TVL is the single most reported metric in DeFi. Every protocol dashboard leads with it. Every aggregate site ranks protocols by it. When someone wants to describe how much liquidity exists on-chain, they reach for TVL. This is not unreasonable: TVL is easy to compute, easy to understand, and broadly available. As a rough indicator of protocol scale, it has served its purpose in the early years of DeFi where the primary question was "which protocol is credible and has enough capital to be worth using."
The problem is that DeFi has moved past the point where "enough capital to be worth using" is the relevant threshold for most serious users. The question is now "how much will this specific trade cost me," and TVL does not answer that question. For anyone making execution decisions at meaningful notional sizes, TVL is close to useless as a pre-trade input. Depth at price impact thresholds is the metric that matters, and the two diverge significantly in practice.
The divergence between TVL and depth in concentrated liquidity pools
The divergence between TVL and execution-relevant depth is largest in concentrated liquidity pools, which have become the dominant pool type for high-volume pairs on most major DEXes since Uniswap v3 was launched. The design of concentrated liquidity allows LPs to specify price ranges for their positions: they provide liquidity only within a band around the current price rather than across all possible prices.
This design is capital-efficient in the sense that LPs who correctly forecast the trading range earn more fees per dollar of capital than they would in a constant-product pool. But the implication for TVL-as-liquidity-proxy is significant: the TVL of a concentrated liquidity pool includes all LP capital, including positions far outside the current price range that contribute zero depth at current prices. A pool with $40M TVL might have LPs with $20M in positions priced between 0.80 and 0.90 USD per ATOM when ATOM is at $10, contributing nothing to current execution quality.
The ratio of depth-contributing to total TVL varies across pools and shifts over time as price moves and LPs reposition. In practice, for volatile assets where LP positions tend to be spread across wider ranges, depth-contributing TVL as a fraction of total TVL can be as low as 20-30% in some pools. A pool that looks three times larger than a competitor by TVL may have shallower actual depth if the competitor's LPs are more tightly concentrated near the current price.
A concrete illustration: two pools, same TVL, different depth
Consider two hypothetical pools for the same volatile token pair. Pool A has $20M TVL with liquidity concentrated in a tight range of plus or minus 2% from the current price. Pool B has $20M TVL with liquidity spread across a 20% range. Both have the same TVL; any TVL-based ranking would treat them identically.
For a $250k swap on the pair, Pool A has materially more capital available within the 1% price impact range. The slippage calculation from the concentrated position means more dollars are absorbing the price impact. Pool B spreads the same dollars across a much wider range, so fewer dollars sit within the 1% band and the swap incurs higher slippage. The difference in execution cost between these two pools for the same transaction size could easily be 0.5-1.5 percentage points, which is material at $250k notional.
This illustration uses hypothetical numbers because we are making a conceptual argument, not publishing measured data from specific pools. The point is that the structural difference in depth at a given TVL level is real and known, not speculative. Anyone who has done the depth calculations for a range of pools will have seen this pattern repeatedly.
How pool type determines the TVL-to-depth relationship
The relationship between TVL and depth is not purely a function of LP positioning. It is also determined by pool curve type, because different curve designs convert capital into depth at different rates at the current price.
For constant-product pools (x*y=k), depth at 1% price impact scales roughly proportionally with TVL. If you double the TVL in a constant-product pool, you approximately double the depth at any given price impact threshold. This linear relationship is why TVL was a reasonable depth proxy in the early DeFi era when most pools were constant-product.
For stableswap curves (Curve Finance's amplified invariant), depth near the peg is much higher per unit of TVL than a constant-product pool with the same capital. The amplification factor effectively concentrates liquidity near the peg price, providing very deep execution within a narrow band at low cost. The tradeoff is that depth drops off sharply outside the amplified range. For stablecoin pairs, stableswap depth is typically 5-20 times higher than an equivalent TVL constant-product pool near the peg. TVL is a particularly poor depth proxy for stableswap pools when you are looking at slippage near the peg versus away from it.
For concentrated liquidity pools, as discussed, depth per TVL depends on how LP positions are distributed. There is no single relationship that holds across pools.
The practical case for depth-first analysis in treasury operations
Treasury teams managing cross-chain allocations in DeFi protocols face this question in concrete form when planning a rebalance. If the question is "can we move $500k from ATOM to USDC on-chain without materially affecting the market," looking at TVL rankings across Osmosis DEX pools and cross-chain alternatives will not answer it. The TVL of the ATOM-USDC pool might be $15M, but whether $500k can be moved at under 1% slippage depends on how much of that $15M is concentrated within the 1% price band at the moment of execution.
Answering this question requires current depth data at the target execution size, not TVL. For a transaction of this size, the relevant inputs are: depth at 1% impact for the best available pool, whether multi-hop routes with split execution across pools provide better aggregate depth, and how depth has varied over the recent period to assess execution timing risk.
This is not a question that requires sophisticated financial modeling. It requires having the right data and applying the correct metric. TVL rankings are the wrong input not because they are inaccurate, but because they answer a different question than the one the treasury team is asking.
What this means for how to evaluate protocol liquidity
We are not arguing that TVL is a meaningless metric. For questions about protocol scale, competitive positioning, or LP incentive attractiveness, TVL is the appropriate measure. The problem is specifically when TVL is used as a proxy for execution quality in contexts where depth is what matters.
For protocol governance decisions about liquidity incentives, TVL is relevant because it reflects LP commitment. For a treasury team's execution planning, TVL is largely irrelevant and depth at execution size is the required input. Using the right metric for the right question is the core principle. The fact that TVL is widely available and depth is harder to compute does not make TVL the right answer to execution quality questions; it makes depth data a gap worth filling.