Why launch activity matters
When a token launches on Pump.fun, everyone starts from the same bonding curve. In theory the first buyers are just early. In practice, a large part of the first buys often comes from bots and from wallets connected to the creator. If these wallets end up holding a big slice of supply, they can sell into every pump and the price struggles to go anywhere. That is why traders look at three groups before buying: snipers, bundlers and insiders.
Snipers
A sniper is a wallet that bought in the very first blocks after the token went live, usually through a bot watching for new launches. Sniping is not automatically malicious — many snipers are just fast traders — but they buy at the lowest possible price and tend to sell into the first wave of demand.
What to look at:
- Supply still held by snipers. This is the number that matters. "9 snipers · 0%" means nine wallets sniped the launch and all of them have already sold: the selling pressure is gone.
- Number of snipers. Many snipers still holding a meaningful share (for example 10%+ combined) is a red flag on a young token.
Bundlers
A bundle is a group of buy transactions that land together, typically in the same Solana slot through a Jito bundle. Creators use bundles to buy a large part of the supply from many fresh wallets at launch, so the holder list looks distributed when in reality one person controls those wallets.
Bundled supply is usually more dangerous than sniped supply, because it is coordinated: the wallets can be dumped together. A token where bundlers still hold 20–30% of supply can collapse the moment they decide to exit.
Insiders
Insiders are wallets linked to the creator. Depending on the data provider, a wallet is flagged as insider when it received tokens directly from the dev wallet through a transfer (not a swap), or when it was funded by the same source as the dev. Insider supply is hidden dev supply: if the dev holds 0% but insiders hold 15%, the dev effectively still holds 15%.
Why every tool shows different numbers
There is no official definition of these groups, so each data provider sets its own rules:
- Snipers: some count buys within the first few slots after the first swap (for example 5 slots), others use the first seconds or the first N buyers.
- Bundlers: some require several buys in the same slot, others also count a few different wallets buying in adjacent slots — a looser rule that can flag normal fast buying.
- Insiders: some only count direct transfers from the dev, others build a graph of funding links between wallets and find many more.
The practical consequence: compare a token with itself on the same tool, not numbers across different tools. Edge Rugcheck shows which provider supplied the values next to the section title, so you always know which definition you are reading. Objective on-chain values — top 10 holders, dev holding, LP burn, mint and freeze authority — should match across tools.
Reasonable starting thresholds
These are the starting values used by the Edge Rugcheck presets. Treat them as a starting point and tune them to your own strategy:
| Stage | Snipers | Bundlers | Insiders |
|---|---|---|---|
| New Pairs | ≤ 15% | ≤ 20% | ≤ 20% |
| Final Stretch | ≤ 10% | ≤ 15% | ≤ 15% |
| Migrated | ≤ 5% | ≤ 10% | not checked |
Key takeaways
- Look at the supply these wallets still hold, not just how many there were.
- Bundled and insider supply is coordinated and usually more dangerous than sniped supply.
- Percentages differ between tools because definitions differ — always check the source.