How Many Likes and Comments Does a Competitor Post Get?

Like counts differ several-fold between platforms, and comment counts sit below one per post everywhere we measure them. Both patterns are stable enough to plan against, and both are counts rather than rates, so nothing in them depends on an estimate.

Interaction counts are heavily right-skewed. A tracked set of fifty accounts will usually contain one or two with audiences orders of magnitude larger than the rest, and because the metric is an average of per-account rates, those accounts set the figure on their own. Everyone else is described by a number none of them is anywhere near.

The practical test is stability. Recompute an average of these counts month after month and it moves by more than an order of magnitude, tracking which large account happened to be in the sample rather than anything about the field. The median over the same accounts and the same months stays inside a narrow band.

The test is worth applying to any published engagement benchmark, including ones that are not ours. If a source quotes an average of per-post interaction counts with no median beside it and no month-to-month run behind it, there is no way to tell whether the number describes the field or its largest member.

Engagement rate

Interactions on a post divided by an audience figure such as followers, reach or impressions, expressed as a percentage, which makes the denominator the most consequential choice in the metric.

Full definition of Engagement rate

Every published engagement rate is really a claim about its denominator. Divide by followers and you flatter accounts with dormant audiences. Divide by reach and you flatter accounts the algorithm currently favours. For a competitor account, you can see neither number directly, so any rate you compute for them is built on a figure you had to guess.

Counts avoid the problem entirely. A post with 58.6 likes has 58.6 likes whatever the account size, and two posts on the same account are directly comparable to each other. That is why this benchmark publishes interaction counts with their account totals, and leaves rate construction to anyone who has first-party audience data to divide by.

Computing per account before taking the median is the step that makes the figure describe a typical account. Pool every post from every account and the number becomes a description of whichever account published most, which on social data is usually the largest one in the set.

The publication floor is 20 contributing accounts. Below that, a single additional account can move the figure by tens of percent, so anything thinner is labelled an observation rather than a benchmark. Every value on this page is above the floor, with its count printed beside it.

Contributing accounts are counted per metric rather than per platform, which is why the account count is printed next to every individual figure rather than once at the top of the page. A number carries the sample it was computed from, and no other.

Instagram is built for a fast double-tap while scrolling, and a like there is close to free. A LinkedIn reaction appears in a professional feed where colleagues and clients can see it, which makes it a small public act. The counts reflect that difference before they reflect anything about the post.

Two comparisons stay valid. Compare a channel against itself over time, which holds the mechanics fixed. And compare the same channel across accounts, which is what these per-platform figures are for.

Scarcity is what makes a metric informative. Likes accumulate on nearly everything an account publishes, which means they separate posts weakly. Comments happen when a reader has something to say, so a post that pulls comments has usually asked a question, taken a position or touched a problem the audience recognises.

The scale is worth sitting with. A median of 0.8 comments per Instagram post means more than half of tracked accounts average less than one comment on everything they publish. On Facebook it is 0.36 across 54 accounts, so the typical account goes multiple posts between comments. Against that field, a competitor post with twenty comments is not a good post, it is an event.

Shares sit between the two. Facebook shares ran at 1.49 per post and LinkedIn at 1.54, so a share is more common than a comment but far rarer than a like, and it carries the strongest intent of the three because the reader attached their own name to the content.

For competitor analysis this is a shortcut worth using. Rank a competitor's recent posts by comments rather than likes and the list you get is much closer to what actually resonated, which is the list worth studying before you write anything of your own.

Interaction counts are late data. A post published on the last day of a month keeps collecting likes and comments for days afterwards, so a month measured while it is still open understates every account in it. Waiting for the month to close removes that bias.

Figures move between recomputations as new data arrives, which is why a published number belongs to a dated snapshot rather than to a live query. Quote 58.6 likes per post with its 51 accounts and its June 2026 date, and the claim stays checkable long after the underlying aggregate has moved on.