Awareness
Review content delivery and available reach or view data. Keep each platform’s definitions visible. Cross-platform totals may count the same person more than once.
Plan influencer marketing measurement across content, traffic, bookings and sales. Define KPIs, tracking, attribution limits and the next campaign decision.
By IRG Media · Published
Influencer marketing measurement starts with the campaign objective and the evidence you can access. Content views, audience responses, website visits and completed actions describe different parts of the journey.
Choose a primary question. Is the audience noticing the content, understanding the offer, exploring it or taking a business action? Name a data owner and collection method before publication.
Review content delivery and available reach or view data. Keep each platform’s definitions visible. Cross-platform totals may count the same person more than once.
Look at relevant audience responses, saves, questions and visits where available. Explain why the selected signals matter to this brief.
Use the brand’s recorded enquiries, bookings, purchases or other agreed actions. Confirm tracking and attribution before describing a campaign connection.
Google Analytics campaign parameters identify referring campaigns. Use consistent source, medium and campaign values; content labels can distinguish creative links. Values are case-sensitive. Naming rules should be agreed across creators before links are distributed.
Name the observable action. A hotel booking confirmation, event ticket purchase and newsletter registration should not be grouped without explanation.
Record where each figure comes from, who supplies it and when it will be available. Keep creator-provided data distinct from brand analytics.
Use a consistent period and definition. Note changes in spend, season, offer, distribution or content format that affect interpretation.
Test the intended landing page and tagged link. Confirm that the expected action can be observed in the brand’s setup before the content goes live.
State the included interactions and the denominator: reach, impressions, views or followers. Rates with different denominators are not directly comparable.
State the cost included and the recorded result used. Creator fees alone give a different view from total campaign cost. Show the underlying values alongside the calculation.
ROI needs a defined financial return and the associated cost. Revenue relative to spend is a different calculation from profit-based ROI. Avoid calling engagement a financial return.
An attribution model assigns credit among touchpoints. The chosen model and window affect the reported result. Attribution alone does not show what would have happened without the campaign.
Separate destination or product interest from recorded actions. Consider journey stage and booking lead time. A link click can show exploration without establishing a purchase.
Distinguish booking-engine visits, enquiries and confirmed stays. Note cancellations, availability and relevant dates where those affect the business question.
Separate views of creator content, visits to release or event information and recorded subscriptions or ticket purchases. Keep launch timing and availability in the interpretation.
Separate content response, product-page visits and recorded purchases. Note changes in stock, offer, season or other activity when reviewing patterns.
A useful report explains what happened, what the evidence can support and what remains uncertain. Label unavailable data. If only content and click data are accessible, keep conclusions at that level.
Record a decision: retain a creator rationale, refine a message, change a delivery step or test a different audience assumption. Keep the observation separate from the proposed explanation so the next cycle can challenge it.
IRG’s performance learning uses the evidence agreed for the engagement. Analytics access, integrations, tracking design and any additional research must be confirmed in scope.
Objective → source → definition → observation → limitation → next decision. Keep that chain visible for every conclusion.
What should we repeat, what should we change and what evidence would change our interpretation?