Affiliate marketing measurement mistakes explained for UK teams

What to take away

Affiliate marketing measurement means deciding which partner, channel and click gets credit for a sale, then reporting that choice consistently. Most mistakes are caused by inconsistent definitions rather than bad tools.

Are you counting the last click and calling it the cause?

Last-click attribution gives all credit to the final partner in the path. It is simple and it is also the most common source of inflated partner claims.

In a typical journey a price comparison site, a content publisher and a voucher partner may all touch the same buyer. Last-click awards the whole sale to the voucher partner, which then looks far better than it is.

Worked example. A programme records 100 sales and £1,000 of commission in a month. Voucher is the last touch on 45 sales and the first on 5. The content publisher is last on 35 and first on 30. The comparison site is last on 20 and first on 65.

Last click pays the voucher partner £450, the content publisher £350 and the comparison site £200. A rule weighting first touch at 40 per cent and last touch at 60 per cent pays the comparison site £380, the content publisher £330 and the voucher partner £290. The ranking reverses and the total stays at £1,000.

Choose one primary credit rule and one secondary view, and state both in the report. If you need a full method for weighting, the measurement and reporting guide sets out how to document a rule and defend it.

The attribution window causes the same trouble. A 30-day window and a 7-day window will rank the same partners differently, so publish the window next to the credit rule.

Do your tracking links survive consent choices?

Tracking depends on lawful data handling, and consent choices change what you can record. A tag that fires before consent is recorded may be unusable as evidence.

The ICO's step-by-step guide to direct marketing for your small business explains consent, legitimate interests and record keeping for smaller organisations, which covers most affiliate promotions.

Recent reform matters too. The Data (Use and Access) Act 2025 changed parts of the UK data framework, so check your consent wording and retention periods against current law rather than a 2023 template.

Write down what each partner is allowed to see, and check that your partner contracts match the tracking you actually run. Consent rates also vary by audience, so one blended figure hides where tracking is weakest.

Illustrative example: two partners each send 1,000 clicks. At a 90 per cent consent rate the first shows 900 measurable clicks, and at 60 per cent the second shows 600. Audience, banner design and the point at which the tag fires drive the gap.

Are you comparing periods that are not alike?

A common error is putting a peak trading week beside a quiet week and treating the gap as performance. Seasonality, promotion calendars and stock availability all move the numbers.

Fix the comparison before you judge the partner. Match like-for-like weeks, note any campaign overlap, and record what changed outside the affiliate programme. Keep a short note of what else ran, such as email or paid search, so a later reader can judge the comparison.

Where a claim rests on audience or market research, treat the source as evidence with a date. The ASA and CAP publish research at the ASA and CAP which is useful when you need to describe how a market or audience claim was reached.

Do you report clicks, sales and commission in one place?

Splitting clicks, conversions and commission across three files invites contradiction. One dashboard, one definition set, one owner.

Most UK programmes pull the click and commission columns from a network such as Awin, CJ Affiliate or Rakuten Advertising, and the sales columns from their own ledger.

Build the view so a non-specialist can trace a number back to its source. A practical layout is covered in the guide to an affiliate marketing reporting dashboard in England, which is worth reading before you commission another report. Name one owner in the document itself, not in a meeting nobody minuted.

Which measurement mistakes should you fix first?

Use the table to match your situation to a sensible choice. The avoid column lists the shortcut that usually costs more time later.

The table maps each row to the eight mistakes listed at the top. Row one covers mistakes 1 and 2, row two covers mistakes 3 and 4, and rows three to six cover mistakes 5 to 8 in order.

Common questions

How many metrics should a small programme track?

Start with four: clicks, conversions, commission and a return figure. Add metrics only when someone has agreed to act on them each month.

Does attribution choice change the reported return?

Yes. Changing the credit rule redistributes the same commission across partners, so the total spend is unchanged while partner rankings move. That is why two teams can quote different returns from the same month of sales.

Can we rely on platform-reported figures alone?

No. Platform figures reflect that platform's own rules and window. Reconcile them against your own sales ledger before quoting a return.

How often should definitions be reviewed?

Annually, and whenever tracking, consent handling or the partner mix changes. Record the date of each change so old reports stay readable.