Choose the group and the observation window
Write down which campaign and join dates you are evaluating before opening the numbers. A useful starting cohort is everyone who joined one campaign during its recruitment window. Follow that same group through submission and reward redemption, even when those events happen later.
Record the report cutoff date. A customer who joined yesterday has had less opportunity to submit than someone who joined two weeks ago. Comparing them as if both had finished can make a healthy campaign look stalled. Keep open obligations visible after recruitment closes.
Use a metric dictionary the whole team understands
Count people consistently. For a one-submission-per-participant campaign, count someone who corrects their proof once as one submitting participant, not two new creators. Keep revision workload in a separate count.
| Measure | Definition | Useful denominator |
|---|---|---|
| Joins | Unique participants enrolled in the chosen cohort | Counted invitations only when reliably recorded |
| Submissions | Cohort participants who have submitted proof at least once | Joins |
| Approvals | Cohort participants whose proof is approved | Submitting participants |
| Redemptions | Cohort participants whose issued reward was redeemed | Approvals with a comparable redemption window |
| Review effort | Minutes spent checking proof and resolving questions | Submissions or approvals; label your choice |
Read a small funnel without overclaiming
Hypothetical example: 40 people join, 24 submit, 20 are approved, and 12 redeem by the cutoff. Submission rate is 24 ÷ 40 = 60%. Approval rate among submitters is 20 ÷ 24 = 83.3%. Observed redemption among approvals is 12 ÷ 20 = 60%. These describe participation, not a sales return.
The most useful next question is why 16 joiners have not submitted. Check their remaining time, instructions, and access before increasing recruitment. For the four submitted but unapproved entries, distinguish pending reviews from requested corrections and declines. The same headline rate can hide different problems.
Attach costs to the same cohort
Record setup time, invitation time, review time, fulfillment cost, and any printing or other spend. Separate costs already incurred from the cost of outstanding reward promises. An unredeemed code can still create a future obligation.
If the example cohort costs $68 so far, cost per approval is $68 ÷ 20 = $3.40. Calling that customer acquisition cost would be misleading: the participants may already be customers, and no new purchase has been established. Use the reward-budget calculator for scenarios, then replace assumptions with actual costs.
End with one decision
Copy this memo into your monthly notes: cohort and cutoff; joins/submissions/approvals/redemptions; pending decisions; actual costs; outstanding rewards; most common customer question; next change; next review date.
Choose one next action that follows from the evidence. A slow review queue calls for a clearer staff handoff. Confusing proof instructions call for a simpler brief. Revenue measurement needs additional purchase data and a credible comparison; a busy weekend after launch does not establish that the campaign caused the sales.
Start a cohort record.
Download the campaign tracker from the kit and record the same participants through redemption.
Get the campaign trackerPlanning your spend? Use the reward-budget calculator. Ready to see the product? Walk through the campaign flow.