Budgeting the Curve: What a Launch Actually Costs

The cost lines of a curve launch

A curve launch has roughly seven cost lines: network fees, launchpad fees, your own first buy, design and asset work, paid promotion, any tooling you pay for, and the reserve you keep for the hours after launch. Write all seven in SOL before you mint. The one that gets cut is the reserve, and it is the one that decides how the second day goes.

Most launch budgets are not wrong because a number was miscalculated. They are wrong because a line was never written down at all, and then had to be paid for out of another line at the worst possible moment. A budget with seven honest lines and modest amounts survives launch day better than a larger budget with three lines and an unspoken assumption behind each of them.

Write the budget in SOL. Nearly everything here is paid in SOL, the network fee is denominated in SOL, and your own first buy is a SOL amount. If you plan in fiat you will spend launch day recalculating a moving target instead of running the launch. Note the rate you were looking at when you wrote the plan, then leave the fiat figure alone.

UNIT

SOL, which has nine decimal places, so small fee lines are expressible exactly

LINES

Seven, of which one is a reserve that is not meant to be spent on day one

OWNER

One named person who authorises every outgoing payment on the day

SIGNED OFF

Before the mint transaction, not during the first quiet hour

Fees you do not set

Two of the seven lines are prices you have no say in. Network fees are charged by the Solana network per transaction and paid in SOL, and they apply to the mint, to every buy and sell, and to every account you create along the way. They are individually small and collectively not nothing, particularly if your plan involves many transactions.

There is one genuine recovery inside this line. Accounts on Solana hold a rent-exempt balance in SOL while they exist, and that balance is returned when the account is closed. If your launch creates accounts you will later close, part of that money comes back. It is a small effect, but it is the only line in the budget that is recoverable by an action you control.

The second price you do not set is the launchpad's own fee. Every platform publishes its own schedule and those schedules change, so the correct move is to read the current figure on the launchpad itself on the day you write the budget. This desk is independent of every launchpad and will not print a rate it cannot stand behind next month.

The creator first buy

Most operators buy some of their own token at the start. This is a legitimate line, and it is also the line most likely to be described dishonestly afterwards. Budget the amount deliberately, decide it before the day, and publish the address that made the buy. An undisclosed creator buy sitting in an unlabelled wallet is a misrepresentation, and it is one that readers find on a public ledger.

Financially this line behaves differently from the rest of the budget. Fees and invoices are gone at the moment of payment. A first buy converts SOL into a token position, and that position can be worth more later, less later, or nothing at all. It is not recoverable, it is at risk, and those are different categories that budgets frequently collapse into one.

Size it so that the loss of the whole amount is survivable and so that the resulting balance is a share of the holder list you are willing to defend in public. Both constraints are real, and the second is often the binding one. The reading procedure for that share is set out in the piece on first buyers and distribution.

Assets and paid promotion

Design and asset work is the one part of the budget that produces something you keep. An image, a set of social assets and a written description outlast the launch day and can be reused. It is still sunk spending in accounting terms, because the money does not come back, but the output has a life beyond the first hour in a way that a fee does not.

Paid promotion is the opposite. It buys placement or attention for a period and then it is over, whatever happened during that period. Budget it as a fixed amount decided in advance rather than as a tap you open when the curve is quiet, because opening the tap during a quiet hour is the single most reliable way to convert a disappointing launch into an expensive disappointing launch.

Be specific about what promotion can and cannot do. It can put your token in front of people. It cannot make them buy, and it cannot make them hold. Nobody controls whether attention converts into buying, or whether buying arrives at all. A budget line that assumes conversion is not a budget line, it is a forecast with an invoice attached to it.

Pricing tooling as a budget line

If you are going to pay for tooling, price it before launch day and put it in the ledger like any other supplier. Services in this category are usually priced either as a package with a stated deliverable, as a percentage of the trading volume they process, or as a subscription for a period. Each pricing model behaves differently when your launch does not go the way you hoped.

Ask a short set of questions before committing budget, and get the answers in writing. What exactly is delivered, in units you can count. What happens if the curve stays quiet. Whether network fees are inside or outside the quoted price. A published figure such as a stated Solana volume bot cost is a starting point for that conversation, not the end of it, and the honest framing is that this is spending, not an investment with an expected return.

Hold the line on categories. Money paid to a service is an expense, whatever the outcome. If a proposal is being sold to you as something that will bring buyers in, the proposal has left the territory anyone can promise. Nobody controls whether buyers arrive, and a supplier who implies otherwise is telling you something about themselves rather than about your launch.

What running activity costs to operate

Automated activity has an operating cost separate from whatever the service charges. Every transaction pays a network fee in SOL. Every wallet involved has to be funded before it can do anything, and funding wallets means moving SOL into accounts that then need to be tracked. The arithmetic is dull and it is the part most likely to be missing from a quote.

Those fees are paid whatever the result. A transaction that lands during an hour when nobody is watching costs exactly the same as one that lands during your busiest hour. Whether you are buying a package or looking at what a SOL volume bot charges to run, the money leaves regardless of outcome, which is the definition of a cost rather than an investment, and nobody controls whether buyers arrive alongside it.

There is also a reconciliation cost in your own time. Funded wallets have to be accounted for, drained and closed at the end, and the movements have to be explicable if anyone asks about them later. If you are unwilling to explain those movements publicly, that is a clear signal about the plan and it should be dealt with at the budget stage rather than at the accusation stage.

The reserve, and why it gets cut

The reserve is money set aside and deliberately not committed before launch. It exists so that on the second day you still have options: a further round of assets, a promotion slot that only appears once the token is trading, a supplier invoice you did not anticipate, or simply the ability to do nothing calmly rather than nothing helplessly.

It gets cut because it is the only line that buys nothing on the day the budget is written. Every other line has an advocate in the room. The reserve is defended by whoever is imagining the quiet hour, and that person is always outnumbered by whoever is imagining the launch going well. Protect it by fixing it as a share of the total before any other line is discussed.

The reserve is also the line that connects to the triage work. A stalled curve is diagnosed before it is spent on, and the sequence for that is set out in the stall triage. Having a reserve does not mean deploying it. Most of the time the correct use of a reserve is to still have it on day three.

An illustrative budget in SOL

What follows is arithmetic, not advice, and every figure in it is invented to show the method. As an illustration, if you budget 40 SOL in total, the seven lines can be set out as below. The launchpad fee line carries a labelled placeholder of 1.0 SOL because the real rate has to be read on the platform, and the total is computed with that placeholder in place.

Read the totals rather than the individual lines, because the individual lines are invented. Committed spending before the curve opens is 25.0 SOL, which is 40.0 minus the 15.0 reserve, or 62.5 per cent of the budget. The reserve is 15.0 of 40.0, which is 37.5 per cent. Genuinely recoverable spending is only the rent portion inside the 0.3 SOL fee line.

That last figure is the one worth sitting with. In this illustration, of 40 SOL budgeted, 8.0 SOL is at risk in a token position, 15.0 SOL is retained if you do not spend it, and almost everything else is gone at the moment of payment. A budget that looks flexible on paper contains, in practice, one line you can still change on the second day.

Stop-loss, records, and what a budget cannot buy

Decide the stop-loss before launch day and write it down with a named person authorised to call it. A limit set in advance is a decision made by people thinking clearly. The same limit revisited at hour four of a quiet curve, with the desk watching and the group chat unhappy, is a negotiation you will lose against your own optimism almost every time.

Continuing the illustration, a workable rule on a 40 SOL budget is that no more than half the 15.0 SOL reserve, so 7.5 SOL, may be drawn in any one day, and that any draw at all requires a written reason recorded at the time. Two rules, both checkable afterwards, both trivially easy to follow if agreed in advance.

Record spend as it happens, not afterwards. One line per outgoing payment with the time, the amount in SOL, the destination and the reason. Wallet history will give you the first three later, but the reason is the part that disappears, and the reason is what makes the review worth doing at all. A ledger without reasons is a bank statement, and nobody learns anything from one of those.

• All seven lines written in SOL and totalled before the mint.
• Launchpad fee read from the platform on the day, placeholder replaced.
• Reserve fixed as a share of total before any other line is negotiated.
• Daily draw cap and a named person authorised to approve draws.
• Creator buy address published alongside the rest of your wallet disclosure.
• Live spend log open on the desk from before the mint transaction.
• Review scheduled for day two, with the ledger closed and reasons attached.

Finally, be clear about what a budget cannot buy. It cannot buy buyers, attention, or a curve that completes. It funds preparation, materials, fees and attempts, and the outcome of those attempts is decided by people who have never heard of your plan. Any budget whose logic is that spending more produces a better result has stopped being a budget and become a hope with columns.

The rest of the phase-by-phase picture, including where each of these lines is decided and by whom, is in the four-phase operating plan and in the launch day hub. If the budget survives to the migration step, the costs and readiness questions that arrive there are covered in planning for graduation.

Questions the desk is asked