Solana volume bot setup, one screen at a time
Nine steps from an empty panel to a run that is actually moving, written for somebody who has never opened one of these screens before. Every field is named, every decision has its consequence attached, and the arithmetic for a small first run is worked in full.
In one line
Setting up a run means filling one form: a token, a venue, a wallet count, a swap size, a budget cap and a duration, then reading the confirmation screen before anything is signed.
The short answer
Setting up a run means filling in one form. You point the panel at a funding wallet, paste a token mint address, choose the venue that token trades on, say how many wallets will send trades and how large each trade is, set a spending cap and a duration, then read a confirmation screen before anything is signed. Everything else on the screen is either a display of what you just chose or a log of what happened afterwards.
The reason it feels harder than that is vocabulary, not complexity. Six words on the form carry real consequences and none of them are explained where you meet them. This page walks the sequence in order and attaches the consequence to each field as it appears.
Before you open the panel
Two of the most expensive mistakes in this whole subject happen before you have opened a settings screen at all: using the wrong wallet, and pasting the wrong token. Both are cheap to avoid and neither is reversible afterwards. Work through this list first.
- A separate funding wallet exists. A fresh wallet, holding only what this run is allowed to spend. Not the wallet holding your other tokens, and not a wallet whose address you have published anywhere.
- The mint address is on your clipboard from an explorer. Open the token on a block explorer, copy the mint from the address field there, and never from a message somebody sent you.
- You know where the token currently trades. Before migration a token sits on a launchpad curve; afterwards it sits in an automated market maker pool. The panel needs the one that is live now.
- You have a number you are willing to lose entirely. Fees are spent whether or not anything comes of the run. Decide the figure before the form asks for it, when you are not looking at a progress bar.
- You have twenty uninterrupted minutes. The first minutes of a run are where funding problems and venue mistakes show up. Starting one on the way out of the door is how a cap gets found the hard way.
- Nothing has asked you for a seed phrase. Not the panel, not a support chat, not a browser popup. If anything has, close it and start again somewhere else.
The nine steps
Panels vary in wording and in which order they put the boxes, but the sequence of decisions is stable. Read it once end to end before doing any of it, because two of the steps are much easier if you knew they were coming.
-
Prepare the funding wallet
Create a new wallet in your browser extension and move in the run budget plus a margin. The margin matters: a run that runs out of SOL mid-way leaves worker wallets holding dust and unfinished positions. A working habit is to fund the wallet with the amount you decided on plus roughly a fifth again for fees, retries and the account rent described further down.
-
Open the panel and connect
The connect button opens your wallet extension and asks for permission to see the address and to request signatures later. Approving that request does not move funds and does not hand over a key. It is a permission, and it can be revoked from inside the extension at any time. If any part of the connection flow asks for a seed phrase or a private key, the flow is not a connection flow.
-
Paste the token mint
Paste the full mint address into the token field. A well-built panel resolves it and shows you a name, a symbol and usually a market. Check all three. Symbols are not unique on Solana and duplicate tickers are the single most common way a run ends up trading a token nobody meant to trade. If the panel shows nothing, the address is wrong or the token has no market the panel can see.
-
Choose the venue
The venue selector decides which program every swap is sent to. Before a token graduates from a launchpad it trades against a bonding curve; after graduation it trades in a pool, sometimes in several pools at once. Picking a venue with almost no depth is how a run produces enormous price impact on tiny trades and burns the budget on slippage instead of activity.
-
Set the wallet count
This is how many separate addresses will send the trades. Each additional wallet adds a fixed cost before any swap happens, because it needs enough SOL for fees and usually a token account created for the token being traded. More wallets spread the activity across more addresses; fewer wallets are cheaper to start and cheaper to clean up afterwards.
-
Set swap size and pacing
Swap size is the amount each individual trade moves. Read it against the depth of the venue you chose rather than against your budget: a size that is comfortable in a deep pool is disruptive on a thin curve. Pacing, sometimes called interval or delay, spreads the trades across the window instead of firing them in a block.
-
Set the budget cap and duration
The cap is the amount the panel may spend before it stops on its own. The duration is the window it spreads the work across. Whichever binds first ends the run, which is exactly what you want: two independent stopping conditions mean a mistake in one of them is caught by the other.
-
Read the confirmation screen
Every reasonable panel shows a summary before it starts: the token, the venue, the wallet count, the totals it derived from your inputs, and an estimated cost. This screen is where a mistyped decimal place becomes visible. Read it slowly; it is the last point at which nothing has been spent.
-
Fund, start, and watch the log
Starting a run normally means approving one transfer from the funding wallet, after which the panel distributes to workers and begins. Stay on the screen until several lines have landed. If the first handful all fail, stop the run: whatever is wrong will not fix itself, and every retry costs a fee.
The whole form in one table
The same fields under a different name on a different panel. If a box on your screen is not in this table, it is almost certainly a display of something already decided rather than a new decision.
| Field | What it accepts | What it changes | Failure mode |
|---|---|---|---|
| funding wallet | A connected address or a deposit target | Where the SOL for the run comes from | Exposing a wallet that holds more than the run needs |
| token mint | A base58 mint address | Which token every swap trades | A duplicate ticker or a copied address with one wrong character |
| venue | A curve, a pool, or an automatic choice | Fee level, depth, and the program the swap hits | Routing to a venue with almost no liquidity |
| wallet count | An integer, usually with a maximum | Fixed overhead, funding time, spread of activity | Overhead consuming a large share of a small budget |
| swap size | An amount in SOL, or a range | Price impact per trade and turnover per trade | A size the venue cannot absorb without heavy slippage |
| slippage tolerance | A percentage | The worst price accepted before a swap cancels | Set too tight, nothing lands; too loose, you overpay |
| interval | Seconds, or a min and max | How trades are spaced across the window | Bunching every trade into a few slots |
| budget cap | An amount in SOL | Where the panel stops itself | Assuming the cap includes fees when it counts notional |
| duration | Minutes or hours | The window the schedule is spread across | A window too short for the number of trades requested |
Two of those rows deserve reading twice. Slippage tolerance is the only field on the form that can silently double the cost of a run, and budget cap is the only one whose meaning genuinely differs between tools. Before a first run, find out in writing whether your cap counts trade notional, fees, or both.
The confirmation screen
The summary is the most useful screen in the whole flow, and the one people skim. It is where inputs that were entered separately are finally shown as a single arithmetic result, which is why a mistake that was invisible in a text box becomes obvious here.
Before you look
Six numbers entered on six different parts of a form, none of which were checked against each other. A wallet count that felt reasonable, a swap size that felt small, a duration picked because it was the default.
After you look
One derived line: this many wallets, this many swaps, this much notional, this much estimated cost, ending at this time. If any of those four numbers surprises you, the form is wrong, not the summary.
Check four things and nothing else. Does the token name match the token you meant? Is the venue the one where the token actually trades? Is the estimated total inside the number you were willing to lose? And is there a stopping condition, rather than a run that only ends when you come back to it? A panel that cannot answer the fourth question is one to leave alone.
What a small first run costs
The arithmetic below is illustrative. Two inputs are protocol constants and the rest are assumptions you would replace with the numbers your own venue publishes on the day. The point is the shape of the calculation, not the total.
Assume ten worker wallets, two hundred swaps of 0.05 SOL each, on a pool charging a one per cent trading fee. Turnover is straightforward: 200 multiplied by 0.05 gives 10 SOL of notional moved. Now the costs, one line at a time.
| Line | Rate used | SOL |
|---|---|---|
| base network fee | 5,000 lamports per signature, a protocol constant | 0.0010 |
| priority fee | 0.0002 SOL per transaction, assumed | 0.0400 |
| token accounts | Roughly 0.00204 SOL rent per wallet, recoverable | 0.0204 |
| venue fee | 1 per cent of each swap, assumed | 0.1000 |
| slippage | 0.4 per cent average, assumed | 0.0400 |
| total | 0.2014 |
Two observations survive changing every assumption. First, the network's own fee is almost irrelevant: the 5,000 lamport base fee per signature is a rounding error next to the venue fee, and anyone worrying about it is worrying about the wrong line. The current fee model is documented in the Solana fee documentation if you want the primary source.
Second, the token account rent is a real number that behaves differently from the others: it is a deposit rather than a payment, held by each account and recoverable when the account is closed. Whether you ever get it back depends on whether your panel closes worker accounts at the end, which is worth asking before you set the wallet count to something large. If cost is the thing you are actually trying to pin down, the breakdown behind a hosted service's Solana volume bot cost is a more useful comparison than any per-transaction figure quoted on its own.
The first ten minutes after start
The run begins and the log starts filling. Most of what you need to know is visible in the first ten minutes, and the useful habit is to treat those minutes as a test rather than as the beginning of the run proper.
What healthy looks like
Transactions appear at roughly the interval you set. A minority fail and are retried, which is normal on a network that produces a new block every few hundred milliseconds and where a transaction can expire before it is included. The running totals in the header climb at a rate that matches your arithmetic from the confirmation screen.
What to stop for
Three patterns are worth killing a run over. Every transaction failing with the same error means a configuration problem, and no number of retries will fix it. A spend rate far above the estimate means your slippage assumption was wrong for that venue. And a log that stops advancing while the header still says the run is active usually means the endpoint the panel uses is struggling, which is a good moment to stop rather than to wait.
What to check on chain
Copy any transaction signature out of the log and open it on an explorer such as Solscan. You are checking three things: that the transaction exists, that it swapped the token you intended, and that the fee shown matches what the panel claimed. If those three agree, the panel is telling you the truth about the rest of it as well.
Six mistakes people make once
Not exotic failures. These are the ordinary ones, in rough order of how often they turn up in questions sent to this desk.
- Connecting the main wallet. The panel only needs a wallet holding the run budget. Connecting one that holds everything else means an unrelated mistake later has a much larger blast radius.
- Trusting a ticker. Two tokens can share a symbol, and one of them is usually the one you did not mean. Only the mint address identifies a token.
- Choosing the wrong venue after migration. A token that has graduated may still show an old curve in some interfaces. Trades routed there behave nothing like trades in the live pool.
- Setting slippage generously to make trades land. It works, and it is the most expensive way to solve the problem. Wide tolerance turns a landing-rate problem into a silent cost problem.
- Reading the cap as a total. If the cap counts notional and not fees, a run stops at the cap having spent more than the cap. Find out which before, not after.
- Walking away during the first minutes. The failure modes that cost real money all appear early. Ten minutes of watching is the cheapest insurance in this entire process.
None of those six require technical knowledge to avoid. They require reading the confirmation screen and staying at the desk, which is a strange thing to have to write down and the reason this page exists.
Where to go from here
Three fields on the form above carry enough weight to have their own pages on this site. The token and venue pickers decide what you are actually trading, wallet count and swap size decide the cost floor, and the caps decide when the whole thing ends. Read those three field guides before a second run, in that order.
If you would rather look at a working screen than read another description of one, most of what is written above maps directly onto a hosted Solana volume bot, where the same decisions appear as one form with a log underneath. Seeing the fields laid out is often faster than reading about them, because the form is forced to name every choice the software makes on your behalf.
The next walkthrough covers the step most people rush: the connection itself, what a signature request is actually asking for, and the three ways a panel can end up touching your wallet. It is worth reading before you press connect rather than after.
Written by The Setup Desk. Field names and screen layouts differ between tools, so this page describes what a setting is for rather than where a particular build puts it; any figure in an example is arithmetic chosen to show the shape of the calculation and describes no real account. The way these pages are put together is set out in how this desk works, and anything that reads wrong can be sent back as a correction.