How to Find Solana Meme Coins Before They Trend
--
Most people discover a Solana meme coin the same way: a screenshot in a group chat, a viral tweet, or a green candle on DexScreener that’s already up 8x. By that point, the easy money is gone and you’re the exit liquidity.
I spent the last eight months building MemeScout — a radar for Solana meme tokens — and along the way I analyzed thousands of launches looking for one answer: what actually separates a token that’s about to trend from one that already has? This is what I learned, and how you can apply it without paying for any tool (including mine).
The trending tab is a trap
By the time a token hits the top of DexScreener’s trending list, it has usually already done 60–80% of its move. Trending lists are rearview mirrors — they rank tokens on 24h volume and recent price action, which means the signal arrives after the price does.
If your edge depends on being early, you need to stop filtering on “what’s popular right now” and start filtering on “what’s quietly lining up.” The patterns are boringly consistent once you know what to look for.
The four-hour window
The majority of 2x+ moves on Solana meme coins happen within the first few hours of the token hitting a usable liquidity threshold. After about four hours, one of three things is true: it’s already pumped, it’s already rugged, or it’s going to flatline. Almost none of the 10x+ tokens I’ve studied were discovered in the “mature” phase.
So the real question isn’t “is this token good?” — it’s “is this token young enough and clean enough that something can still happen?”
Filter 1: Liquidity-to-market-cap ratio
The single most useful filter I use is the ratio of liquidity to market cap. If a token has a $300K market cap but only $15K in liquidity, that’s a 5% ratio — any sell of meaningful size will crater the price and you’ll never get out at the price the chart shows.
My rough rule: I ignore tokens where liquidity is less than 10% of market cap, and I pay extra attention to ones in the 15–30% range. Above 30% usually means the bonding curve is still active or the pool is freshly funded — both interesting.
You can compute this from any token page on DexScreener or Birdeye. It costs you nothing.
Filter 2: On-chain authority checks
Before a token even has a price history, you can learn a lot from its on-chain metadata. Three things I check every time.
Mint authority. If the developer still holds mint authority, they can print unlimited new tokens and dilute your bag to zero. This should almost always be revoked. Unrevoked mint is the single biggest “hard no” in my scoring.
Freeze authority. If this is active, the dev can freeze your wallet and stop you from selling. It’s rare on legitimate launches. Rarer on ones you’d actually want.
Top-3 wallet concentration. If three wallets hold more than 35–40% of supply outside of the liquidity pool, you’re one coordinated sell away from getting dumped on. I treat anything above 50% as untradeable.
You can check all of this on Solscan or through a tool like RugCheck in about 30 seconds.
Filter 3: Order flow, not price
Price is a lagging indicator. Order flow is a leading one. When I analyzed 30 days of radar data, the tokens that went on to hit 2x+ had two things in common: a buy/sell ratio above 2.5 (meaning way more buyers than sellers at the moment of detection), and a 5-minute-to-1-hour volume ratio above 0.25 (meaning volume is accelerating, not decaying).
Tokens that were flatlined or pumping-and-dumping usually had buy/sell ratios near 1.0 and decelerating volume. The chart can look the same in the first 90 seconds — the order flow is what tells you which one is real.
Filter 4: A threshold beats a gut call
One of the hardest lessons: your discretionary judgment is worse than a checklist. When I started MemeScout, I was tuning a radar score by feel. Then I ran the numbers on a month of calls and found tokens scoring 30–35 were responsible for 14 out of 18 flatlines, while tokens scoring 42+ had a 78% hit rate on a 1.25x move. I raised the threshold and it cut about 60% of the noise overnight.
The point isn’t the specific number — it’s that a boring, mechanical filter beats a clever gut call almost every time. Whether you use my tool or a spreadsheet of your own, commit to a threshold and stick to it.
The manual workflow
Here’s the one I use when I’m hunting without any automation. Open a Solana new-pairs feed (DexScreener “New Pairs” filtered to Solana, min liquidity $20K). Sort by age ascending — anything older than 4 hours is out. Check liquidity-to-market-cap ratio and toss anything below 10%. Click into RugCheck and confirm mint revoked, freeze revoked, and top-3 concentration reasonable. Look at the last five minutes of trades — are buys outpacing sells by at least 2:1, and is volume accelerating? If all five checks pass, size small and set a stop.
This takes about 90 seconds per token once you’ve done it a few times. You will pass on a lot of tokens that end up pumping — that’s fine. You will also stop buying 80% of the rugs and flatlines you would have bought on vibes.
Where a tool earns its keep
I’ll be honest about where MemeScout earns its keep. It watches 24/7 and applies these same filters automatically, so you don’t have to be glued to DexScreener to catch the four-hour window. It also layers in things that are hard to do manually — bot detection on copy-trading wallets, call-tier classification based on market cap at first detection, and an Alpha Score for wallets that consistently buy within minutes of a radar call.
But the filters above are the filters. The tool is just faster execution, not a secret formula. If you internalize the framework, you’re already ahead of 95% of people aping off a Telegram screenshot.
If you want the daily version of this — the handful of tokens that pass every filter each day, plus rug alerts and a weekly wrap-up — you can get it in your inbox at memescout.io/digest. It’s free. If you want the live radar with the filters running in real time, that’s at memescout.io.
Either way: stop chasing the trending tab. Start filtering the firehose.