Someone with 200,000 followers just agreed to promote your product for £300. Sounds like a deal. But half those followers were bought in 2023, their engagement rate is inflated by a pod, and the last five brands they promoted got zero sales from the partnership.
You wouldn't know any of that from looking at their profile for thirty seconds. And that's exactly how most brands pick influencers — a quick scroll, a gut feeling, and a bank transfer.
Vetting influencers properly takes about fifteen minutes per creator. That fifteen minutes can save you hundreds of pounds and months of wasted marketing budget.
Check the engagement rate (but don't trust it blindly)
Engagement rate is the most-cited metric for influencer quality. Take the average likes and comments per post, divide by follower count, multiply by 100. Anything above 3% on Instagram is generally considered good. Below 1% and something is probably wrong.
But engagement rate alone is misleading. A creator can have a 5% engagement rate because they're in an engagement pod — a group of creators who like and comment on each other's posts to inflate numbers. The engagement looks real in the numbers, but it's other creators propping each other up, not actual customers.
Look at who's engaging. Open a few recent posts and read the comments. Are they from real people saying real things? Or are they generic ("love this!", "amazing!", fire emoji repeated three times) from accounts that are clearly other influencers or bots?
Real engagement from real people looks messy. Questions about the product. Personal anecdotes. Tagging friends. That's what you want to see.
Look at previous brand partnerships
Scroll back through their last 50-100 posts. How many are sponsored? What brands have they worked with? A creator who promotes a new brand every three days has trained their audience to ignore sponsored content.
Pay attention to what happens when they promote something. Do the comments mention the product, or do they ignore it entirely? If someone posts a glowing review of a skincare product and every comment is about their outfit, the audience isn't interested in product recommendations from this person.
Try to find brands that are similar to yours. If you sell coffee and the influencer has previously promoted coffee or food products, check those specific posts. Did the audience engage with the product or scroll past it?
If you can, reach out to one of those previous brand partners and ask how the campaign went. Most will be honest if you're polite about it. "Did you see sales from working with [creator name]?" is a reasonable question between brand owners.
Audit the follower quality
Follower counts are the easiest metric to fake and the least useful on their own. What matters is the quality of those followers.
Open their follower list and scroll through randomly. Look for:
Red flags: Accounts with no profile picture, no posts, following thousands but followed by few, usernames that are strings of random numbers. A few of these are normal (every account has some ghost followers). A lot of them suggests purchased followers.
Good signs: Followers with real bios, real posts, and engagement of their own. Followers located in your target market (not a random geographic spread). Followers whose interests align with your product.
Some tools can audit follower quality automatically — HypeAuditor, Modash, or Social Blade give rough estimates of fake follower percentage. The free versions give enough data to spot obvious problems.
Check their audience demographics
A creator can have a perfectly genuine audience of 100,000 people, and it still won't work for your brand if those people aren't your target customers.
If you sell women's fashion in the UK and the influencer's audience is 70% male and 40% based in Brazil, no amount of engagement will drive sales. Instagram's Creator Marketplace and TikTok's Creator Portal both show audience demographics — but only the creator can share this data with you.
Ask the creator for a screenshot of their audience demographics before agreeing to anything. Age, gender, and location breakdowns. Any creator worth working with will share this happily. If they refuse or make excuses, that tells you something.
Test before committing
The smartest approach: don't start with a paid partnership. Send the creator free product and ask them to post about it only if they genuinely like it. This accomplishes two things — you see how they create content about your product, and you see how their audience responds to it.
If the organic post gets a good response and you see even a few sales come through, that's your signal to invest in a paid partnership. If nothing happens, you've lost the cost of one product instead of a £500 campaign fee.
When you do move to paid, use tracked links from the start. Give the creator a unique link so you can see exactly how many clicks and sales they drive. With LinkOwl, this takes about two minutes to set up — create a link, send it to the creator, done.
After the first paid campaign, you have hard data. If the creator drove 15 sales from 300 clicks, that's a 5% conversion rate — excellent. If they drove 300 clicks and zero sales, their audience isn't buying and you shouldn't rebook regardless of how pretty the content was.
The minimum checklist
Before paying any influencer:
- Engagement rate above 2% with real, relevant comments from real people
- Previous brand partnerships that aren't constant and show audience interest
- Follower audit — no obvious signs of purchased followers
- Audience demographics match your target customer (age, gender, location)
- Product test — they've tried your product and their audience responded
- Tracked link in place so you can measure actual sales from day one
Skip any of these and you're gambling. Do all six and you'll still occasionally pick wrong — some creators just don't convert despite looking perfect on paper. But you'll catch the obvious duds before they cost you money, and you'll have the data to make better decisions next time.
The goal isn't to find perfect influencers. It's to stop paying bad ones.