A wallet score compresses observable activity into a signal. The campaign turns that signal into a decision: allow participation, hold a reward for review or exclude a claim. The cost sits in that decision, including the cases where the signal is wrong.
A Sybil attack involves one actor presenting multiple identities to gain an advantage. In a reward program, that can mean collecting benefits intended for distinct participants. John Douceur's foundational paper describes the identity problem. It does not establish the accuracy of any modern wallet score.
Separate identity, eligibility and risk
Identity asks who or what the record represents. Eligibility asks whether the published campaign conditions are satisfied. Risk assessment asks whether the observed behavior deserves closer inspection. A wallet address alone does not answer all three.
A legitimate new user can have a new wallet. An old wallet can participate in abuse. Two wallets can share an activity pattern for ordinary reasons. Scoring is useful when it informs a defined process that accounts for those possibilities.
| Signal | What it can contribute | What it cannot establish alone |
|---|---|---|
| Wallet age | History available for observation | Whether the owner is legitimate |
| Transaction history | Evidence of supported onchain activity | Product interest or a unique person |
| Repeated timing or funding patterns | A reason to inspect a group of records | Common ownership or fraudulent intent |
| Verified product event | Evidence that a qualifying action occurred | Whether another account earned duplicate credit |
| Prior campaign participation | Context about recorded participation | Future retention or reward entitlement |
The relevant signals depend on the networks, data and integration available. Missing history should remain missing history in the report, rather than silently becoming evidence of abuse.
Define the action a score can trigger
A low-risk introductory task may need a different review policy from a valuable reward. Start with the reward obligation and decide which uncertainty warrants intervention.
An illustrative process has three outcomes: the record meets the rule, the record needs review, or the record does not meet a published condition. Define who handles the middle group and how long a participant should expect to wait. Keep the reason for a decision so the team can explain it later.
Reward caps, duplicate-event handling and clear qualification windows still matter. A score at entry does not prevent the same verified event from being credited twice if the reward system handles retries incorrectly.
Read the denominator before claiming lower cost
Suppose a campaign spends $100,000 and produces 50,000 participant records. That is $2 per record. If 15,000 records complete the qualifying product action, the cost is about $6.67 per qualified record, before any excluded operating costs.
Those illustrative figures do not imply that the other 35,000 records are fraudulent. They might be incomplete journeys, ineligible participants, duplicate records or people who chose not to continue. The qualification rule determines what the denominator means.
Now imagine a stricter gate reduces reward spend. That is only part of the result. It may also reduce qualified completions, increase support work or exclude valuable newcomers. Compare the same outcome, time window and cost categories before calling the new policy more efficient.
Review both sides of the decision
Sample accepted cases as well as rejected ones. Reviewing only rejected claims can reveal mistaken exclusions while missing suspicious participation that passed. Reviewing only the accepted group has the opposite blind spot.
Use a consistent review process and record where evidence is inconclusive. A manual review is not automatically perfect ground truth. Track appeals, overturned decisions, reward concentration and the time required to resolve a case.
The useful operating questions are concrete:
- How many legitimate qualifying cases did the rule appear to exclude?
- How much suspicious reward activity still passed?
- What did review and support cost?
- Did participants understand the pending state and appeal route?
- Did abuse move to a different step in the journey?
Use wallet context where it improves the decision
Claimr Wallet Intelligence provides wallet context that can inform campaign segmentation and eligibility. Confirm the available signals and controls for the intended integration before designing the policy around them.
Begin with a limited campaign and inspect the decisions it produces. A useful scoring process protects the reward budget while preserving a workable path for eligible participants. Its value is established through those outcomes, including the errors, rather than the score distribution alone.
Keep learning
Further reading
Continue with the sources and practical guides behind this article.
The foundational account of one actor presenting multiple identities. It describes the identity problem, not the accuracy of a modern wallet score.
How qualification and reward timing change acquisition cost, with a worked comparison and the limits of the evidence.
Work through the audience, participant journey, rewards, launch checks and evaluation for your own campaign.
