Measurement · Reporting
How to write an influencer campaign report
A good report makes the next campaign easier to decide. Show what you set out to do, what the evidence says, what is still unknown and what you would change next time.
Explore the guideStart with this
Start with the campaign question and the reporting window, then show the evidence that answers it.
Separate observed counts, calculated rates, estimates and figures supplied by a creator or another system.
End with a decision supported by the evidence, rather than a page of unexplained totals.
Write the scope at the top
Name the objective, intended audience, campaign type and included deliverables. State whether the report covers public creator posts, content delivered to your brand, or both. A UGC production report can describe delivered assets and feedback without inventing audience results for content that has not been published.
Specify the publication dates, data capture date and time zone, and the comparison window. If you compare each post seven days after publication, say so. If you simply captured all current counters on one date, say that instead: the older post has had longer to accumulate activity. The campaign's calendar dates and the age of each post are different pieces of context.
Keep an evidence table behind the summary
Create one row per included deliverable. Keep its link or file reference, creator, format, publication date and agreed scope alongside the figures. Record the source for each metric and preserve the creator's original report where they supply private insights. This makes a later correction possible without guessing how the summary was assembled.
Use a consistent spelling for unavailable and explain the gap. A blank can be mistaken for an accidental omission; zero is a claim that no event occurred. If a metric covers only some posts, show the coverage beside it. Record the currency and included cost categories before calculating efficiency.
| Record | What to include | Reason |
|---|---|---|
| Deliverable | Creator, platform or intended channel, format, post/file reference | Makes the unit being evaluated clear |
| Timing | Publication date, capture date, time zone, observation period | Makes comparisons repeatable |
| Source | Public count, native-platform export, creator-supplied screenshot or other named record | Lets the reader inspect where the figure came from |
| Definition | Metric name, denominator and included actions | Prevents unlike numbers being combined |
| Cost | Currency, fee basis, included extras and missing amounts | Keeps efficiency calculations matched to their scope |
Build a summary someone can act on
Lead with the result against the original question, followed by a few supporting figures. Include enough method to make those figures interpretable. A useful summary can be short without hiding missing inputs or turning a count of actions into a count of people.
When calculating campaign engagement rate, sum the included actions and matching views rather than averaging post percentages. For cost efficiency, use the same set of deliverables in the numerator and denominator. A report can show both creator-fee efficiency and a fully scoped cost view, provided they have different labels.
Illustrative report extract · invented campaign data
Three sponsored product demonstrations
Objective: learn which demonstration approach produces relevant product questions. Scope: three videos, observed seven days after each was published, using the same available engagement fields. Creator fees total $1,800; sourcing, products, shipping and any other costs are excluded from this fee-only comparison.
The three posts recorded 120,000 total views and 3,600 engagement actions. The engagement rate by views was 3%; cost per 1,000 views was $15; fee-only cost per engagement was $0.50. These are activity and fee-efficiency measures, not unique reach or profit.
A separate review of 90 selected comments found 18 product-fit questions. The next test should demonstrate fit more clearly and retain the same comment-selection rules. The sample does not establish what all viewers thought, and no sales conclusion is available from these figures.
Separate a reading, a calculation and an estimate
A public view count captured on a date is an observation. Dividing an agreed cost by that count produces a calculation. Classifying comment sentiment introduces an interpretation or estimate. A screenshot of private analytics is supplied evidence whose account, metric and period still need to be identified. Labelling these differences makes the report easier to trust.
The platform may itself estimate a metric. On YouTube, unique viewers is an estimated audience measure; watch time and average view duration are separate from views. Preserve those metric names when copying figures into your report. Do not convert a public view total into unique reach or assume a missing private measure can be reconstructed from it.
Only make sales claims from sales evidence
If your business also measures traffic or purchases through a separate system, identify that system and explain its attribution rules. Attribution assigns credit to touchpoints under a chosen method; a change in the method can change the credit shown. Do not treat a platform click, a website session, a buying-interest comment and a completed order as interchangeable events.
For any revenue or profit statement, specify the event definition, reporting period and cost basis, and explain how refunds or cancellations were handled. If the evidence is unavailable, report that limitation. A campaign can still provide useful content and audience learning without a defensible sales-attribution result. The report structure here is a manual reporting method, not a promise of an integrated sales-reporting service.
Explain what to repeat, change or test
Compare results with the objective and with genuinely comparable work. Separate differences in creative approach from differences in format, audience, post age, spend and distribution. A small campaign rarely isolates all of those factors, so distinguish a promising observation from proof that one choice caused the outcome.
Finish with a small number of concrete next actions. Name the evidence for each one and the uncertainty it should resolve. That is more useful than declaring a campaign successful from a high total alone. Retain the source table so the next campaign can use the same method and the comparison becomes more informative over time.
Illustrative recommendation format
Observation → interpretation → next test
Observation: the demonstration post received repeated compatibility questions in the reviewed comments. Interpretation: the content may not have shown which devices work with the product. Next test: put the compatible devices in the opening demonstration, then review the next campaign with the same comment-selection method.
This recommendation identifies something the team can change. It does not claim that answering those questions will guarantee more sales.
Check the report before sharing it
Read the summary as someone who did not work on the campaign. They should be able to tell what was commissioned, which results are included and why the proposed next action follows. Keep the main report concise and put detailed evidence in an appendix or accompanying sheet.
- The goal, deliverables and intended audience are stated.
- Every headline figure names its unit, period and source.
- Rates use a clear denominator and matching included posts.
- Costs use a stated currency and consistent scope.
- Missing data, estimates and supplied evidence remain visible.
- Total post views are not labelled unique campaign reach.
- Revenue claims, if any, have separate supporting records.
- The recommendation names a practical change and its evidence.