Micro influencer marketing measurement mistakes that survive a board review

What to take away

Counting reach when the decision is about sales

Reach is cheap to collect and easy to defend in a meeting, so it crowds out better numbers. If the decision is whether to renew a creator, reach alone cannot answer it.

Ask what the campaign was meant to change, then pick one primary metric and two supporting ones. A mistake qualifies for this list if it distorts a decision a UK client or board has to make.

For a full method, the measurement and reporting pillar sets out how to match metrics to objectives before any creator is briefed. Reach still has a role as a diagnostic, never as the headline in a renewal paper.

Ignoring disclosure as a measurement input

Disclosure is usually filed under legal, not measurement, which is a mistake. If a creator did not label paid content, the engagement figures describe a different audience response than you think you bought.

The Advertising Standards Authority copy advice service checks campaign copy before it runs, and its output is evidence you can log against each creator.

Keep a per-post column: label present, label wording, date checked, reviewer. When a client asks, the file answers in seconds.

Treating platform dashboards as audited accounts

Platform numbers are directional. They come from the business that sells the placement, and definitions change without notice. TikTok's advertising policies for branded content govern what can be posted, but they are not an assurance opinion on the metrics.

Reconcile platform figures with your own analytics, then state the difference rather than hiding it. A variance line builds more trust than a tidy table.

Letting fake engagement pass into the dataset

The Competition and Markets Authority has published guidance on fake reviews that is worth reading alongside creator vetting. Sudden follower jumps, comment clusters and generic praise are flags. If you cannot explain a spike, exclude the post from the headline number and say why.

Skipping the control group

Without a comparison, every uplift is a story. A holdout group, a matched region or a pre-period baseline gives the number something to stand against. For example, a team paying £400 a month for creator content might see a 6 per cent sales lift, but that means little until it is compared with a region that received no content.

A reporting dashboard makes control comparisons repeatable rather than bespoke, so build the template once and reuse it.

Reporting averages instead of distributions

One strong creator can flatter an average and hide eight weak ones. Report the median, the range and the number of creators who hit the agreed threshold. This shows whether a programme is repeatable or dependent on a single lucky post.

Scoring rubric

Use this before any number enters a board pack. Score each metric from 0 to 2 and drop anything scoring zero on evidence.

A metric scoring below 6 out of 10 should be labelled provisional in the report.

Common questions

How many metrics should a micro influencer report contain?

Three to five, with one primary metric tied to the original objective. More than that and the reader stops knowing which number matters.

Should I include reach if the campaign was about sales?

Yes, as a supporting diagnostic. Keep it out of the headline and out of any renewal decision on its own.

What evidence should sit behind a disclosure claim?

A dated check against the ASA's copy advice guidance, the label wording as published, and the name of the person who signed it off.

Can I report platform figures without reconciliation?

You can, but label them as platform-reported and state the variance against your own analytics. Unlabelled platform figures lose credibility fastest in a review.