Measurement
Part of Does ecommerce SEO measurement need dashboards or evidence?
Before you trust an ecommerce SEO dashboard, check these measurement mistakes
A listicle of ecommerce SEO measurement mistakes English store teams make, from attribution switches to unsourced benchmarks, with a worked pound example.
What to take away
- The most common mistake is reporting platform activity as business outcomes, so sessions rise while nobody can say which product lines gained revenue.
- Set inclusion criteria firstone reporting period, one currency, named data sources, and a written owner for every number.
- Geography matters. Consumer rules are not always identical across England, Scotland, Wales and Northern Ireland, so label where a source actually applies.
- Treat platform announcements as change signals, not results, and keep an evidence trail: a figure that cannot be traced to a source and a date does not belong in a board pack.
1. Confusing traffic growth with store performance
A dashboard showing more organic sessions feels like progress, but it is not evidence of trading improvement. A category page can gain visits from broad informational queries and still convert worse than before.
Traffic vs commercial metrics
Traffic metric
- Organic sessions
- More visits
- Category page
- Broad queries
- Date range
- Same range
- Pair with
- Revenue per session
Commercial metric
- Organic sessions
- Not trading proof
- Category page
- Worse conversion
- Date range
- Same range
- Pair with
- Orders per thousand
Pair every traffic metric with a commercial one from the same date range, such as revenue per session or orders per thousand sessions.
2. Ignoring how attribution changes the story
Attribution models reassign credit between channels, so the same month can look different under different settings. Comparing periods across a model switch measures two different things.
Freeze the model for a reporting cycle and record it alongside the figures. For a fuller treatment of cycles and owners, the measurement and reporting guide sets out how to structure the year before dashboards are built.
3. Reporting platform announcements as outcomes
Search advertising and analytics platforms ship changes continuously. Google Ads maintains a public record of new features and announcements that is useful for spotting what has shifted in reporting.
Reading it is not the same as measuring your store. Note the change, then check whether your own series stays comparable across the period, and flag any reporting break rather than smoothing it over. A break caused by a platform update should be documented, not absorbed into a trend line.
4. Benchmarking against numbers with no stated source
A benchmark is only useful if a reader can find where it came from and what it covers.
Where to source English benchmarks
Use published statistics from the Office for National Statistics, regulator publications, or your own historical data with the period stated. Our benchmark research in England explains how to document sources so a figure survives scrutiny in a budget meeting.
How to label what a figure covers
Say whether a rule or dataset applies to England only, to Great Britain, or to the whole UK, because consumer-facing duties can differ. GOV.UK's help for consumers is written for shoppers, and it is a reminder that reporting should not imply rights or coverage the source does not support.
5. Leaving advertising claims out of the measurement plan
If paid search feeds the same store, your reporting touches advertising compliance. The ASA's AdviceOnline library holds detailed guidance for marketers on what claims and evidence need to line up.
Build a check into the monthly cycle: any promotional claim should have evidence recorded next to the metric it cites.
6. Using illustrative figures without labelling them
Worked examples must be labelled as examples. Consider a store paying £1,200 a month for SEO support and £300 a month for analytics tooling, a combined £1,500. If reported organic revenue is £30,000 and the previous quarter was £27,000, the £3,000 rise is not profit.
Subtract the £4,500 quarterly cost and the contribution is a negative £1,500 before margin. Label every figure as illustrative, and keep real figures tied to a named source.
Common questions
What is the single biggest ecommerce SEO measurement mistake?
Reporting activity rather than outcomes. Sessions and rankings describe work done, not value created, so they should sit beside a commercial measure for the same period.
How often should an English store review its measurement setup?
At least each quarter, and immediately after any platform reporting change. A quarterly check catches broken tracking, model switches and stale benchmarks before they reach a budget decision.
Do I need an external benchmark at all?
Not always. Your own history, clearly dated, is often stronger evidence. External figures help when you have no prior period, provided the source and its coverage are stated.
Who should own the numbers in a report?
Name one person per metric. Shared ownership across a marketing and finance team usually means nobody checks whether the source still supports the claim.
