Straight answer: the metrics that matter are engagement rate calculated on true reach (not followers), cost per acquisition through a tracked link or code, and saves/shares as a proportion of likes, because those three tell you whether people are buying attention or buying behaviour change. Follower count and total likes tell you almost nothing about sales. If your influencer report only shows reach and impressions, ask for the rest before you pay the invoice.
Why most influencer reports are dressed-up vanity metrics
I spent years on the other side of this. Back when I was a top social media influencer with a following in the hundreds of thousands, brands would send me briefs asking for “reach and impressions.” That was it. No promo code, no tracked link, no follow-up survey. I could post something that got 40,000 views and the brand would call it a win, even though nobody could tell them a single person bought anything because of me.
That’s the uncomfortable bit nobody in this industry likes to say out loud: most influencer campaigns are reported on the numbers that are easiest to pull, not the numbers that mean anything. Reach is easy. Impressions are easy. A screenshot of a follower count going up is easy. Attribution is hard, slow, and often incomplete, so agencies default to the easy stuff and dress it up in a nice PDF with a brand colour scheme.
I’m not saying reach is worthless. I’m saying it’s a leading indicator at best, and treating it as proof of ROI is how brands end up spending five figures a month on influencer marketing with no idea if it’s working.
The metrics that predict a return
Here’s what I now tell clients to track, in order of how much they tell you about real business impact.
1. Engagement rate on reach, not on followers
The standard formula everyone uses is (likes + comments) divided by followers, times 100. Throw that out. It’s wrong because it assumes every follower saw the post, and on Instagram in 2026 that’s rarely more than 15 to 30 percent of a following, even for accounts under 50,000 followers.
The better version: (likes + comments + saves + shares) divided by reach, times 100. This tells you what proportion of the people who saw the content bothered to do anything with it. A micro-influencer with 12,000 followers who I worked with on a client campaign in early 2025 had a reach of 9,200 on her sponsored post and 774 total engagements. That’s an 8.4 percent engagement-on-reach rate, which is excellent. If we’d calculated it on her follower count instead, it would have looked like 6.4 percent, which is still fine, but it hides the fact that almost everyone who saw the post engaged with it.
2. Cost per acquisition, tracked with a unique link or code
This is the one most brands skip because it takes an extra ten minutes of setup. Give every influencer their own UTM-tagged link or their own discount code. No exceptions, no matter how small the fee.
On that same campaign, we spent £450 on the micro-influencer’s fee. Her unique code was used 340 times over four weeks, generating email signups for a lead magnet at a cost per acquisition of £1.32. We were running Meta ads to the same lead magnet at the same time, and those were costing £4.50 per acquisition. The influencer wasn’t just “engaging,” she was outperforming paid ads by more than three times on cost, and we only knew that because of a five-minute code setup.
3. Saves and shares as a percentage of total engagement
Likes are cheap. A save means someone wants to come back to this later. A share means someone is willing to put their own name next to it. If saves and shares make up less than 10 percent of total engagement on a post, that content performed but didn’t persuade. On content that converts, I typically see saves and shares sitting at 20 to 35 percent of total engagement.
4. Follower growth on your own brand account after the campaign
Check your own account’s follower growth in the 48 hours after a post goes live, and compare it to a normal week. If an influencer with real influence over their audience posts about you, you should see a bump you can point to on a graph, not just a vague sense that “things felt busier.”
5. Direct traffic and branded search lift
Pull Google Search Console or your analytics tool and look at branded search volume (people typing your company name) in the week after a big influencer push. A genuine influence spike shows up here even when someone doesn’t click the link in the caption, because plenty of people will just go and google you instead. I’ve seen branded search rise 20 to 40 percent in the week following a well-matched influencer post, even when the tracked link only captured a fraction of that activity. This is one of the reasons I wrote about the wider problem of proving content impact in how to measure your content when it feels invisible, because influencer content has the exact same attribution gap as any other content marketing.
The uncomfortable truth about attribution
Here’s the part that will annoy people who want a tidy answer: you will never get perfect, last-click attribution for influencer marketing, and anyone who promises you clean ROI numbers on every single post is either lying or measuring something that isn’t ROI. People see an influencer post on their phone on the train, don’t click anything, forget about it for eleven days, then type your brand name into Google on a laptop at work. No platform on earth ties that whole journey together for you.
What you can do is build a stack of proxy metrics that, taken together, tell a consistent story. If engagement-on-reach is strong, cost per acquisition on tracked links beats your paid channels, saves and shares are healthy, and branded search moves in the right direction after each post, you have enough evidence to keep spending. If two or three of those four things are flat, stop and ask harder questions before renewing the contract.
I go into more detail on the mechanics of tracking this in the easiest way to increase your influencer marketing ROI, including a free tool for pulling this together without spreadsheets.
A step-by-step process for measuring your next campaign
- Before you agree fees, give every influencer a unique tracked link or discount code, no exceptions, even for a single Reel or a one-off Story.
- Set a baseline in the week before the campaign: normal branded search volume, normal daily follower growth, normal website traffic from social.
- Ask the influencer (or their agency) for reach and impressions per post, not just at the end of the campaign but within 48 hours, when the numbers are most accurate.
- Calculate engagement-on-reach for each post using the formula above, not engagement-on-followers.
- Log saves and shares separately from likes and comments so you can see the ratio.
- Pull your branded search and direct traffic numbers seven days after the campaign ends and compare to baseline.
- Calculate cost per acquisition on the tracked link or code and compare it directly to your paid ad cost per acquisition for the same offer.
- Score the campaign against at least three of these five metrics before deciding whether to repeat it, never on one number alone.
What good looks like, by influencer size
Benchmarks shift a lot by follower count, and this is where a lot of brands compare apples to oranges. Roughly, for engagement-on-reach:
- Nano influencers (1,000 to 10,000 followers): 6 to 12 percent engagement-on-reach is healthy, and cost per acquisition should usually beat your paid channels comfortably because fees are low, often £50 to £300 per post.
- Micro influencers (10,000 to 100,000): 4 to 9 percent is healthy. Fees typically run £300 to £2,000 per post depending on niche and platform.
- Mid-tier (100,000 to 500,000): 2 to 5 percent is normal, and you should expect fees from £2,000 to £15,000 per post, so cost per acquisition needs closer watching because the volume is higher but the intimacy is lower.
- Macro and celebrity (500,000 plus): often 1 to 3 percent engagement-on-reach, sometimes lower, and this is where brand awareness has to be the honest goal rather than direct sales, because the numbers rarely justify a cost-per-acquisition case at this tier.
If a mid-tier influencer’s agency shows you a 15 percent engagement rate, ask which follower count and reach figure they used to calculate it, because that number is almost always calculated on followers, not reach, and it’s making a mediocre result look brilliant.
Where this fits with your wider marketing stack
Influencer measurement doesn’t sit in isolation. If your business is small and you’re doing this yourself without a marketing team, the discipline of tracking codes and comparing cost per acquisition across channels is exactly the kind of work I cover in how I use AI every day as a non-technical business owner, because AI tools can now pull engagement-on-reach calculations and branded search comparisons for you in minutes rather than the hours it used to take me with a spreadsheet in 2019.
If you’re running influencer campaigns at a scale where you need someone senior thinking about attribution across the whole funnel, that’s usually the point where a fractional CMO for the AI era earns their fee, because they’ll build the measurement framework once rather than you rebuilding it for every new influencer relationship. And if you’re deciding whether you need an AI consultant or a traditional marketing consultant to sort your tracking and reporting, I’ve laid out the difference in AI consultant vs traditional marketing consultant.
A quick story that shows why this matters
A client of mine, a small skincare brand, had been working with a mid-tier influencer for eight months, paying £3,500 a post, based on reach numbers alone. The reach looked great, around 180,000 per post. When we added tracked links and pulled branded search data, we found cost per acquisition was sitting at £38, against a £4,200 average lifetime value they were used to seeing from a customer, which sounds fine until you compare it to their email marketing, which was converting at a cost per acquisition of £6. The influencer wasn’t a bad choice, but the brand had been treating “reach looks impressive” as the same thing as “this is our best channel,” and it wasn’t, it was their third best. They kept the influencer relationship but cut the spend by 40 percent and moved that budget into email and a smaller roster of nano influencers with tighter audiences. Six months later, blended customer acquisition cost across the influencer budget dropped by almost a third.
Frequently asked questions
What is the single most important influencer metric to track?
Cost per acquisition through a unique tracked link or discount code, because it’s the only metric that connects spend directly to a business outcome rather than to attention alone.
Is engagement rate still worth measuring in 2026?
Yes, but only if you calculate it against reach rather than follower count, since most reports still use the follower-count version and it flatters small, unengaged followings.
How do I measure influencer impact if people don’t click the link?
Track branded search volume and direct website traffic in the seven days after a campaign and compare it to a normal week, because a real influence spike shows up there even without a click.
What engagement rate on reach counts as good for a micro-influencer?
Between 4 and 9 percent engagement-on-reach is healthy for an influencer with 10,000 to 100,000 followers, and anything above that with a low fee usually means excellent value.
Sources worth reading
The same argument, in conversation: the C-Sweet Talks conversation on vanity metrics and where to spend your time.
