Affiliate Reporting Discrepancies: 8 Publisher Checks
Use this affiliate reporting discrepancies checklist to compare dates, click rules, statuses, currency, adjustments, and exports when publisher totals change.
On this page (8)
What Should You Compare Before Choosing?
Affiliate reporting discrepancies can appear when a network report, publisher dashboard, export, and analytics tool show different clicks, sales, or earnings. Before assuming that data is missing, compare the exact report definitions and settings used for each view.
This checklist focuses on detecting changes over time in one publisher account. It is not an EPC calculator and does not claim that any network recently changed its reporting. Save a baseline, identify what changed, and ask support a question that includes a date range and transaction evidence.
Quick answer: compare the report setup before the totals
Keep the same account, advertiser, date range, time zone, status filters, currency, and export fields when you compare two report snapshots. If a number still differs, isolate the changed dimension and check whether the source system defines the metric differently.
| Check | First comparison | Evidence to save |
|---|---|---|
| Dates | Same start, end, and reporting time zone | Filter settings and export timestamp |
| Clicks | Same gross, unique, or filtered click definition | Network definition and analytics rule |
| Transactions | Same pending, approved, declined, and paid statuses | Transaction IDs and status history |
| Currency | Same report currency and conversion method | Currency field and conversion date |
| Adjustments | Same bonuses, refunds, reversals, and corrections | Adjustment row and reason |
| Export | Same fields, filters, and file version | Original file and column list |
Eight checks for a changed affiliate report
1. Confirm the report window and time zone
Start with the date picker. For affiliate reporting discrepancies, match the beginning and end dates, the time zone, and whether the view groups records by click date, order date, validation date, or payment date. A sale can appear after its referral day, so analytics and transaction reports may use different event dates.
Save a screenshot or settings export with the exact range and time zone. If a report has a saved preset, check whether the preset changed. Do not compare a calendar-day total with a rolling twenty-four-hour window and treat the gap as a tracking loss.
2. Check filters, dimensions, and hidden rows
Record the selected advertiser, campaign, placement, device, region, and transaction type. Some reports show only offers with activity during the chosen period. A filter change can make a program disappear from a table even when its account still exists.
Compare the active filters with the saved baseline. If the platform has a status selector, note its exact value instead of relying on a report name such as “sales” or “earnings.” A screenshot of the filter panel can explain a surprising total faster than a new spreadsheet.
3. Verify what the platform counts as a click
Click totals are not automatically comparable across a network and a site analytics tool. One system might count repeated link taps as separate clicks while another groups visits or filters automated traffic. Check the network's own definition and any account setting that affects duplicate or invalid clicks.
Rakuten Advertising's discrepancy guide explains that its click counts can differ from third-party analytics because systems count repeated clicks and bot traffic differently. Treat that as an example, not a rule for every network. If the definition is undocumented, label the comparison unresolved.
4. Separate attribution date from order date
Check which date each report assigns to the sale. Return-day attribution can credit a publisher's earlier referral after a customer returns directly to the advertiser. A network can therefore show a sale on a date with no same-day publisher clicks.
Look for a transaction ID, advertiser, click reference, and order or validation date. Ask the network which event timestamp drives each column. Do not infer that the customer made an invalid purchase solely because an analytics system credited a later search visit.
5. Match transaction and payment statuses
Compare pending, approved, declined, paid, and reversed rows separately. A report that includes only approved transactions will not equal one that also shows pending actions. A payment statement may omit approved commissions that have not yet reached the payment threshold or payment cycle.
Awin's Publisher Transaction Report guide describes date, payment-status, transaction-type, advertiser, and order-reference filters. It distinguishes pending, approved, and declined states. Use the status label from your export and keep the advertiser's validation period beside it.
6. Check currency and rounding
Confirm the transaction currency, the report currency, and whether conversion occurs at click, order, validation, or payment. Even when two reports contain the same commission, rounding at different stages can create small differences across totals.
Keep the original currency amount and the displayed conversion in separate columns. If a platform changed its default currency or conversion setting, note when it changed and compare the same underlying transaction before and after. Do not estimate a conversion rate from a rounded display value.
7. Find adjustments, reversals, and manual entries
Review bonus, correction, refund, chargeback, and reversal rows. A revised total may reflect an explicit adjustment rather than a changed click or sale. Save the adjustment type, status, date, reason, and transaction reference.
If the platform exports adjustments separately, include that file in the audit. A report with the original commission and a second report with a later reversal can be correct even though the headline earnings differ. Ask support which ledger view contains the final amount owed.
8. Compare export fields and report versions
Preserve the original CSV or spreadsheet, not only a screenshot of the dashboard. Compare the column names, row count, delimiter, empty fields, and identifier values with a prior export. A renamed field or changed default can break a formula or import without changing the underlying transactions.
Keep a dated copy of the export definition or documentation page. If the network does not publish a version number, record the download date, account, filters, and file checksum in your own log. Do not overwrite the older file when the new report arrives.
Run a compact drift audit
Use this affiliate reporting discrepancies procedure whenever a report changes unexpectedly:
- Save the current report and its filters before editing anything.
- Re-run the prior date range with the same account and settings.
- Compare one dimension at a time, starting with the date window and status.
- Select a small set of transaction IDs and trace each into the network ledger.
- Note whether the difference is a definition, an adjustment, a delayed update, or an unexplained gap.
- Ask support about the smallest reproducible example and attach redacted evidence.
- Update the baseline only after the network explains the change or the source documentation confirms it.
The goal is to detect a change, not force every analytics product to agree. If an adjustment recalculated the source report, keep both files so the change date stays clear. This affiliate reporting discrepancies checklist helps separate a definition change from an unexplained gap.
Ask support with a reproducible comparison
Include the account, advertiser, report name, exact date range, time zone, filters, export time, and a few transaction references. State which two totals differ and which metric definition you believe each uses. Ask whether a report rule, status, currency, or default filter changed during that period.
Remove customer details, email addresses, authentication data, and private tracking tokens from screenshots. A ticket should carry only the information support needs to locate the affected records. Keep the response, case number, and any dated documentation update with the report snapshots.
How this differs from comparing EPC offers
When the goal is to compare the economics of two offers, use the separate affiliate EPC comparison worksheet. This reporting discrepancies checklist helps investigate why one report changed or differs from another source. The affiliate earnings per click versus commission per sale guide explains why payout models and click denominators need clear labels.
Publishers who research software offers can review the Rewardful affiliate program, FirstPromoter affiliate program, and Tapfiliate affiliate program listings, then confirm their own live account reports before making an earnings comparison. Browse the FindAffiliates directory to find more programs.
Source confidence and limits
Rakuten's public help article supports the point that click counts and sales attribution can differ between the network and third-party analytics. Awin's publisher report guide documents transaction filters, statuses, and export use. These examples describe those platforms' published documentation; report fields and definitions vary by account and can change.
This audit does not establish that a network changed its rules or that a discrepancy is an error. Use dated account evidence and the current provider documentation to identify what happened.
FAQ
Why do affiliate reports show different click totals?
Systems can use different click, visit, duplicate, and invalid-traffic definitions. Match the report window and compare each system's documented counting rule before treating the gap as missing traffic.
Why does a sale appear without clicks on the same day?
The sale may follow a referral from an earlier day, or the report may use order or validation date instead of click date. Check the transaction and attribution timestamps.
Should pending affiliate commissions count in earnings?
That depends on the report and the decision. Keep pending and approved amounts separate unless the provider defines a combined total. Do not treat a pending transaction as paid.
What should I save when affiliate reporting changes?
Save the original export, report name, filters, date range, time zone, currency, status selection, download time, and a few transaction references. Keep a dated copy of any documentation that explains the fields.
