KPIs by objective
The KPIA metric tied directly to an objective. follows the objective written in the strategy. Picking measures because they are available, rather than because they show whether the objective was met, is the most common reason creator reports convince nobody.
| Objective | Primary KPI | Supporting measures | Main source |
|---|---|---|---|
| Awareness | ReachUnique people who saw an ad. among the target audience | Views, watch time or completion, audience match, branded search in the campaign window | Creator insights, search data |
| Consideration | Engaged actions that show interest | Saves, shares, product questions in comments, site visits, time on site | Creator insights, analytics |
| Sales | Tracked orders or sign-ups | Click-through rate (CTRShare of impressions that result in a click.), conversion rate (CVRShare of visits or clicks that convert.), code redemptions, order value | Analytics, shop platform |
| Content for ads | Paid performance of licensed content | Hook rate, cost per result against other creative | Ad platform |
A KPI tree links these to the business outcome above them. For an awareness programme, the tree might run from new customers down to new visitors from the target audience and branded search, then down to target-audience reach per creator. The tree also stops vanity measures sneaking into the report: a number that does not connect to a branch above it does not belong on the page.
One primary KPI per campaign is the rule. Supporting measures explain why the primary moved or did not; they do not replace it when it disappoints.
Define each measure before the campaign, not after. "Engagement" can mean likes, comments, saves and shares combined, or any subset, and can be divided by followers, reach or views; creators and tools calculate it differently. Write down which actions count and what they are divided by, and ask every creator to report the same way. The same applies to reach, which some platforms report as unique accounts and others as something closer to impressions. A shared definition is what makes two creators' numbers comparable.
UTM links and naming
A UTMURL tags identifying traffic source. link adds campaign parameters to a URL so the analytics tool can attribute the visit to its source. For creator work, each creator, and ideally each post, gets its own link. Without parameters, visits from a creator's post arrive in analytics as social, referral or direct trafficVisits recorded as direct because they arrived with no campaign parameters and no referrer, whether typed, bookmarked or from a lost source., and cannot be told apart from anything else.
A workable pattern uses the source for the platform, the medium for the channel type, the campaign for the campaign, and the content field for the creator and post. The exact values matter less than consistency; the naming conventions and UTMs guide sets the rules the brand should already be following, and creator links should use the same ones. Google Analytics help documents how campaign parameters are read and which reports show them.
Practical points that save data:
- Build links from a sheet, not by hand. A shared builder with fixed values for each field prevents typos that split one creator into several rows.
- Test the full path. Open each link on a phone, inside the platform's in-app browser, and check that the parameters arrive on the final page after any redirects.
- Use a short link where the platform shows the URL. Long URLs look odd in captions and stickers. Check that the shortener passes parameters through.
- Keep a register. Record each link with the creator, post and date, so results can be matched to content later.
Where a programme involves many small creators at once, the tutorial on running a micro-creator programme makes that register part of a single roster sheet that also generates every link and code.
When the goal is an app install rather than a site visit, a UTM link is not enough on its own, and the guide to influencers for app installs replaces it with one device-routing link per creator from the app's measurement partner.
Discount codes and dedicated landing pages
Links capture people who tap through. Many do not. They watch, remember, and buy later on another device, from a search or by typing the address. A creator-specific discount code catches some of those people, because they enter it at checkout whatever route they took. Comparing code redemptions with link-tracked orders is one of the simplest ways to see how much the links miss.
Codes have their own weakness: they leak. Coupon sites collect them, and people who never saw the creator redeem them. Short-lived codes, codes tied to a first order, and monitoring which codes appear on coupon sites all help. Treat redemptions as an indicator, not a precise count of creator-driven sales. The discount code request SOP issues each creator's code with a register line, a checkout test and an end time, so a code found on a coupon site can be traced to the creator it was issued to.
A dedicated landing page does two jobs. It gives the creator a short, memorable address to say aloud, which works for people who do not tap, and it lets the brand shape the page for the creator's audience. The page should continue the creator's message: same product, same offer, similar tone. The message match guide explains why a mismatch between ad and page costs conversions, and the effect is stronger with creator traffic, whose expectations were set by a person rather than an ad. A page per creator is possible for large partners; most programmes use one page per campaign with creator-specific links.
Platform insights and creator reports
Much of the data that matters for awareness and consideration lives in the creator's own platform insights: reach, impressions, views, watch time, saves, shares and audience breakdowns for the specific post. The brand cannot see it unless the creator shares it or a partnership format exposes it.
Make reporting a written condition. The contract should require the creator to share insights for each deliverable within an agreed period after posting, and the brief should say exactly which measures and in what form, usually screenshots of the post insights and the audience breakdown. Ask for the same set from every creator so results can be compared. Where partnership ad formats give the brand access to post data, the platform help centres document what is visible; use it, but do not assume it covers everything the creator can see.
Check screenshots for consistency. Dates should match the posting window and the post shown should be the contracted one. Most creators report honestly, but a standard request makes errors and omissions easier to spot. For reporting on the brand's own social channels alongside creator results, the social media reporting guide covers the organic side.
Data gaps: dark social and screenshots
Some of the effect of creator content is invisible to every method above, and a report should say so rather than pretend otherwise.
Dark social. People share creator posts in private messages and group chats. The recipients may visit the site with no referrer and no parameters, appearing as direct traffic.
Screenshots and memory. People screenshot a product, or simply remember it, and search for it days later. The visit arrives as branded search or direct.
In-app browsers and consent. Platform in-app browsers and consent choices can prevent analytics from recording visits that did happen.
Offline and retail. Where the product is sold in shops or on marketplaces, sales influenced by creators may never touch the brand's own site.
These gaps are why links understate creator impact. Indirect signals help fill them: branded search and direct traffic in the campaign window compared with similar periods, "how did you hear about us" questions at checkout, and marketplace sales trends. They are approximate. The ROI and incrementality guide covers how to estimate the full effect with tests rather than inference, and the influencer measurement hub shows where both guides fit in the brand's wider measurement plan.
Creator reads on podcasts share these gaps with broadcast advertising, and the guide to measuring TV, radio, podcasts and print shows how codes, vanity addresses, search lift and planned on-off schedules can be combined in proportion to the spend.
Reporting per creator and per campaign
Report at two levels. The campaign view answers whether the programme met its objective. The creator view answers which creators to re-book, which content to license for ads, and which tier or style works. Creators who come out well more than once are candidates for a standing deal, and the guide to building long-term creator partnerships covers that decision and the reviews that follow it.
| Level | What it shows | Key columns |
|---|---|---|
| Per creator | Who delivered and what it cost | Deliverables, reach, engagement, link visits, CTR, code redemptions, tracked orders, cost, cost per outcome |
| Per post | Which content worked | Format, reach, completion, saves, clicks, comments of note |
| Per campaign | Whether the objective was met | Primary KPI against target, supporting measures, total cost, indirect signals, notes on gaps |
A per-creator report for a meal kit launch
Pantry Lane, an invented meal-kit company, reports on 6 creators after a three-week campaign. Creator A: reach 48,000, 620 link visits, 41 tracked orders and 18 code redemptions for a £700 fee, about £12 per order counting both. Creator B: reach 95,000, 410 visits, 12 orders and 25 redemptions for £1,200, about £32 per order. Branded search in the campaign weeks runs 30% above the same weeks last year. The team re-books A, tests B's video as a paid ad because its completion rate was the highest, and notes that the codes caught more of B's sales than the links.
Timing matters as much as layout. Pull link and code data at fixed points, such as a week after each post and again at the end of the campaign window, because creator posts keep earning visits for some time and an early read undercounts them. Note the date of every pull in the report so later comparisons use like for like.
Share the creator view with the creators. A creator who sees what worked makes better content next time, and sharing results builds the relationship.
Issuing those links and codes is easiest to get right before launch, and the influencer campaign launch checklist asks for a unique link and code per creator and agreed success measures before anything is posted.
Once verified impressions arrive, working out a sponsorship's CPM lets the deal be compared with paid social and with other creators measured the same way.
Common mistakes
- Reporting what is easy, not what matters. Follower counts and likes fill pages without answering the objective.
- One link for all creators. Results cannot be split, so nobody knows who to re-book.
- Untested links. Redirects and shorteners that strip parameters lose the data before the campaign starts.
- Treating code redemptions as exact. Leaked codes inflate them; use them as an indicator.
- No reporting clause. Without a written requirement, creator insights arrive late, partial or not at all.
- Changing definitions mid-campaign. If engagement or reach is calculated differently from one report to the next, the trend means nothing.
- Hiding the gaps. A report that claims links capture everything will be found out, and the programme's credibility goes with it.
