A scorecard is useful when two people can calculate the same number and reach the same understanding. It becomes misleading when the denominator changes, pending orders are mixed with approved sales or a publisher label is treated as a verdict about quality.
Use the 13 metrics below as a definition sheet. They are not industry targets. Set review periods, currency, order status and publisher classifications before entering values. Compare a result with your own economics and a genuinely comparable period, not an unexplained universal benchmark.
Start with participation and cost
1. Active affiliate rate: partners with at least one approved sale in the chosen 90-day window divided by eligible approved partners in that window. Document eligibility and exclusions. A recently approved partner may not have had enough time to launch.
2. Approved sales per active partner: approved sales value divided by partners active under that same definition. This is an average; it can hide a heavily concentrated program.
3. Network fee ratio: network fees divided by approved sales value on the defined basis. Match the period and relevant contract fee components. Invoice timing can differ from transaction timing.
4. Commission ratio: commission expense divided by approved sales value on a consistent basis. State whether bonuses are included. Do not mix pending commission with approved sales without explaining the difference.
Describe concentration and partner mix
5. Top-ten concentration: approved sales attributed to the ten largest publishers divided by total approved sales. High concentration identifies a dependency to understand, not automatic misconduct.
6. Content share: approved sales attributed to publishers classified as content divided by total approved sales. State the classification rule. This does not prove those orders were incremental.
7. Coupon, loyalty and extension share: approved sales attributed to that defined group divided by total approved sales. Review overlapping categories consistently. The label alone does not establish low value or a policy violation.
8. Subaffiliate share: approved sales credited through defined subaffiliate arrangements divided by total approved sales. Record what partner-level visibility is available. A visibility gap is a question to resolve, not proof of fraud.
Check cohort health and order economics
9. New-cohort activation: partners in a defined new cohort with at least one approved sale within 90 days, divided by cohort members with a full 90-day observation period. Do not compare an immature cohort with a mature one.
10. Year-over-year approved-sales growth: the change in approved sales divided by the comparable prior-year value. Keep credit rules, currency and order-status definitions aligned. If the earlier value is zero, report the percentage as not applicable and show the amounts.
11. Approved average order value: approved sales value divided by approved order count. Explain how partial returns and taxes are handled.
12. Commission per approved order: commission expense divided by approved order count. Read this with order value and margin; a lower number is not inherently better.
13. Disclosure sample pass rate: inspected items meeting your documented review criteria divided by inspected items. Record the sample size and criteria. It describes that sample review, not legal certification of the whole program. FTC guidance emphasizes context and clear disclosure of material connections.
A fictional comparison that changes the conversation
Suppose a fictional program has 100 eligible approved partners and 25 produced an approved sale in the chosen window. Its active rate is 25%. If 40 of those 100 joined very recently, the overall figure does not tell you whether onboarding improved. Review a mature joining cohort separately before deciding that recruiting or onboarding failed.
Likewise, an increase in content share may reflect a change in classification or attribution. Attribution allocates credit under rules. It does not independently establish what sales would have happened without the partner.
Download and add the evidence
Download the editable CSV: /downloads/affiliate-benchmarking-scorecard-2026.csv. It contains the same 13 metrics, blank working fields and definition notes. Add your evidence source, chosen period and target rationale beside each value.
Keep operating facts in separate notes: attribution rules, tracking method, audit scope, review dates and unresolved questions. Server-to-server tracking is an implementation choice to evaluate against requirements, not a universal performance metric or proof of compliance. Finish the review by naming the next decision and the evidence it needs.