Photo: Liza Summer, Pexels
Blog · Measuring influencer marketing
Influencer marketing builds brands. Stop judging it like a promo code.
Promo codes show what an influencer campaign sold in its first few weeks, but the IPA’s first influencer study says the bigger return comes later. Here’s how to judge it in the short term and the long term.
The trouble with promo codes
Promo codes are the most honest number in influencer marketing, because they count real sales from real people, influencer by influencer, and that is exactly what makes them dangerous.
A promo code only sees the people who bought straight away with the code, so it can’t see the person who remembered your brand 3 months later and paid full price. Influencer campaigns end up with what I call the four-week verdict: judged on what sold before the code expired, and booked again, or not, on that alone.
The industry is leaning further in: in Linqia’s survey of more than 200 US marketers, conversions as a top-3 measure of success rose from 35% to 46% in a year while engagement rate fell from 69% to 48%, so we’ve swapped likes for sales, but both are still four-week numbers.
Even the sharpest minds feel the pull: in 2018 Mark Ritson paid micro-influencers to post a pixelated photo of his backside as art, to show how little follower counts say about influence, and his fix was click-through rates, which is another four-week number.
The most important word is “and”
In 2013, Les Binet and Peter Field went through 996 campaigns from 700 brands in the IPA’s databank for The Long and the Short of It, and found that advertising does 2 jobs that pay back over very different timescales:
- Sales activation (offers, discounts, “buy now”) spikes fast and fades.
- Brand building (being known, liked and remembered) starts slower, keeps building and pays back more.
The best results came from roughly 60% brand building and 40% activation, with both measured, because “short term metrics alone can be misleading”. Even then, the IPA was warning about “very short-term online metrics” becoming the main measure, and influencer marketing grew up in the decade that followed.
It was never a case of long beating short, or as Mark Ritson puts it, “the most important word in the long and the short of it is ‘and’.”
Influencer marketing is a slow burner
In October 2025 the IPA published its first influencer database, covering 220 campaigns from 144 brands in 28 markets, and scored every channel’s return on investment against an all-channel average of 100.
Judged at 4 weeks, influencer marketing is bang on average, but given time it beats the average by about half.
There is one honest caveat, which is that 18 campaigns is a small base and the IPA’s director of effectiveness, Laurence Green, calls it “still early days”. Even so, it suggests influencer marketing is being marked before the exam is over, and that spend didn’t decide the winners either, as the IPA points instead to the fit between brand and influencer and the quality of the content.
Testing pays early, measuring pays later
Binet and Field have a habit of spoiling a comfortable assumption, and here they compared campaigns that pre-tested their adverts with campaigns that kept tracking their effects, counting which went on to report very large sales effects.
Testing gets you a better first month, while measuring campaign after campaign is what pays off over the years, so you need both, which brings us back to “and”.
How to measure influencer marketing ROI
Binet and Field closed their study with a balanced scorecard, so here is mine for an influencer campaign, with where Justify comes in.
In the short term: before launch and the first few weeks
Will people know it’s you?
Only 27% of influencer adverts link clearly to the brand, System1 reported in December 2025, so check each video shows the brand early and sticks to the brief.
In Justify: run a creative test on each influencer video before you approve it, and here’s how to test an influencer video.
Was it seen?
Reach, views and engagement for every post.
In Justify: views and engagement for every post and every influencer, in one place.
Did it sell?
Promo code and link sales for each influencer, treated as the floor rather than the full count.
In Justify: connect your Shopify or WooCommerce shop and see what each influencer sold from their promo code or tracked link.
In the long term: the months after
Did people’s view of the brand change?
Compare awareness and consideration among people who saw the campaign with people who didn’t, and do it every campaign to build the track record Binet and Field found pays off.
In Justify: Brand Lift studies, with real people who saw your campaign and people who didn’t (see an example Brand Lift report).
Are people choosing you at full price?
Over 3 to 12 months, look at searches for your brand name, sales after the promo code ends and whether you can hold your price, because Binet and Field found it’s brand building, over time, that makes people less sensitive to price.
Are you building with the right influencers?
The IPA points to fit, so keep the ones who worked and find more like them.
In Justify: every influencer’s views, engagement and sales side by side, so you can book the ones who worked, and here’s how to decide which influencers to book again.
And once a year, if you spend enough to model it, look at the whole mix with econometrics, which Justify will cover when its marketing mix modelling arrives (it’s coming soon).
So next time someone asks what the campaign sold, ask what it will still be selling next year, because that is the long and the short of it.
Quick answers
Does influencer marketing work in the long term?
The early evidence says yes, as the IPA’s 2025 study gave it a longer-term score of 151 against 77 for paid social, on a base of 18 UK campaigns.
Are promo codes a good way to measure influencer marketing?
For the short term, yes, but a promo code only counts the people who bought straight away with the code, so it misses anyone the campaign persuaded to buy later, or at full price.