Beyond Engagement (IPA) — influencer marketing payback
Original study- Publisher
- IPA
- Evidence:
- Observed behaviour
- Vendor interest:
- Moderate
- Sample:
- 220 paid influencer campaigns from 144 brands across 36 sectors and 28 markets, covering £133m in spend; sample generally consists of larger brands using marketing mix modelling, and all campaigns had at least some positive influencer result — campaigns where influencer spend produced no result at all are excluded from the databank, so IPA itself says the bottom of its ROI range is significantly under-reported and every ROI index in the report is therefore read as an upper estimate.
- Methodology
- Disclosed
- Geography
- Global
- Published
- Oct 2025
Summary:
This report from IPA, the UK trade body for advertising agencies, looks at whether influencer marketing pays back in sales and how it compares with other ad channels in the short and long term. It was built with agency and analytics partners, including Annalect, Circana, dentsu, Ebiquity, Ekimetrics, Monks, Publicis, the7stars and WPP Media.
IPA pooled marketing mix modelling (MMM) results into an influencer benchmarking databank. It covers 220 paid influencer campaigns from 144 brands across 36 sectors and 28 markets, with £133m in influencer spend. To compare influencers with other channels, IPA linked the databank to Thinkbox Profit Ability 2, which holds ROI data for 10 ad channels. The report does not say which years the campaigns ran.
Key findings:
• Short-term ROI is about average. In the UK, influencers have a short-term ROI index of 99 (100 = all-channel average) and drive 4.5% of short-term sales, close to TV on efficiency. Paid social drives more volume but has a lower ROI.
• Long-term effects are much stronger. The UK long-term ROI index is 151, with 6.2% of sales. The long-term multiplier of 3.35 is the highest of any channel, just ahead of linear TV at 3.27.
• Health & beauty and telco lead by sector. Skin care reaches an index of 132. Telco reaches 199, but this rests on only 10 campaigns.
• Markets look similar. Most land near the all-channel average, from 92 in Germany to 125 in Belgium.
• Media settings don't explain results. Spend share, campaign length and CPM show almost no link with ROI (R² under 0.02), so the report argues that creator fit and content matter more.
Margin notes:
IPA represents agencies that sell influencer work, and the sample only includes campaigns that produced some result, so weak campaigns are likely undercounted. Headline ROI and long-term figures are UK-only, and the long-term results rest on 18 observations. The channel comparison matches two separate databanks rather than coming from a single study.