How to Vet an Influencer Before You Pay: Spotting Fake Followers, Odd Engagement, and Audience Mismatch

The most common regret after hiring an influencer goes like this: the account had well over a hundred thousand followers, yet after the post went up there were almost no inquiries, almost no clicks, and no real customers in the comments. The problem usually is not that the influencer did not try. It is that nobody vetted the account first. Fake followers, odd engagement, and audience mismatch can all be spotted before any money changes hands.

Plenty of business owners say, “I am not a data expert, how would I know?” In fact, most of the warning signs need no tools. Spend half an hour on the account and the outline becomes clear. For anything you cannot see from outside, ask the creator for their account data, and finally confirm everything with one small partnership. This guide walks through those steps in order: the signs you can check yourself, the data to request, how to design a pre-partnership test, and a checklist you can use as is.

Why follower count cannot be the only measure

Follower count is a running total. It reflects how appealing the account was at some point in the past, not how many people will actually see a post today. An account may have grown on one viral video and then changed direction, leaving followers who no longer care. Its followers may also be mostly overseas, which means nothing to a local business.

The other problem is that followers can be inflated. Buying followers, follow-for-follow swaps, and engagement groups all make the numbers look good without producing real purchases. How influencer tiers and fees work is covered in KOL vs KOC: Which Should Your Brand Choose? and The Complete Guide to Influencer Pricing in Taiwan. This guide deals with one thing only: confirming that the numbers you see are real people, and the people you actually want.

Fake-follower signs you can check yourself

Open the account and look at these in order:

• The follower growth curve: some third-party tools show historical follower counts. Normal growth is usually a slow climb, with steps that line up with a few viral posts. A sudden spike during a period with no obvious hit, followed by a slow decline, is worth asking about
• What the followers look like: open a few dozen followers at random and see whether many have no profile photo, no posts, or names that are strings of random letters or numbers
• The following-to-follower ratio: an unusually high following count sometimes means the audience was built through follow-for-follow swaps
• Audience language and location: if most comments are in another language or unrelated to your area, the follower base may not match your market

No single sign proves fraud. Long-running accounts inevitably get followed by bots. Look at whether several signs appear together.

A practical way to keep this fair is to write down what you see for each sign rather than forming an overall impression. Note the period of any sudden jump, roughly how many of the followers you sampled looked empty, and where the commenters seem to be from. When you compare several candidates later, written notes are far more useful than a memory of which account “felt off,” and they give you specific questions to raise with the creator instead of a vague accusation.

How to read unusual engagement

Engagement reflects the real state of an account better than follower count, but it can be manipulated too. Look at posts from the last one or two months, not the pinned highlights:

Is engagement consistent? Normal accounts rise and fall from post to post, but they do not show a few posts with huge numbers while the rest are almost ignored, unless those few covered unusually hot topics.

Do the comments relate to the post? If most comments are generic lines like “So good!” or a string of emojis that could sit under any post, and the same group of accounts keeps commenting, you may be looking at an engagement group or bot comments. Real comments ask questions, mention specific details, and reply to each other.

Do views and engagement fit together? A short video with very high views but almost no likes or comments, or the reverse, lots of likes but ordinary views, is an unnatural combination worth asking the creator to explain.

How do sponsored posts compare with regular posts? This is the most useful check for a brand. Find the creator’s past paid partnerships and see whether engagement is clearly lower than on everyday content, and whether anyone in the comments asks how to buy. If followers ignore sponsored posts, your partnership will probably be ignored too.

When something looks unusual, ask about it before drawing conclusions. A creator may have a perfectly ordinary explanation: a video that was picked up by another account, a giveaway that pulled in a wave of new followers, or a format change that shifted who watches. How they answer is useful information in itself. A creator who can talk through their own numbers clearly is usually someone who pays attention to results, which is exactly what you want in a partner.

Audience mismatch: more common than fake followers

Many partnerships fail not because of fake followers but because the followers are real, just not your customers.

Imagine a brunch café with a single location in Taichung (a city in central Taiwan) hiring a food creator whose followers mostly live in the north. However good the content, few viewers will make a special trip. Or picture a skincare brand aimed at mature customers working with an account whose followers are mostly students. Engagement is lively, but the intent to buy is low.

Check audience fit from three angles:

• Location: this matters most for brick-and-mortar businesses. Are the followers concentrated in the area you serve?
• Age and life stage: do they overlap with your main customers, including their spending power?
• Interest context: does the reason people follow this creator have anything to do with your product? Followers of a creator known for comedy will not necessarily take skincare advice from them

The account data to request before a partnership

What you can see from outside is only the surface. Account analytics are what confirm the audience makeup. Reasonable requests include:

1. Audience profile screenshots: location, age, and gender breakdown
2. Average reach and views on recent posts: specify a recent period rather than letting the creator pick their best single post
3. Results from past partnerships: reach, saves, shares, and link clicks
4. Average Stories views: important if the partnership includes Stories

When you ask, explain that the data is only for evaluation and will not be shared. If you are worried screenshots could be edited, ask the creator to open their analytics live on a video call. Creators who are willing to cooperate usually care more about results. If someone refuses everything, use the test approach in the next section to limit your risk.

Keep the request short and specific. Something like “Before we confirm, could you share your audience location and age breakdown and your average reach over the last month? We only use it to check fit with our customers” is easier to say yes to than an open-ended request for “all your stats.” Asking every candidate for the same items also makes the comparison fair, and it signals that you are evaluating professionally rather than looking for a reason to haggle.

Run a small test before committing long term

Even when the data looks fine, results are not guaranteed. The safest approach is one small partnership first, and let actual results speak:

• Choose a light format: a single Story or one post, rather than a long contract for several videos right away
• Set up tracking: a dedicated discount code, a link with tracking parameters, or a special in-store phrase, so results can be attributed to this creator
• Agree on the metrics in advance: clicks, how many people used the code, or the number of inquiries, rather than likes alone
• Confirm disclosure: even for a test, any paid or gifted partnership should clearly state the relationship. That is a basic responsibility to consumers and platforms; for the principles, see Word-of-Mouth Marketing Compliance in Taiwan

When the test ends, look at the numbers you agreed on before you look at anything else. It is tempting to be won over by a lively comment section or a nice-looking video, but the question you set out to answer was whether this audience takes action on your offer. Also read the comments for intent: are people asking where the shop is, whether the product suits them, or how long the offer lasts? Questions like these are a better sign than praise for the creator.

If the test goes well, discuss a longer or larger partnership. If the results are flat, you have spent only a limited amount and learned how this kind of audience responds to your product.

Pre-partnership vetting checklist (use as is)

Spend half an hour per candidate and tick each item:

• Engagement on posts from the last two months is consistent, with no extreme gaps
• A random sample of followers shows only a few suspicious accounts
• Most comments relate to the content and do not come from the same repeat accounts
• Past sponsored posts show normal engagement and comments, including people asking how to buy
• You have screenshots of audience location and age, and they match your target customers
• You have recent average reach or views, not a single best result
• Past partnerships were clearly disclosed
• The content style does not clash with your brand and there is no history of controversial remarks
• A small test and a tracking method are planned

If a candidate ticks fewer than half, put them on a watch list and hold off on talking price.

Common mistakes

Negotiating on follower count alone. Bigger numbers cost more but are not necessarily more effective. Vet first, and then you have a basis for the price.

Looking only at the highlights the creator provides. Media kits usually show the best results. Scrolling recent posts yourself is closer to reality.

Not recording what you found. Keep every candidate’s vetting results in one spreadsheet. Next time you can compare directly, and over time you learn which kinds of creators suit your brand.

---

The point of vetting is not to find a perfect influencer. It is to put your budget where real people are watching, and where the people watching will actually buy. If you already have a few candidates but are not sure how to compare them, ask NETVANA to help with influencer screening and test planning, or start by looking at our word-of-mouth marketing services.

Further reading: For influencer tiers and when each fits, see KOL vs KOC: Which Should Your Brand Choose?. For estimating partnership fees, see The Complete Guide to Influencer Pricing in Taiwan. For the terms to confirm before signing, see The Influencer Contract Checklist. For spreading the word through everyday customers instead of big influencers, see The Complete KOC Seeding Playbook. For disclosure principles, see Word-of-Mouth Marketing Compliance in Taiwan. After vetting your list and getting ready to send invitations, read How to Run an Influencer Press Trip or Media Event.