An influencer campaign report should tell stakeholders what happened, how reliable the evidence is, and which spending decision follows. Separate recorded results from your explanation of them. State what remains unknown. Then recommend a renewal, a smaller test, a pause, or a stop, with a budget and an owner. A report can support that decision without claiming the campaign caused every tracked sale.
Put the decision before the results
Write the question the budget owner must answer at the top. "Should we renew these three creators at the current fees?" gives the report a job. "September influencer results" leaves the reader to find one.
Next, name the rule you will use. A sales campaign might require enough contribution from matched orders to justify another limited spend. An awareness campaign needs a different rule. Agree on it before reviewing the results, rather than selecting whichever measure looks strongest afterward.
Modash's reporting guide recommends connecting metrics to the campaign objective and documenting the attribution model. Both belong in the report. Add a separate line for the decision and the evidence that would change it.
Keep four types of statement distinct:
| Statement | What belongs here | Example wording |
|---|---|---|
| Observation | A recorded result with a source and cutoff | "The order ledger contains these code-matched orders." |
| Interpretation | Your explanation, with its assumptions | "These orders support a limited renewal at the current cost." |
| Unknown | Something the evidence cannot answer | "We do not know how many buyers would have ordered anyway." |
| Action | A choice, owner and spending limit | "The campaign lead will renew Creator A within the agreed cap." |
This is a recommended reporting method, not a platform requirement. It prevents an explanation from quietly becoming a fact.
A completed hypothetical renewal report
Everything in this example is hypothetical, including the dates, creators, costs, order counts and decision thresholds. The example covers organic creator placements with no paid amplification. It uses a custom order-ledger rule, not a platform analytics export.
Decision requested
Renew Creator A for one comparable placement. Defer Creator B until finance checks the margin assumption. Do not renew Creator C at the current fee. Keep the rest of the budget uncommitted.
The purpose of this campaign was to find creator partnerships worth repeating under a contribution-based screening rule. The rule supports a spending choice; it does not establish incremental profit.
Scope and counting rules
| Hypothetical report field | Completed entry |
|---|---|
| Campaign period | September 1 through September 14, 2026 |
| Eligible order period | September 1 through September 21, 2026, UTC |
| Data cutoff | September 28, 2026, 12:00 UTC |
| Primary evidence | Order ledger, discount-code field and campaign cost ledger |
| Matched order | Paid, non-cancelled order using one assigned creator code, with positive retained merchandise value at cutoff |
| Duplicate rule | One row per order ID; each included order has one creator code |
| Retained merchandise value | Item value after discounts and recorded refunds; excludes tax and shipping |
| Campaign cost | Creator fees, allocated staff cost, samples and sample delivery |
| Contribution assumption | Finance estimates 40% of retained merchandise value remains after product and variable order costs, before campaign cost |
| Renewal screening rule | Estimated matched-order contribution must reach at least 1.2 times campaign cost |
This ledger counts orders, not unique buyers or new customers. It excludes orders without an assigned code. Later refunds may change retained value. Code matching records a relationship to the promotion; it does not establish who saw a post or why they bought.
Observed results and calculations
Hypothetical data and calculated estimates:
| Measure | Creator A | Creator B | Creator C | Total |
|---|---|---|---|---|
| Campaign cost | $1,500 | $1,000 | $1,500 | $4,000 |
| Matched orders | 100 | 50 | 25 | 175 |
| Retained merchandise value | $6,000 | $3,000 | $1,500 | $10,500 |
| Estimated contribution at 40% | $2,400 | $1,200 | $600 | $4,200 |
| Estimated contribution minus campaign cost | $900 | $200 | -$900 | $200 |
| Contribution divided by campaign cost | 1.60 | 1.20 | 0.40 | 1.05 |
The calculations are reproducible. Multiply retained merchandise value by the assumed contribution rate. Subtract campaign cost for the remaining amount. Divide contribution by campaign cost for the screening ratio.
For Creator A, that is $6,000 × 40% = $2,400; $2,400 − $1,500 = $900; and $2,400 ÷ $1,500 = 1.60. The campaign total is $10,500 × 40% − $4,000 = $200.
That final amount is an accounting estimate attached to matched orders. Calling it incremental profit would require evidence about what would have happened without the campaign. For a fuller cost treatment, use the campaign cost ledger guide before copying these calculations.
Interpretation and uncertainty
Creator A clears the renewal screen. Creator B meets it with no buffer above the threshold. Creator C falls below it. The total also misses the screening rule, so renewing everyone is unsupported by the agreed criterion.
Test the assumption that matters to this decision. Suppose finance revises contribution from 40% to 35%. This is a hypothetical sensitivity check, not a confidence interval.
| Hypothetical sensitivity at 35% | Creator A | Creator B | Creator C | Total |
|---|---|---|---|---|
| Estimated contribution | $2,100 | $1,050 | $525 | $3,675 |
| Contribution minus campaign cost | $600 | $50 | -$975 | -$325 |
| Contribution divided by campaign cost | 1.40 | 1.05 | 0.35 | 0.92 |
A still clears the screen. B no longer does. This makes checking B's margin more useful than debating which headline sales number looks better.
Three uncertainties remain unresolved:
- Buyers could have purchased without the creator placement. There was no randomized comparison.
- Orders without codes are absent. The report cannot estimate those missing orders by adding a guessed uplift.
- Refunds after the cutoff could lower retained value. The current results are provisional on that point.
Missing orders do not prove that the campaign is undervalued. Existing demand among code users could move the incremental result in the other direction.
Renewal instructions
Hypothetical action register:
| Decision | Owner | Limit and condition | Review date |
|---|---|---|---|
| Renew A once | Campaign lead | Maximum $1,500 total campaign cost for comparable scope; no automatic increase | October 28, 2026 |
| Defer B | Finance lead | No commitment until order-level costs support the 1.2 screening ratio; otherwise renegotiate scope and cost | October 2, 2026 |
| Stop C at current fee | Campaign lead | No repeat purchase at the current terms | September 30, 2026 |
| Refresh original results | Analyst | Reconcile later refunds for the same order cohort | October 28, 2026 |
The recommendation limits exposure while the team learns. It does not predict that A will repeat the same result, or declare C ineffective in every setting.
Keep attribution language within the evidence
Google Analytics defines attribution as assigning credit for actions across touchpoints. Its models differ, and its data-driven methodology includes counterfactual modeling informed by randomized-trial data. Record the actual model and settings when using its reports.
The example above uses code matching. Neither its counts nor a sales increase after posting constitute a campaign-specific randomized experiment.
NIST's randomized-design guidance describes random assignment of treatments to experimental units. That design step is absent from this ledger. Do not add "lift," "caused," or "incremental" to its sales totals. If leadership needs a causal answer before increasing spend, use the incremental-impact planning guide to frame a separate study.
NIST also recommends writing and prioritizing experiment objectives before choosing the design. Decide which spending question that study must resolve before asking for a lift number.
Send the report with an approval request
Before sending, have a colleague reproduce one calculation from the underlying records. Check that observations have sources, estimates have assumptions, and unknowns remain visible. Use the reporting-window guide if the team has not agreed when results become mature enough to review.
End the email with the requested approval: authorize the capped renewal, assign the unresolved checks, and confirm the review date. The stakeholder should be able to approve or reject the next spend without reconstructing the analysis.
