Why Influencer ROI Is the Wrong Metric — and What to Measure Instead

Every brand I've worked with in the last five years has asked the same question at the end of an influencer campaign: "What was the ROI?"

It's a reasonable question. It's also, almost always, the wrong one.

ROI tells you what happened. It doesn't tell you why.

When I was leading influencer campaigns at Ogilvy for L'Oréal and Wella Professional, we tracked everything: reach, engagement, sales lift, share of voice. The numbers were good. Sometimes they were great. But the metric that actually predicted whether a campaign would compound over time wasn't ROI. It was trust.

Specifically: did the audience believe that the creator actually used the product?

That sounds obvious. But most brands measure what's easy to measure, not what actually matters.

The number that looks good on a slide deck isn't the number that builds brands

I've seen campaigns deliver 8x ROI on a single activation and leave zero lasting impression on the brand. I've also seen campaigns with modest immediate returns generate customer loyalty that kept paying out for years.

The difference was almost never the creator's follower count. It was whether the brand had chosen a creator whose audience already trusted them on exactly this topic.

Three years ago, I realized we were celebrating the wrong campaigns. The ones with the best ROI weren't always the ones clients mentioned six months later. The ones they remembered — the ones that had actually moved something — were almost never the top performers on the dashboard.

That observation changed what we asked before any campaign launched:

Does this creator's audience actually overlap with our buyer? Not demographically — behaviorally. What do they buy? What do they watch? What do they share?

Does this creator have credibility in this category, or just reach? A creator with 200K followers who is genuinely known for skincare recommendations will outperform a creator with 2M followers who dabbles in everything — every time.

What does success look like six months from now, not six days from now?

That last question is the one most brands skip entirely.

What happened when we stopped optimizing for ROI

One campaign that shifted my thinking involved a mid-size beauty brand trying to break into a crowded skincare category. Instead of casting for reach, we spent three weeks identifying 23 micro-creators — none with more than 80K followers — who had spent years building genuine authority in sensitive skin communities. People who answered comments at midnight, who posted about their own skincare struggles, who their audiences actually trusted.

The immediate ROI was unremarkable. Six months later, the brand had moved from unknown to recommended — organically — in three of the top skincare communities on Reddit and TikTok. That's not something you can buy with a media budget.

The metric I actually care about

If I had to choose one leading indicator beyond ROI, it would be what I think of as earned conversation rate — the percentage of people who saw creator content and then talked about the brand unprompted. In comments, in DMs, in their own posts, in real life.

That number tells you whether the content landed as advertising or as a genuine recommendation. There is an enormous difference between those two things in terms of long-term brand value.

Brands that optimize for ROI get efficient campaigns. Brands that optimize for earned conversation get cultural momentum. Those are not the same thing.

What this means practically

ROI should be the floor, not the ceiling. The question isn't "did we get a return?" It's "did we build something that compounds?"

The brands I've watched win treat creators as relationships, not transactions. They brief them like collaborators. They give them creative latitude to produce content that actually sounds like them. And they measure what happens after the post goes live — not just in the first 48 hours, but in the weeks and months that follow.

That's where the real ROI lives. It's just harder to put on a slide.