Measurement
Part of A method for measuring and reporting short form video performance that avoids vanity metrics
Short form video measurement mistakes to avoid
A practical short form video reporting checklist covering view definitions, platform updates, spend denominators, compliance and review routines.
What to take away
A Manchester ecommerce team spending £3,000 a month on vertical video can usually say how many views it bought, rarely how many orders it earned. That gap is the mistake this list is built around.
- Choose one commercial outcome per campaign before any spend, and write it down with a date.
- Keep platform figures in a separate column from your own order or enquiry data.
- Fix the attribution window and leave it alone for at least a quarter.
- Record the geography of every source where available, and apply requirements relevant to the UK audience and placement.
- Set the review date before the first video goes live.
Criteria and geography: these items suit teams and agencies producing or buying short form video for UK audiences, across paid and organic work. They do not rank suppliers, and none comes from product testing.
1. Are you counting views as if they were customers?
Views describe how a platform delivered a video. They do not describe what a buyer did next. TikTok, Instagram Reels and YouTube Shorts each report platform-defined view measures, so their totals are not automatically like-for-like. Thresholds for a view vary by platform and can change, so record the definition used for each report.
Platform views vs checkout orders
Platform-reported
- Views
- Platform-defined
- Conversions
- Platform-reported
- Spend
- £2,000 monthly
- Agreement
- May differ
Checkout data
- Views
- Not a person
- Conversions
- Assisted orders
- Spend
- £2,000 monthly
- Agreement
- Source of truth
Put your outcome beside it: a Leeds retailer paying £2,000 a month for video can compare assisted orders in its checkout data with platform-reported conversions in the same table. If the columns never agree, definitions are the problem, not creative.
For each placement, record the platform metric and its definition, spend, chosen outcome, attribution window and matching order or enquiry data. Keep platform-reported conversions separate from your own commercial results.
2. Do your metrics survive a platform update?
TikTok Ads Manager, Meta Ads Manager and YouTube Analytics organise their reporting differently. If a platform renames or changes the definition of a view metric, a report built on one exported column can break quietly. Keep the raw export as well as the published summary.
Keep reports alive through platform updates
- Keep the raw export
- Keep the published summary
- Version definitions with a date
- Note what changed
- Explain the series to finance
Version your definitions with a date and a note on what changed. If a metric disappears, you can still explain the series to a finance lead instead of restarting it.
3. Are you reporting spend without a denominator?
A single cost per result across several channels hides where the money went. Split spend by placement and give each its own denominator: a thousand impressions, a completed view or a booked call.
For TikTok and Instagram Reels placements, record spend beside the result measure used for that campaign in TikTok Ads Manager or Meta Ads Manager. Keep those figures separate from YouTube Shorts rather than blending them into an average.
For YouTube Shorts placements, label the campaign type and the reported measures in Google Ads. Keep those distinct from YouTube Analytics figures so readers can see which system each number came from.
4. Do you treat compliance as someone else's problem?
Measurement and compliance meet when you report a claim. Advertising claims in the UK are judged against the CAP Code. Any brand also running broadcast or on-demand services sits under media regulation grounded in the Communications Act 2003.
Agree who signs off claims for UK campaigns before a video is scripted, not after it runs. Note in your report which claims were substantiated and by whom.
5. Do you review on a fixed schedule?
Ad hoc reviews produce arguments about anecdotes. A monthly review with a fixed agenda is easier to defend. Put traffic, leads and revenue in that order, and leave creative debate to a separate meeting.
Benchmarks belong in the same conversation. Compare published benchmarks only when the source, audience, format, period and metric definitions are clear. If those details are unavailable, compare like with like within your own campaign history instead.
Glossary
- Viewa platform-defined event, not a person, and not a customer.
- Attribution windowthe period in which a conversion is credited to an exposure; record click and view windows separately and keep the settings consistent when comparing reports.
- Denominatorthe unit you divide spend by to get a cost.
- Blended costtotal spend divided by total results, which hides channel differences.
Common questions
How many metrics belong in a monthly report?
Five or six. One commercial outcome, two delivery metrics and one cost figure per channel is enough, with detail in an appendix.
Weekly or monthly reporting?
Monthly for decisions, weekly for fast fixes such as a campaign spending without delivering. Keep the same definitions across both cycles.
What if platform data and our own data disagree?
Expect it. Platforms credit conversions on their own rules. Report both, label them, and treat your own system as the commercial record.
Do we need a control group?
Not always. If you cannot hold spend back, compare matched periods and say plainly that the comparison is not a controlled test.
