Audit an influencer’s Instagram comments: signals you can count yourself
Follower counts are easy to inflate and easy to look up. Comments take more effort to fake convincingly, and you can inspect them directly. These are things to check, with the formulas, not a verdict on anyone.
Updated Nov 17, 2026 · 8 min read
What this can and cannot tell you
An export gives you the public comments on a post: who wrote them, what they said and when. It does not give you follower lists, audience location, ad spend or whether anyone paid for engagement. So you cannot prove that comments are fake from the file. You can see whether the comments look like a real conversation, and you can see where they look unusual enough that you should ask the creator questions or look closer.
Real audiences also produce short, repetitive, emoji-heavy comments. Giveaways, “comment for the link” posts and posts in some niches look odd on every metric below and are still genuine. That is why the last step is always comparison: with the same account’s other posts, and with a similar account.
Step 1: export several posts
Do not judge from one post. Pick three to five recent public posts or Reels, a mix of formats, and skip any that were giveaways (they would distort everything). Paste each link into the IGComments tool; up to 200 comments per post are free. Paste the exports into one sheet and add a post column, as in combining exports across posts.
Step 2: the numbers to compute
Column letters assume the standard file: C is username, E is comment, G is likes, H is is_reply. Work on one post at a time first. Ranges end at row 5000, so change them to match your data.
- Comments per commenter. Total comments divided by unique commenters:
=COUNTA(C2:C5000)/COUNTA(UNIQUE(C2:C5000))(Excel 365, Google Sheets). A value near 1 means almost everyone commented once. High values mean a few people wrote a lot, which is normal on a tight community and on giveaways. - Duplicate comment text. In a helper column:
=COUNTIF($E$2:$E$5000, E2). Filter for values above 1. A few repeats of “🔥” are normal. The same longer sentence from many different accounts is worth reading. - Emoji-only or very short comments. A rough test for comments of four characters or fewer:
=IF(LEN(E2)<=4,1,0), summed and divided by the total. For emoji-only, filter the column to rows with no letters; there is no single formula for that, so sample and read. - Generic phrases. Count a few stock comments such as “nice”, “great post”, “love this”:
=COUNTIF(E2:E5000,"*great post*"). Add your own list from what you read. - Reply share.
=COUNTIF(H2:H5000,"yes")/COUNTA(H2:H5000). Many replies, especially from the creator, usually means actual conversation. - Likes on comments.
=AVERAGE(G2:G5000)and the share of comments with 0 likes. Most comments on any account have few or no likes; look at the pattern, not one number. - Timing. A large share of comments inside a few minutes, far from the rest, may be worth a look. Use
posted_at_utcand see timestamps and time zones.
Step 3: read, do not only count
Numbers tell you where to look. Reading tells you what you are looking at. Open the file and read fifty comments from the middle, not just the top. Questions that quote something specific in the post, replies that answer other commenters and comments that mention a detail only a viewer would know are signs of real attention. Comments that could appear under any post, in a different language from the account’s audience, or that follow the same pattern with different usernames deserve a second look. Search for a repeated phrase to see how many accounts used it.
Look at the stats panel too: top words and most active commenters show whether a few accounts carry the post.
Step 4: compare before you conclude
Run the same checks on three posts from this account and on three from a comparable account in the same niche and size range. One unusual number is a question for the creator. A set of unusual numbers repeated across all posts is a stronger reason to ask for more, such as audience screenshots or a past campaign’s results. Do not publish accusations based on a spreadsheet: the file cannot show intent, and hidden or filtered comments are not in it.
Limits of the data
The export covers public comments that Instagram serves. Comments caught by spam filters, hidden-word filters or restrictions are not served, so the file is typically smaller than the displayed count. That works both ways: it can make a comment section look cleaner than the raw one. The file also contains no profile data, so it cannot tell you whether a commenter has followers, a profile photo or a real history. For more on what is and is not collected, see the legal and privacy guide.
FAQ
Can an export prove that an influencer’s engagement is fake?
No. It shows how the public comments look: duplicates, length, timing and replies. Those are reasons to look closer or ask questions, not proof.
What is a normal ratio of comments to commenters?
It varies by niche and post type. Compare against the same account’s other posts and a similar account instead of using a fixed number.
Why exclude giveaway posts?
Giveaway comments are short, repetitive and heavy on tags by design, which would distort every signal.
Can I check an account’s followers with this?
No. The export has comment data only: usernames, comment text, time, likes and replies.
Do I need the influencer’s permission to export comments?
IGComments reads public posts without a login, so you do not need access to the account. It is still good practice to be open with the creator about how you evaluate them.