An AMM route is a public function of public state: you can compute the expected output before you sign. An intent asks a set of solvers to compete to fill a desired outcome, and the winner chooses the path. The second design often produces a better price, which is why it has spread quickly. It also moves the point of trust from a deterministic contract to an auction with a small number of repeat players, and that change is easy to underprice.
Under direct AMM path routing the execution path is fixed in the transaction: a sequence of pools, a minimum output, a deadline. Anyone can simulate it, and failure is legible. Under intent-based solver auctions the user signs a statement of preference and an off-chain auction decides who fills it, against which liquidity, and in which order relative to other flow. That flexibility lets a solver net opposite orders internally, use private inventory, or batch across venues, all of which can beat the public route. The cost is that the user can no longer enumerate the execution before it happens.
An intent replaces a verifiable execution path with a competitive auction, and the quality of the result depends on how many solvers actually bid.
Part One: What is actually being compared
Strip the marketing away and the question in on-chain trade execution reduces to a threshold. Below a certain scale, direct AMM path routing is cheaper because it avoids a commitment; above it, intent-based solver auctions wins because the commitment is amortised. Everything else in the debate is about where that threshold sits and how confident you are that the workload will stay on one side of it.
A useful way to see the trade is to ask which cost survives a bad quarter. Under intent-based solver auctions the committed cost stays even when utilisation collapses, so the effective unit price rises exactly when demand is weakest. Under direct AMM path routing the cost falls with use, which is more comfortable operationally but removes the incentive to build spare capacity ahead of time.
Part Two: The arithmetic that decides it
Consider a swap large enough to move a pool by roughly 50 basis points. A direct route pays that impact plus a fee, and the trader can compute it. A solver with an offsetting order in the same block can internally match part of the flow, cutting realised impact to perhaps 10 to 20 basis points, and pass some of the saving back as price improvement. But the split of that improvement is set by the auction: with three active solvers the competitive share passed to the user is high; with one dominant solver the same mechanism leaves most of the surplus with the filler. The mechanism's performance is a function of the number of credible bidders, not of the smart contract.
One second-order effect deserves its own line. Once intent-based solver auctions is deployed, teams begin designing around the property it makes cheap, and that induced demand can consume the entire saving within a couple of planning cycles. The same thing happens in reverse with direct AMM path routing: a constraint that looks like a limitation often acts as a useful discipline, because it forces the design to stay inside a budget that everyone can see.
Part Three: Where the simple model breaks
This is where the analysis has to be careful. A quote is not an execution, and price improvement measured against a stale public reference can be manufactured by choosing a flattering reference. The honest benchmark is the execution a user could actually have obtained on the reference route at the same block, minus fees and minus any delay cost. Repeat solvers also see the order flow before it settles, which creates an information asymmetry that looks exactly like better pricing in aggregate statistics and may be partly rent extraction in individual cases. Disaggregated fill data is needed to separate the two.
A 50 basis point public impact can fall to 10 to 20 basis points when opposite flow is netted, but who keeps that difference is decided by bidder concentration, not by the contract.
Part Four: The cost that arrives later
There is an operational bill as well. Intent systems need an auction service, a reputation or bonding mechanism for fillers, a dispute path for unfilled or mis-filled orders, and monitoring for liveness. Each of those is a component that can fail off-chain, and the user's recourse is usually slower than the on-chain revert they would get from a direct route. Concentration is the long-run risk: if filling requires inventory and low-latency infrastructure, the set of competitive solvers naturally shrinks, and the auction gradually stops being an auction.
The adoption curve is likely to be S-shaped and gated by tooling, not by the core technology. The physics or the algorithm is already good enough; what is missing is the boring layer of monitoring, migration and default configurations that lets an ordinary team choose intent-based solver auctions without hiring a specialist. That layer is usually what separates a conference result from a procurement line.
Part Five: What this changes outside the engineering team
For an ordinary user the visible difference is a cleaner interface: one signature, one outcome, no gas-token juggling. That is a real gain in adoption, and it is why wallets keep adding intents. But it also means the fee is now embedded in the quoted price rather than itemised, so the cost of the service becomes hard to compare across providers. A market where price is easy to see and quality is hard to see tends to compete on the former, which is precisely the wrong variable when the product is execution quality.
A serious evaluation reports realised slippage against the counterfactual route in the same block, the number of active solvers per auction, and the distribution rather than the mean of price improvement. Those three numbers say more than any advertised average.
A market sizing sanity check
If even a third of retail on-chain swap volume migrates to intent-style execution by 2030, the infrastructure layer of auctions, filler bonding, monitoring and settlement could plausibly support $0.8B to $2.5B in annual revenue and associated service value, depending on volume growth and on how much of the spread is retained by fillers. The wide band reflects one assumption: whether order flow stays fragmented across many wallets, which keeps auctions competitive, or consolidates behind a few aggregators, which lets a small set of solvers price their own reference.
All figures here are an author's synthesis of published ranges and a stated model, not a measurement of a shipped system. The point of the arithmetic is to locate the crossover and expose which assumption drives it; a vendor quote will differ, and the direction of that difference is more informative than the final number.
The comparison that matters
The comparison that matters is between a deterministic path with legible failure and an auction with better expected price but a new counterparty. Intents win when solver competition is real and measurement is honest; direct routing wins when the trade is small, the venue is deep, or the cost of an opaque failure is high.
It is worth stating what would reverse this conclusion. If the cost of the scarce input falls sharply, or if a standard emerges that removes the integration overhead, the balance tips toward intent-based solver auctions far faster than a linear extrapolation suggests. Conversely, a single reliability incident in direct AMM path routing can move the decision back by years, because trust is accumulated slowly and spent all at once.
Measure execution against the counterfactual route in the same block and report the distribution of outcomes, because a flattering average can hide systematic rent extraction.
SOL: HCqPmPJM3NVtS6YrmVaTZYnuP8XZ2PFgQZKT94VK1LkS
Creating this takes real effort — thank you for the support.
Top comments (0)