1. Who this is for
Anyone. You do not have to be a customer, and you do not have to own the address in question. If you are a token team, a deployer we have labelled, or a holder who thinks a scan misread something, this is the route.
Every verdict we publish is an automated assessment produced by the method at /methodology, not a statement of fact about any person. If the underlying data or our reading of it is wrong, we want to fix it — a wrong verdict is a defect in the product.
2. What to send
Email [email protected] with the subject line Contest a verdict, and include:
- The address. The token mint or the deployer wallet, in full.
- What is wrong. Which specific number, label or finding you are disputing — for example "the top-holder figure counts a locked treasury" or "this wallet is not the deployer".
- Why, with evidence. Transaction signatures, a lock or vesting contract address, an exchange attestation — whatever shows it. On-chain evidence is strongest, because it is what we can verify independently.
- How to reach you. An email address we can reply to.
You do not need a solicitor to use this route and we would rather you did not need one.
3. What we check
Every challenge gets the same review:
- Re-run the scan clean. A fresh, non-degraded trace, because a partial trace is one of the known ways a verdict goes wrong.
- Verify the disputed input. Was a pool, treasury or locked allocation counted as a real holder? Is the deployer attribution right, given that pump.fun ownership can be transferred? Did an exchange or infrastructure wallet get read as a funder?
- Re-apply the published thresholds to the corrected inputs, without hand-tuning the outcome for the individual case.
- Decide, and write down why.
4. What we will and will not change
| If | Then |
|---|---|
| An input was wrong — misidentified wallet, pool counted as a holder, wrong deployer | We correct the data, rescan, and the verdict moves with it. Cached and shared results are invalidated. |
| A threshold produced an outcome we agree is unreasonable across the board | We change the threshold for everyone and note the change at /methodology. |
| The data is right and the verdict follows from the published method | The verdict stands. We will explain which signal fired and point you at the transactions. |
| You would simply prefer a lower score | The verdict stands. We do not remove or soften verdicts on request, for payment, or under pressure. |
5. Timing
We acknowledge within 2 business days and give a reasoned decision within 10 business days. If a re-scan needs data we cannot get quickly, we will say so rather than let it go quiet.
6. What happens to a corrected verdict
Corrections are recorded, not silently applied. We keep a log of every challenge, its outcome and its date, so a verdict that was changed is identifiable as having been changed. If a corrected assessment was already shared or embedded, the correction propagates to those surfaces when they next refresh.
7. Reporting something else
- Security vulnerability: [email protected] — see security.txt.
- Billing or a key problem: [email protected], or the refund policy.
- Privacy request: [email protected], or the privacy policy.