The Affcalculator quality score rates an affiliate cohort from 0 to 100 on what happens after the FTD. It exists because most affiliate reports stop at deposit volume, which is exactly where the real analysis should begin. Volume is not quality; sometimes your biggest affiliate is your most expensive leak.
Eight KPIs, each normalized between a "poor" (0 points) and "good" (100 points) benchmark, then weighted. Fields you leave blank are excluded and the remaining weights rescale automatically, so a short-form score and a full-form score are always comparable.
| KPI | Weight | Poor (0) | Good (100) |
|---|---|---|---|
| FTD to 2nd deposit rate | 20% | 15% | 45% |
| NGR per FTD at 90 days | 20% | €0 | €150 |
| Active at day 30 | 15% | 10% | 35% |
| Bonus cost, % of deposits | 15% | 40% | 15% |
| Registration to FTD rate | 10% | 10% | 40% |
| Early withdrawal rate | 10% | 50% | 15% |
| NGR per FTD at 30 days | 5% | €0 | €60 |
| Wagered per FTD at 90 days | 5% | €0 | €2,000 |
The 2nd deposit rate and 90-day NGR carry the most weight because they are the two numbers that separate players from bonus tourists. The wagered benchmark assumes casino traffic; sportsbook turnover per FTD runs far lower.
Fraud and risk are not KPIs to average away, so they cut the score after weighting: responsible gambling flags reduce it by 4% per percentage point (capped at 40%), duplicate accounts by 5% per point (capped at 50%), and chargebacks by 5% per point (capped at 30%). A KYC pass rate below 85% raises a review flag without changing the score, because verification failure is a compliance conversation, not a discount.
A (75-100): scale, raise caps, protect the terms. B (60-74): keep and review quarterly. C (40-59): renegotiate, usually away from flat CPA. Below 40: a leak; audit before the next invoice.
Separately from quality, the calculator computes total 90-day cost (CPA × FTDs + fixed fees + rev share on GGR or NGR as contracted), ROI against cohort NGR, and payback in months at the 90-day run rate. Quality and deal structure are scored apart on purpose: good traffic on a bad deal loses money, and a rev share deal can make mediocre traffic acceptable because cost tracks value.
The default thresholds are calibrated against published industry data rather than invented. The sources worth reading:
The 7.3% responsible-gambling base rate comes from Auer & Griffiths (2023), who found 7.3% of 37,986 new registrants at a European operator became high-risk within 90 days, and that first-week behaviour predicts it. Cohorts far above that baseline deserve scrutiny regardless of revenue.
Retention and deposit benchmarks reference Optimove's iGaming Pulse (monthly active retention around 70%, average monthly deposits per depositing player of roughly $544 US and $259 globally in late 2025).
Day-30 activity bands (15-25% industry, 30-40% best in class) and composite scoring practice, including weighting registration-to-FTD conversion and chargebacks, follow operator guidance such as Track360's traffic-quality scoring guide.
Bonus-abuse contrasts (high allocation with low wagering, healthy GGR with near-zero NGR) draw on Sumsub's iGaming Fraud Report (bonus abuse is roughly 64% of all iGaming fraud) and EveryMatrix's bonus-abuse analysis.
Every benchmark and weight is editable in the full model, because your market, product mix and margins should override defaults.
This is a screening tool. It works on aggregated cohort metrics, not player-level data; it cannot detect sophisticated fraud like gnoming that survives duplicate checks; and a score is a reason to investigate or renegotiate, never a substitute for doing so. Validate before terminating any partner.