Compare token calls
What this is
Compare token calls puts between two and five tokens side by side and shows how the calls Cognivo has observed for each one compare. A call is a public post where someone named a token. Cognivo compares who called them, when, how widely, and what happened after.
The page states its own scope at the top: this compares observed call signals only. It is not a safety check, not a holder or liquidity comparison, not a price forecast, and not advice.
When to use it
Use it when several tokens are being talked about at the same time and you want to see the shape of that talk rather than the volume of it. It answers questions like: which of these did people start naming first, which one has been named by the widest range of callers, and are the same few callers behind more than one of them. It does not tell you which token to buy.
Where to find it
In the dApp sidebar, open Intelligence and select Compare Calls. The page itself is titled Compare token calls.
- Choose the network the token lives on. It starts at Network not set, and the same symbol can belong to many contracts, so the network and the contract address are what tell Cognivo which token you mean.
- Select Add token to put that token on the list, then repeat until you have between two and five tokens.
- Select Compare once at least two tokens are listed. The button stays greyed out until then, and comparing costs no credits.
You can compare tokens on Ethereum, Base and BNB Chain. Cognivo identifies a token by its network and its contract address, so a symbol on its own is not enough. If you enter a symbol that matches more than one contract, Cognivo shows Which token did you mean? and asks you to pick the exact contract instead of guessing for you.
What you get back
At the top sits the Cognivo Verdict with a plain quick read. It lands on one of four states: Clear signal leader, Narrow edge, Mixed category leaders, or Insufficient evidence. Cognivo names a leader only when the observed calls support one, and a token being listed first is never the result.
Below that, the evidence is laid out in sections you can read in order:
- In this comparison lists the tokens Cognivo actually resolved, by network and contract.
- Score and timing gaps shows how far apart the tokens are on the Cognivo Signal Score, and which was seen first.
- Who leads each category covers earliest detected, widest caller breadth, and strongest measured follow through.
- Why it leads, What limits this signal and What would change this explain the verdict and name what is missing.
- How fresh the evidence is describes the newest call observed for each token.
- Callers who named more than one flags overlap. Callers who named more than one of these tokens are not independent evidence for each of them.
- The original calls links out to the posts Cognivo could verify. Where it has no verified link, it says so rather than pointing at a guess.
Every measurement carries its own state, and the difference matters. Measured means Cognivo measured it from the calls it observed. Not enough evidence means Cognivo has nothing measured there, and that is not the same as a zero. Not available here means Cognivo does not measure that for that token. Cannot be compared means Cognivo cannot line the tokens up fairly, so it shows no figure. If an entry does not resolve at all, it appears under Cognivo could not identify every token, which means Cognivo has no observed calls for it, not that nothing exists.
Limits and cost
Comparing costs no credits, and there is no confirmation step, because nothing here charges. The card is marked Free.
Five tokens is the maximum in one comparison, and each token can appear only once. A higher Cognivo Signal Score is not a reason to buy, and a lower one does not mean a token is dangerous. Comparing callers rather than tokens is not available yet while Cognivo builds out its verified history of past calls. Token comparison stays available.
Next
To see the calls behind a single token rather than a comparison, open KOL Intelligence. To understand how Cognivo separates what it measured from what it could not verify, read Not verified, not assessed, augmented.