Article

Influencer Vetting: 6 Pass/Fail Checks Before You Pay

15 minutes13.10.2026321
Influencer Vetting: 6 Pass/Fail Checks Before You Pay

In a survey of 117 US marketers by EMARKETER and Viral Nation, more than half said they spend 30 minutes or less vetting a single influencer. Half an hour is enough to scroll a feed and read a media kit. It is too little to learn whether the followers are real people in your market. It says nothing about whether the price fits the views you'll actually get.

Influencer vetting is the set of checks you run on a creator before you sign or pay: whether the audience is real, whether it lives where you sell, whether the account is safe for your brand, and whether the rate matches the reach. We run it as six pass/fail checks with the cut-offs written down before we look at the creator, so a good-looking feed can't talk us out of them. Three checks are pure data and run before outreach. Pricing needs a quote, so it comes after first contact. Comments and brand safety need a person reading, so they wait for the shortlist.

The six influencer vetting checks with our pass and walk-away cut-offs. Anything in between goes to negotiation.

Why the media kit can't settle it

A media kit is the creator's sales document. The follower count is real in the sense that the platform shows it. The engagement rate is often an average of posts the creator picked. The audience screenshot comes from a date range the creator chose. None of that is lying; it's the best case, the way a résumé lists only the wins.

Bought audiences are common enough to check every time. In a HypeAuditor study of about 2.9 million Latin American Instagram accounts with 1,000+ followers, about 1 in 12 had at least half of their audience flagged as suspicious. The study covers one region and comes from a company that sells audits, so take the exact share as an indication. The split by size is the useful part: 8.8% of micro accounts (10K–50K followers) and 8.2% of nano accounts crossed that line, against 0.7% of accounts above a million. In that sample roughly one micro account in eleven crossed the line. Your market may differ, but it's reason enough to check every time.

There is a legal side in the US too. Since October 21, 2024, the FTC's rule on reviews and testimonials has made it unlawful to buy or sell fake followers, views, likes or comments that the buyer knew or should have known were fake, when they misrepresent influence for a commercial purpose. The rule binds the person who buys them, usually the creator. Your exposure as the advertiser comes from the FTC's endorsement guidance, which says plainly that "your company is ultimately responsible for what others do on your behalf". A vetting record is the cheapest evidence that you had a reasonable process.

1. Count the followers who are real people

This check goes first because it kills the most candidates for the least effort. You want one number: the share of the audience that is real, active people, as opposed to bots, abandoned accounts and mass-followers who follow thousands of profiles to farm follow-backs.

You can't get that number by scrolling the follower list. A bot with a profile photo and three posts looks like a person on a phone screen. You need an audit that classifies the whole audience and compares it with accounts of the same size. In Yoloco every creator report carries an audience quality score from 0 to 100 with five classes, Bad, Low, Normal, High and Best, built from bot detection, suspicious follower spikes and engagement authenticity. On Instagram it adds the hidden likes rate and the share of notable users among followers.

Our cut-off: Normal or better passes. Low means the creator stays on the list only if the price comes down to match the real audience (check 6 shows the math). Bad is a walk-away whatever the content looks like.

A hypothetical audit that fails on three checks at once: low audience quality, a growth spike and an audience mostly outside the brand's market.

Take Nora Vance, a made-up baking creator with 84K Instagram followers, pitching a made-up US cookware brand we'll call Larkfield. Her feed is lovely and her kit says "mostly US". The audit puts her audience quality at 38, in the Low class. Her Reels average 6.2K views, which is thin for 84K followers. On its own that's a negotiation point. Checks 2 and 3 settle it.

2. Read the growth curve, especially the jumps

Follower growth from real content is lumpy. A creator posts something that travels, gains a few thousand followers over a week, then settles. Bought followers arrive differently. Thousands land in a day or two with no post behind them, and comments and views stay where they were.

Open the follower chart for at least the last 12 months and look for four shapes. A steady slope passes. A spike that lines up with a viral post passes too; open that post and check it really did travel. A spike with no post behind it fails, and so does a sharp drop, which often means the platform removed fake accounts in a purge.

Nora's chart shows +19K followers in three days in March and no post that week with more than her usual views. That is about a fifth of her audience arriving at once. We'd stop here, but in a real review it's worth running the remaining checks anyway, because you'll want the full record if a colleague asks why she was dropped.

If you're screening at scale, Yoloco's search lets you filter by follower growth over the last three months, which catches both rising creators and suspicious jumps before you open a single report.

3. Check the audience lives where you sell

Creator location and audience location are different numbers, and the second one decides whether your money reaches buyers. A creator in Austin can have most of her followers in Brazil because one Reel went viral there two years ago.

The test is the share of the audience in the country (or region, or language) your offer works in. How strict to be depends on the offer. A US-only promo code needs a high share, because every follower outside the US is reach you can't sell to. A global app or a brand-awareness push can live with a lower one.

Our cut-off for location-bound offers: at least half the audience in your market passes, under a third is a walk-away, and the band in between goes into the price (check 6). Nora's audit shows 41% of her audience in the US, so she lands in the negotiation band on this check alone.

Two practical notes. First, ask for the audience report from a tool, since a screenshot from the creator's own insights is one date range on one post. Second, filter by audience geography before outreach: in Yoloco you can set location, language, age and quality for the audience separately from the creator, so "US audience" stops depending on what the bio says.

4. Read the comments, then compare the engagement rate

Engagement rate is the check people run first and should probably run fourth. On its own it's easy to misread. Rates fall as accounts grow and differ by platform and format. Worse, a creator in an engagement pod can post a healthy-looking number every time.

Start with the comments on the last 10 to 12 posts. Real comments respond to the post or to another commenter. Pod and bought comments are single emojis, generic praise ("Love this!!") that could sit under any post, and the same few dozen accounts showing up within minutes of every upload. HypeAuditor's definition of a suspicious audience includes exactly these: empty or repetitive comments and participation in engagement-exchange groups.

Then look at the engagement rate across those posts. A real account wobbles: a recipe carousel does better than a selfie, a collab does better than a repost. A rate that sits almost flat across every post, whatever the content, points to coordinated engagement. Compare the average with creators of the same size on the same platform, never with a universal "good" number. The free engagement rate calculator shows where a creator sits against platform benchmarks from numbers you type in, and our guide on how to calculate engagement rate covers which formula to use where.

Pass: comments that talk about the content, and a rate in the normal range for the creator's size that moves from post to post. Fail: comments you could paste under anyone's photo, or a rate so flat it looks scheduled.

5. Brand safety, past sponsors and how they disclose

This is the check that needs a person with the sound on, and it catches the problems that get a campaign pulled. A clean audience report can't tell you that the creator mocked your category last spring.

Do three things. Go through six months of content, with sound for video. Search the creator's name with your competitors' names and words like "scam", "drama" or "apology". And list every sponsor from the last 90 days. A creator who promoted your direct competitor last month is either locked by an exclusivity clause or about to look like a billboard to their audience. Ask; most will tell you.

While you're listing sponsors, look at how the ads were disclosed. On Instagram, creators paid in any form, including gifted products, are expected to add a Paid Partnership label. The FTC adds that a platform tool may not be enough on its own and that the disclosure should sit with the endorsement itself. A creator whose past ads carry no label at all will probably do the same with yours, and the FTC expects you to monitor that.

Pass: disclosed ads, no direct competitor in your exclusivity window, nothing in the archive you couldn't defend to your boss. Fail: ads with no disclosure, a competitor deal inside the window, or any post you'd have to explain upstairs.

For high-profile deals (a celebrity, a long ambassador contract, a regulated category like pharma or finance) a person-level background check from a reputation-screening firm is worth its fee. For a $600 post it isn't; the six-month read is enough.

6. Price the post on views in your market

The last check turns the first five into money. Follower count is the wrong base for a price, because it includes the fake, the inactive and the people in other countries. Price on what you actually buy: views, and specifically views from your market.

The formula is two lines:

CPM on views        = fee ÷ average views × 1,000
CPM in your market  = fee ÷ (average views × share of audience in your market) × 1,000

Here is how it plays out for Larkfield on three hypothetical creators, all quoting for one Reel.

Hypothetical quotes for one Reel. The cheapest CPM on views becomes the most expensive once you count only US viewers.

Creator A asks $1,200 and averages 30,000 views: a $40 CPM, the best of the three. But only 40% of her audience is in the US, so Larkfield gets about 12,000 US views and pays $100 per thousand of them. Creator B asks $600 for 10,000 views, a $60 CPM, with 80% of the audience in the US: $75 per thousand US views. Creator C asks $2,500 for 50,000 views at 70% US, which comes to about $71.

Use the average views of recent posts, never the best one. Take the audience share from the audit. Then compare the quote with an outside reference. Yoloco's creator report includes a fair-price estimate for exactly this; whatever tool you use, our cut-off is a quote within about 1.5× of the estimate. Above that, the creator needs a reason you can name: usage rights for paid ads, exclusivity, a deadline, or a format that takes days to shoot. If there's no reason, counter with the in-market CPM math.

What changes on TikTok, YouTube and Telegram

On each platform a different number stands in for reach, and fraud hides in a different place.

Which number to judge reach by on each platform, and what to watch for.

On TikTok, the For You feed pushes videos far beyond a creator's followers, so engagement per follower looks inflated. Judge by engagement per view, and look at the median of the last 10 videos, because one viral clip can carry a whole account's average. TikTok analytics in Yoloco reads engagement by views for that reason.

On YouTube, subscriber counts age badly. A channel that grew on one hit years ago can keep hundreds of thousands of subscribers while new videos get a few thousand views. Price on views of the last few months of uploads, and check that the audience geography matches the language of the videos. The YouTube channel analytics page covers the same checks for integrations.

On Telegram there are no likes to inspect and placements are usually paid up front, so average views per post and reactions do the work, together with the channel's mention network: which channels repost it and which it reposts. Channels that only ever appear in each other's feeds are worth a second look. Telegram channel analytics shows subscriber growth, average views, reactions and audience geo without admin access to the channel.

On Instagram, add the hidden likes rate to check 4 and use Reels views for check 6. A creator who won't share Reels insights can still be priced from public view counts.

Who runs influencer vetting, and in what order

Run the checks in the order that kills candidates cheapest. Checks 1 to 3 are data: with a tool, they take a few minutes per creator, and they can run on a list of 50 before anyone writes a pitch. Check 6 needs a quote, so it comes after first contact. Checks 4 and 5 need a person reading content and go last, on the shortlist only.

Thirty minutes per creator is fine when it goes to the five who passed the data checks. Spread across thirty unscreened names, it buys a scroll and a guess. In Yoloco's influencer search you can apply audience quality, audience country and growth as filters across 400M+ profiles, so the list you start from has already cleared most of check 1 and check 3. Our article on AI influencer marketing covers which of these steps an AI agent can take over and which it can't.

Keep a record for every creator you pay. In the EMARKETER and Viral Nation survey, 96.6% of marketers said they want documentation of how influencers were vetted and only 25.6% always get it. If you work with an agency, ask for this. If you are the agency, send it before the client asks:

Creator:            name, handle, platform
Date checked:       YYYY-MM-DD, tool used
1 Audience quality: score and class                 PASS / NEGOTIATE / WALK
2 Growth:           spikes and what explains them   PASS / NEGOTIATE / WALK
3 Audience market:  % in target country             PASS / NEGOTIATE / WALK
4 Comments and ER:  notes, ER vs same-size peers    PASS / NEGOTIATE / WALK
5 Brand safety:     sponsors last 90 days, issues   PASS / NEGOTIATE / WALK
6 Price:            fee, CPM in market, vs estimate PASS / NEGOTIATE / WALK
Decision:           go / counter at $___ / drop, and why

FAQ

What is influencer vetting?

Influencer vetting is checking a creator before you pay them: whether the audience is real, where it lives, whether the content is safe for your brand, and whether the price fits the views you'll get. It sits between finding creators and signing contracts.

What percentage of fake followers is acceptable?

There's no official threshold, and tools measure "suspicious" differently, so compare scores within one tool. On Yoloco's 0–100 quality score we pass creators in the Normal class or above. Low goes to negotiation; Bad is a walk-away. In HypeAuditor's Latin American study, about 1 in 12 accounts had half or more of their audience flagged, so that level is far past any reasonable line.

What are the best tools for influencer vetting?

You need two kinds. An audience analytics tool covers checks 1 to 3 and 6, plus the engagement-rate half of check 4: audience quality, growth, geography, engagement and pricing (Yoloco, and several others, do this across platforms). For check 5 on high-stakes deals, reputation-screening firms run background checks on the person. For most campaigns a careful read of six months of content does that job.

How long should vetting take?

With an audit tool, the data checks take a few minutes per creator. The human read of six months of content is where the time goes, more so for long-form video. If that sounds like too much, run fewer creators through it by filtering harder first.

Can I vet an influencer for free?

Partly. You can read comments, review content, list past sponsors and work out the CPM yourself at no cost, and the Instagram audit page has a free engagement-rate calculator with no login. Audience quality and audience geography need a tool with data; in Yoloco the full audit runs on a free account.

Should I still pay a creator whose audience is outside my market?

Depends on the offer. For a global product or pure awareness, a mixed audience is fine if the price reflects it. For anything location-bound, price on views in your market (check 6); if fewer than a third of the audience is in your market, we'd walk away.

What to do this week

  1. Write your cut-offs for the six checks into your brief template before you look at the next creator, and get your manager or client to agree to them.
  2. Take the last three creators you paid and run checks 1, 3 and 6 on them. If one fails, you've found your benchmark for why this matters.
  3. For your next campaign, filter by audience quality and audience country before outreach, and send pitches only to creators who pass.
  4. Ask for average views of the last ten posts with every quote, and counter with the in-market CPM.
  5. Save a vetting record for every creator you pay, using the template above.

Similar articles

Find creators who actually move your audience

Define any audience, get the creators who really influence it — with engagement and audience quality up front. Free to start.

Sign up freeNo card needed