KOL in crypto: how to check a KOL's track record before you follow a call
A KOL (key opinion leader) is an account whose token calls move prices. Before you follow one, check five things: how many of their calls were actually measured, whether the hit rate counts spikes or real results, how many calls came after the pump, whether the record was built from winners only, and whether the first half of their history looks like the second. Of 6,995 accounts in our database, 5,192 have zero measured calls, so most 'proven KOLs' are unproven.
What a KOL is
In crypto a KOL is an account on X or Telegram whose calls make other people buy. The word says nothing about whether those buyers made money. Our earlier post explains the term; this one is the checklist.
Step 1: how many calls were measured?
Show as table
| Value | Note | |
|---|---|---|
| Zero measured calls | 5,192 | nothing to judge |
| 1 to 9 measured | 1,655 | page exists, no verdict |
| 10 or more clean calls | 148 | verdict possible |
Most accounts people call KOLs have never had a call measured by anyone. A screenshot is not a measurement. We need 10 clean measured calls before a page says anything, and the page shows the count.
Step 2: what does "hit" mean?
If a tracker counts a call as a hit when the price touched +50 % at any moment in 24 hours, it is counting wicks. Accounts with 85 to 100 % such hits had median 24-hour results of −84 % to −99 % in our data. Ask for "price higher one hour later" and "median 24 h result". Those are what a follower could have had. See hit rate.
Step 3: how many calls came after the pump?
A late call is posted after the price already doubled in the previous two hours. Depending on the source, 10 to 30 % of measured calls are late. They are exits, not entries, and a record that includes them is a record of other people's exits.
Step 4: how was the record built?
Many trackers find callers by starting from tokens that pumped and asking "who posted this first?". Every account found that way looks good: you only ever see their winners. We tag those sources and leave them out of the public numbers; the honest record is built from an account's whole timeline, losers included.
Step 5: does the first half look like the second?
Split the calls in time. Of 39 accounts with 20+ clean calls in our data, two were positive in both halves, and random shuffling of the same outcomes gives one such account anyway. A good month followed by a bad one is the shape of luck. The month-by-month table on every profile shows it.
What to do with a KOL that passes
- Follow calls on tokens between $200k and $1M market cap (the only band where the 1-hour result was not badly negative in our data).
- Never enter a late call; never enter under $30k.
- Plan the exit before the entry; our demo accounts sell half at +15 % and the rest with a 15 % trailing stop.
- Expect most calls to lose anyway: the best accounts beat the median, they do not beat gravity.
Members see each account's live calls and our entry rule per account. Not financial advice.
Compare tools: Best crypto analysis tools
MemeSonar does this for you in real time: every new Solana token, every influencer call and every tracked wallet, scored and filtered for rugs.
Request accessNot financial advice. Numbers are measurements of past tokens and calls; meme coins are extremely risky and most go to zero.