How screeners rank new tokens, and what you can influence

Nobody publishes their ranking formula, and they all change it. What doesn't change is the raw material: every screener reads the same public chain data, so the inputs are knowable even when the weights aren't.

Worth knowing which of those inputs you can affect.

The inputs

Five, and they show up everywhere.

Volume over a trailing window, usually 5 minutes, 1 hour, and 24 hours computed separately.

Transaction count, often split into buys and sells so the ratio can be weighted.

Unique traders or makers, meaning distinct addresses rather than trade count.

Price movement, generally weighted toward upward moves over the same windows.

Liquidity, working as a threshold and as a factor at once, since many boards exclude pairs below a floor entirely.

Some screeners add holder change, some add pair age, and a few weight social signals. Core 5 are common to all of them.

The property that decides everything

Trailing windows.

Your score comes from a window that moves continuously, so trades from 90 minutes ago have already left the calculation entirely, which means rank is never a state you reach and hold but a value recomputed every few minutes against whatever happens to be active at that moment.

So what does that feel like? Your numbers hold perfectly steady and your position drops, because 30 other coins got busier while nothing changed on your side. Why my coin fell off trending covers that specific case.

Why two screeners disagree

Four reasons, all mundane.

Different windows, so one reads 5 minutes while another reads an hour. Different weights, so one favours unique traders and another favours volume. Different fields, since a Solana-only board has a smaller comparison set than a multi-chain one. And different refresh intervals, so one simply hasn't recomputed yet.

None of that means either is wrong. How Birdeye trending works covers one that weights ownership more heavily than most.

What you can influence

Three of the five, honestly.

Volume, trade count, and unique traders all come from activity, and activity is something you can produce or attract. That's the entire mechanism a campaign operates through.

Price movement you cannot control, and trying to control it means buying your own coin, which shows up on your creator wallet permanently. How dev buys affect the curve covers why that trade is bad.

Liquidity is structural before graduation and pool-determined afterwards.

The part that isn't in any algorithm

What happens after somebody clicks.

Ranking gets you onto a list. Everything past that point is your page, your holder distribution, and your creator history, none of which any screener measures. How holders check your dev wallet covers the 60 seconds that decide it.

So a coin can rank well and convert badly, and buying more ranking repeats the ratio at higher cost. Check whether unique holders rose during your last visible period before spending on the next one.

The hour matters more than people think

Same activity produces different ranks depending on when it happens.

A board at 04:00 UTC has a much thinner field than the same board at 20:00, so 15 SOL an hour might place you in the visible range during one and nowhere during the other. Nothing about your coin changed. The competition did.

Which is a real argument for longer windows: 12 hours crosses several of these conditions instead of betting everything on one. How long a campaign should run covers picking a duration.

What you can move, and what you cannot

Screeners read volume, trade count, unique traders, price movement, and liquidity over trailing windows, recompute continuously, and weight all of it differently from each other. You can influence the first 3 by producing activity, and nothing you do affects what happens after the click. The console exposes volume, wallet count, and duration separately, with the full cost shown before anything is signed.