVL is not risk free either. Valuing every unit at one dollar can hide a change in the asset’s actual price or redemption conditions. Our stablecoin peg explainer describes why a target price and reliable redemption are distinct concepts.
No TVL threshold removes the need to examine permissions, dependencies, withdrawal rules and the nature of the assets. A bigger number can mean more capital exposed to a system; it does not certify that system.
How to make a useful comparison
Begin by choosing the same measurement time, currency and inclusion settings. Identify whether a chart describes one application, one chain or an aggregate, and check whether staking, borrowed assets and receipt-token positions are treated consistently.
Then separate price effects from balance changes. Inspect the asset mix and look for events that could explain a discontinuity. Compare related activity measures only after checking their definitions and periods.
Finally, state the conclusion narrowly. “The dollar value of counted assets increased” is supported by a comparable TVL series. “New users deposited that amount,” “the token is undervalued” and “the protocol is safe” require additional evidence.
TVL earns its place in DeFi analysis because it measures an important dimension of scale. Its usefulness improves when the boundaries are explicit. Read it as a defined asset measurement, pair it with flows and activity, and resist turning a single headline number into a verdict on the entire protocol.