Volume bot cost breakdown, every line

Most volume bot pricing pages show one number and call it the cost. There are six components, they behave differently, and only one of them is set by the vendor. This page lists all six, shows which are fixed and which are auction-priced, and works a complete campaign budget from measured block data rather than assumptions.

Volion Research Updated Jul 30, 2026 4 sections

The six things you pay for

A volume campaign costs: the vendor service fee, the Solana base fee of 5,000 lamports per signature, a priority fee you bid per compute unit, a Jito tip per bundle if you route that way, account rent when a swap creates a new account, and the fee on every transaction that fails. Price impact is a seventh cost that is not a fee but still reduces what your SOL buys.

Only the first row is negotiable between vendors. The rest belong to the network and to your configuration.
Component	Set by	Fixed or variable	Scales with
Service fee	Vendor	Fixed	Target volume
Base fee	Protocol	Fixed at 5,000 lamports	Signatures
Priority fee	You, bidding	Variable	Account contention
Jito tip	You, bidding	Variable	Bundle count and competition
Account rent	Protocol	Fixed per account	New wallets and token accounts
Failed transactions	Nobody	Variable	Contention and routing target
Price impact	Pool depth	Variable	Trade size vs liquidity

The reason vendors quote only the first row is that it is the only one they control, and it is usually the smallest. That framing is not exactly dishonest, but it produces budgets that run out early with predictable regularity.

The line item nobody quotes

Failed transactions are charged in full and produce no volume. Because failure rates on DEX programs run several times the network average, this line is frequently larger than the service fee. The correct way to handle it is to divide your target swap count by the measured success rate of your routing target, and budget from the resulting attempt count.

This is arithmetic rather than opinion. A 30% failure rate means you attempt 1.43 transactions for every one that lands, so your fee spend is 43% above a naive estimate. At 50% you attempt two for every one. Our measured failure rates per DEX program, with the sampling method written out, are on the measurement page.

Two things drive the rate. Account contention causes transactions to lose their lock and be dropped, which is worst on new bonding curve launches where snipers compete for identical accounts. Compute limit misconfiguration causes transactions to run out of execution budget mid-flight, which is an engineering failure rather than a market condition and should be near zero in a competent implementation.

When comparing vendors, the single most useful question is not what they charge but what their landed-swap cost is on your routing target, and whether they can show you the measurement rather than assert the number.

A complete campaign, costed from measured data

Take a target of 3,000 landed swaps at 0.15 SOL average trade size. Target volume is 450 SOL, so the 2% service fee is 9 SOL. Then apply the measured failure rate of your routing target to get the attempt count, multiply by the measured median fee, and add tips and rent on top. The network portion is small in absolute terms and large as a multiple of the naive estimate.

Worked from measured data, not estimated. Tips, rent and price impact are left as variables because publishing a fixed figure for an auction-priced cost would be misleading.
Line	Basis	Value
Routing target	Measured program	Pump.fun AMM (post-graduation)
Swaps to land	Your choice	3,000
Average trade size	Your choice	0.15 SOL
Target volume	swaps x trade size	450.0 SOL
Measured failure rate	Sampled blocks	23.3%
Transactions to attempt	swaps / success rate	3,912
Median fee per attempt	Sampled blocks	8,243 lamports
Network fee subtotal	attempts x median fee	0.0322 SOL
Volion service fee	2% of target volume	9.000 SOL
Jito tips	per bundle, auction-priced	variable
Account rent	once per new account	variable
Price impact	pool depth	not a fee

Two observations from the table. First, network fees are genuinely small next to the service fee at this trade size, which is why they get ignored; at larger fleet counts and smaller trade sizes the ratio inverts. Second, the attempt count is materially higher than the swap count, and that difference is the entire reason this page exists.

Price impact is listed as "not a fee" deliberately. It is not charged to you by anyone, but on a thin curve a series of buys moves price against you, so the volume you produce costs more SOL than the same volume on a deep pool. On very thin pairs it exceeds every fee in the table combined.

What actually reduces cost

Four things: routing to a target with a lower measured failure rate, setting compute limits accurately instead of padding them, spacing trades so your own wallets stop competing with each other for the same accounts, and batching efficiently so tips are spread across more transactions per bundle. Chasing a lower vendor percentage while ignoring these is optimising the smallest variable.

Route where failures are lower. Post-graduation pools fail measurably less than contested bonding curves. If your token has already migrated, there is no reason to accept curve-level failure rates.
Set compute limits correctly. Padding the limit raises your priority cost at the same bid price; underestimating it causes failures that still charge. Both are avoidable with per-instruction measurement.
Space your own trades. A fleet trading the same token in the same slot contends with itself. Spreading across slots reduces both failures and priority pressure, which is one reason duration is a real setting.
Batch efficiently. Tips are charged per bundle. More transactions per bundle means fewer bundles and less tip spend for the same activity.

Notably absent from that list: paying a smaller service percentage. The vendor fee is the most visible cost and usually not the largest, which is exactly why it is the one advertised.

That applies to us as well, so it is worth stating plainly: our fee is a flat percentage of routed volume and everything else on this page is charged on chain. The inputs behind each line are measured on the Solana volume bot overview, and the same arithmetic run against someone else quote is on the cost floor page.

The one line that moves most between campaigns is the failure cost, and it moves because of the venue. Our sampling puts the Pump.fun bonding curve at the top of that range, Jupiter aggregator routing close behind, and settled pools on PumpSwap, Raydium, Meteora DLMM and Orca Whirlpools well below both. Budget from the number for your own route rather than from a network average.

Questions

What is the biggest hidden cost in a volume bot campaign?

Failed transactions. They are charged in full, produce no volume, and occur at rates several times the network average on DEX programs. Because no vendor pricing page includes them, budgets built from advertised figures routinely fall short.

Is a 2% service fee cheap or expensive?

It is the wrong question on its own. What determines your total is the service fee plus the network cost of the attempts required to land your target swaps. A lower percentage on a routing target with a high failure rate can cost more overall than a higher percentage on a reliable one.

Does account rent get refunded?

Rent on Solana accounts is a deposit rather than a charge, and closing an account can reclaim it. In practice, whether a campaign recovers it depends on whether accounts are closed afterwards, so treat it as a cost unless your tooling explicitly reclaims it.

Why can nobody quote a Jito tip figure?

Because it is an auction bid, not a price. What you must pay depends on what other bundles are bidding for the same block, which changes continuously. Any specific figure is accurate for about a minute, which is why we publish the mechanism instead.

Does price impact count as a cost?

Not as a fee, but yes in practice. On a thin pool or an early bonding curve, a series of buys moves price against you, so producing a given volume consumes more SOL than it would on a deep pool. On very illiquid pairs it is larger than all fees combined.

Cost calculator Measured transaction costs Solana priority fees Jito tips explained Free volume bots, costed