How to spot a volume bot on chain, in about five minutes

Manufactured activity leaves fingerprints when it's done cheaply. Learning to read them takes about 20 minutes, and it pays off twice: once when you're judging somebody else's coin, and again when you're checking whether the campaign you just paid for produced anything worth having.

Everything below is public data, and none of it needs a paid tool.

The five tells

Open a coin's transfer list in Solscan and look for these.

Metronome timing. Trades landing at suspiciously even intervals. Every 41 seconds, forever. Real flow arrives in clumps and gaps, because humans and markets are irregular.

Repeated sizes. The same amount over and over. 0.4 SOL, forty times. Real traders pick numbers for reasons that vary, so a tape repeating one figure is a script with a hardcoded constant.

Narrow wallet set. Twelve addresses producing 400 transactions. High trade count with a stuck unique-wallet count means the same participants cycling rather than new people arriving.

Fresh addresses with no history. Click through the buyers. If each one's first ever action was funding, then buying this coin, then nothing, that's a fleet rather than a crowd.

A common funder. Follow where those addresses got their SOL. All roads leading to one wallet 10 minutes before the activity started is the loudest signal available, and clustering tools draw it as a star. Bubblemaps and holder analysis covers what that looks like visually.

The five-minute version

You don't need all of it. Do this:

1. Open the mint in Solscan and go to transfers
2. Count distinct addresses in the last 50 transactions, not transaction count
3. Glance at the timestamps for regularity
4. Glance at the amounts for repetition
5. Click 3 buyers and check whether they existed before this coin

Steps 2 and 5 catch most of it. Fifty transactions from 9 addresses that were all created this morning is not ambiguous.

What competent activity looks like

Worth describing the other side, because "spot the bot" implies all of it is detectable and that isn't true.

A well-built campaign spreads across hundreds of wallets, each with its own funding path and prior history. Trade sizes vary inside a band. Gaps between trades come from a distribution rather than a constant. Wallets aren't reused across campaigns, so they don't accumulate into a permanent identifiable cluster.

At that point you're looking at real swaps, from wallets that look like wallets, arriving the way trading actually arrives. Whether it was paid for isn't visible, because the only thing that would reveal it is a pattern, and the pattern is what the work removed. How a fleet gets funded covers why that's genuinely hard rather than a matter of settings.

Why this matters if you're buying activity

Why does it matter? Your buyers run this exact check.

Pay for a campaign that produces 12 addresses on a metronome and you've bought something worse than nothing. The coin now has visible manufactured activity attached to it, and anybody evaluating it reaches a conclusion you can't argue with, because it's right there in the transfer list.

So the check runs both ways. Before you pay anybody, ask for a mint they've worked on and run these five tells against it. A service that won't provide one has answered the question. How to judge a volume service puts that first for exactly this reason.

What it doesn't tell you

Two limits, stated honestly.

Spotting a fleet doesn't tell you the coin is a scam. Plenty of legitimate projects buy visibility, the same way they'd buy an ad. It's a marketing spend, and the ethical question is what the creator tells holders rather than what the wallets did.

And a clean tape doesn't prove nothing was paid for. Good execution is indistinguishable from organic trading by design, so absence of evidence is weak here even though presence of evidence is strong.

Running it on your own coin

Do this after any campaign, whoever ran it:

• Distinct addresses across the window, not transaction count
• Timing spread across hours rather than clustered into minutes
• Size variation rather than one repeated figure
• Buyers with histories predating your mint

If your own coin fails the check, you know before your buyers do. How to use Solscan covers where each number lives, and the console shows fleet size and window as explicit settings before anything is signed.