Most tools will show you your comment ratio. Almost none will tell you it runs backwards to reach
Comment ratio runs backwards to reach: 1.00 views per follower under 1 percent, 0.33 above 4. What that means and what it does not. 26 September 2026.
What the numbers say
Here is a number every social media tool can show you and almost none can interpret. We took 2,132 Reels and worked out what share of their engagement was comments rather than likes. Reels where comments were under 1 percent of engagement reached a median of 1.00 times their follower count. Reels where comments were over 4 percent reached 0.33. The relationship runs the wrong way from the advice: more comments as a share of engagement went with less reach, by about three times. Before anyone acts on that, the honest reading is that this is mostly mechanical rather than causal. A Reel that travels far beyond its own audience collects a great many passive likes from strangers, which pushes the comment share down. A Reel that stays inside a tight community collects proportionally more comments. The ratio is telling you which kind of distribution you got, not what to do next.
A high comment ratio is not a failure and a low one is not a win. Read it as a description of your audience: high means engaged and contained, low means broad and shallow. A tool that hands you the number without that sentence has handed you something you can easily misread.
Sample of 2,132, measured on 2026-09-26. Scrptly production database, queried 2026-09-26. 2,132 Reels deduplicated by shortcode, each with over 1,000 views, at least one like, a comment count and its account's follower count.
The measurement
Comment share is comments divided by likes plus comments, as a percentage. Views per follower is the Reel's views divided by its own account's follower count. Reels under 1,000 views are excluded because a ratio over a tiny denominator is noise. The four bands hold all 2,132: 592 plus 360 plus 372 plus 808. Read this as a description rather than an instruction: nothing here says to pursue fewer comments, and doing so would be acting on the arithmetic rather than on the audience.
| Comments as share of engagement | Reels measured | Median views per follower | Top tenth reaches |
|---|---|---|---|
| Under 1 percent | 592 | 1.00 | 23.19 |
| 1 to 2 percent | 360 | 0.58 | 7.02 |
| 2 to 4 percent | 372 | 0.39 | 5.34 |
| 4 percent or more | 808 | 0.33 | 4.81 |
One Reel, beat by beat
The counterexample, included because a rule you only illustrate with agreeing cases is not a rule. One real Reel from the corpus, not a composite: 768 followers, seen 278,304 times, which is 362.38 times the following. Its comment share is 9.44 percent, which puts it in the band that does worst on average, and it reached further than almost anything else in the corpus. It runs 91 seconds. The structure is why: a value loop that runs a full minute and is built to be argued with.
| At | What happens |
|---|---|
| 0:00 to 0:04 | A large claim tying childhood nostalgia to adult behaviour. Absurd on purpose |
| 0:04 to 0:12 | The reframe: you thought it meant one thing, it means another |
| 0:12 to 1:12 | The value loop. Four to six rapid personality readings, name then verdict then next. A full minute of it |
| 1:12 to 1:22 | Payoff on the funniest or most pointed one, with a real reaction |
| 1:22 to 1:30 | A closing line that reframes everything before it, and gives people something to disagree with |
What to do next
- Work out your own comment share: comments divided by likes plus comments. Then read it as a description of who you reached, not as a score to improve.
- If it is high and your reach is low, you have an engaged audience and a distribution problem. Nothing about the comments is wrong; the Reels are not leaving the room.
- If it is low and your reach is high, that is broad distribution doing what broad distribution does. It is not a sign that people care less.
- Be careful with any tool, ours included, that ranks this number as good or bad. The example above sits in the worst band on this measure and reached 362 times its following, which is what happens when you treat a descriptive number as a target.
- One honest limit: these are Reels from accounts somebody chose to analyse, so the corpus skews towards content that already worked. And this is an association across Reels, not an experiment.
- Scrptly is a social media AI tool and it deliberately reports structure rather than scoring your ratios for you. It returns the hook score, the retention score and the beat order for Reels that travelled. The first analysis does not need an account.