An affiliate dashboard can show more leads, a stronger approval percentage, and less money at the same time. That is not necessarily a contradiction. The percentage may use a different denominator, the newest leads may still be pending, or the advertiser may have changed the status rules that turn an order into a payable action.
For a cash-on-delivery (COD) campaign, approval rate is useful as an operating metric because it connects traffic to a later business outcome. It is easy to misuse as a standalone verdict. A rate without a numerator, denominator, time window, and status definition is a label, not a diagnosis.
Start with the event the contract pays
“Lead,” “confirmed,” “approved,” and “delivered” can refer to different events. Their meaning depends on the offer and its terms. In the DR.CASH Terms of Service currently linked from its site, the payment section describes payment for confirmed applications and says the advertiser must send complete and timely lead-status information. The document is labeled version 22.06, so treat it as a dated public reference: check the agreement and terms attached to the active offer before using it as a current rule. (DR.CASH Terms of Service).
When someone reports a “30% approval rate,” ask what the platform counted as an approval, which statuses it excluded, whether the figure covers the same GEO and traffic source, and when it measured the cohort. Then compare that definition with the event that earns commission under the current terms.
Two rates can describe the same funnel
The broadest common calculation is:
Lead-to-approved rate = approved leads ÷ submitted leads × 100
A second calculation can help isolate what happens after a customer has been reached or has confirmed the order:
Confirmed-to-approved rate = approved leads ÷ confirmed leads × 100
These measures answer different questions. The first describes how much of the submitted traffic reaches the chosen approval event. The second describes conversion from confirmation to approval. Neither is automatically “the” approval rate. If the offer card, tracker, and account manager use a different definition, document that exact definition and use it consistently.
Do not add mutually exclusive status buckets together as if they were sequential events. “Pending” is not the same as “rejected.” “Confirmed” may be an intermediate state, while “approved” may be a later status or a payment event. Draw the actual status map from the offer’s current documentation and sample records; do not infer it from familiar labels.
Keep the cohort intact while it matures
A daily dashboard often compares leads submitted today with orders submitted last week. That mixes different ages. An advertiser may not have finished calling, validating, shipping, or updating a portion of today’s leads. The newest cohort can therefore show a low early approval rate that rises later. The reverse is also possible if some statuses can be revised after the first confirmation.
DR.CASH’s public terms describe status reporting and network statistics as part of tracking and settlement. A DR.CASH educational article also explains postback setup between a network and tracker. These references support the need to reconcile status data; they do not establish a universal status delay or maturity window for every offer. (Terms of Service; postback setup guide).
Use a cohort table with the same submission date, offer, GEO, source, and creative grouping. Record counts by status at more than one age, such as the first report and a later reconciled report. Choose the observation points from the offer’s actual calling and status-update process; do not assume a fixed number of hours or days applies across advertisers.
A practical record can include:
- leads submitted and unique lead identifiers;
- confirmed, approved, rejected, duplicate, and still-pending counts as the network defines them;
- the cohort’s submission window and the timestamp of each status snapshot;
- traffic source, campaign, creative, GEO, and landing-page version;
- payable event, payout, media cost, and other included costs.
The identifiers must follow the network’s rules and the applicable privacy requirements. Keep personal customer data out of general campaign reports when it is not needed for analysis.
A worked example: denominator changes the story
The following numbers are hypothetical and exist only to show how the calculations differ. They are not DR.CASH results, a market benchmark, or a forecast. Assume both campaigns have fully matured under their own offer terms and that an approved order is payable at $24.
| Measure | Campaign A | Campaign B |
|---|---|---|
| Submitted leads | 200 | 200 |
| Confirmed orders | 90 | 60 |
| Approved orders | 50 | 45 |
| Lead-to-approved rate | 25% | 22.5% |
| Confirmed-to-approved rate | 55.6% | 75% |
| Payout at $24 per approved order | $1,200 | $1,080 |
Campaign A has the higher lead-to-approved rate and more approved orders. Campaign B has the higher confirmed-to-approved rate. If you report only one percentage, you can conceal either the upstream drop or the downstream strength.
Now include the example’s costs. If A spent $700 on media and $80 on other included variable costs, its contribution after those stated costs is $420. If B spent $840 on media and the same $80 in other costs, its contribution is $160. The result does not prove that one campaign is better in general; it shows why approval must be read alongside traffic cost and the rest of the cost base. These figures exclude any cost not named here, so “contribution” is not accounting profit.
A useful expected-value relation is:
Expected payout per submitted lead = payout per payable approval × mature lead-to-approved rate
In this example, A yields $6.00 per submitted lead before costs ($24 × 25%). B yields $5.40 ($24 × 22.5%). If other variable costs are $0.40 per lead, an approximate break-even media cost per lead would be $5.60 for A and $5.00 for B. This ceiling only applies if the payout, approval definition, cohort behavior, and costs used in the model remain representative. It is not a bidding recommendation based on a 200-lead hypothetical sample.
Use the rate to choose the next diagnostic
An approval rate describes what happened. It does not identify the cause. When the rate shifts, investigate the change that could plausibly have affected it:
- Submitted leads fall out before contact: check form quality, duplicate rules, phone-data validation, source mix, and whether the network received the expected identifiers.
- Contact or confirmation changes: compare call coverage, language, local time windows, and customer expectations with the actual offer process. Treat these as questions for the network or advertiser, not conclusions from the percentage alone.
- Confirmed orders fail later: check whether the customer-facing claim, price, product, delivery conditions, and confirmation script describe the same transaction. Ask for the applicable rejection reasons when the system makes them available.
- A change appears in only one segment: compare GEO, source, placement, creative, landing version, and cohort age before changing the whole campaign.
- A change appears across every segment: verify status mapping, postback delivery, account reporting, offer terms, and any operational change with the network.
Change one plausible input at a time where practical and keep a record of the comparison. If several inputs change together, the next report may describe a different campaign rather than explain the original one. Small cohorts also produce volatile percentages: ten approvals out of twenty leads is 50%, but one additional approval changes the rate by five percentage points. Report counts next to percentages so readers can see the sample behind them.
Separate what the affiliate controls from what it can only observe
The affiliate can usually control or document traffic source, targeting, creative, landing-page version, expectation setting, and tracking identifiers. The advertiser or network may control call-center staffing, acceptance rules, product availability, delivery, and status updates. Exact responsibilities depend on the contract and operating arrangement.
This separation keeps diagnosis fair. A low approval percentage does not by itself prove poor traffic, weak calling, fraud, or a broken offer. It tells you that fewer of the tracked leads reached the defined status in the measured cohort. The next step is to locate where the funnel changed and gather evidence from the party responsible for that part of the process.
DR.CASH’s public terms place responsibility on advertisers to provide lead-status data and describe the network’s statistics as the record used for service volumes and value calculation. Because the accessible terms PDF is dated and offer rules can differ, reconcile discrepancies through the current account process and retain the applicable offer terms with the report. (DR.CASH Terms of Service, clauses 8.1–8.2; verify the version governing your account.)
Make the metric useful in a weekly review
A decision-ready review should answer four questions in plain language:
- What is the event? Name the status and payment rule for this offer.
- Which leads are in scope? State the cohort dates, GEO, source, and maturity point.
- What moved? Show counts and event rates beside a comparable prior cohort.
- What evidence supports the next action? Identify the event, responsible party, and one change or verification step.
If those answers are missing, do not scale or shut down solely because a dashboard percentage changed. First verify the denominator and status feed. Then separate an operational signal from a measurement error, a mix shift, or ordinary variation.
Approval rate becomes a business metric when it is linked to a payable event, a mature cohort, and campaign economics. It becomes a misleading headline when its denominator and timing disappear. Preserve the full path from lead to payment, and use the percentage to ask a sharper question about the operation behind it.
