CommentsToSheets

Guides

How to chart Instagram comments per day

When did the conversation peak, and how fast did it die down? A comments-per-day chart answers that, and a comments-per-hour chart tells you when your audience is awake. Both take a few minutes from an export.

Updated Nov 17, 2026 · 5 min read

Start from local time

The export gives posted_at_utc as text, YYYY-MM-DD HH:MM:SS, in UTC. Convert it once, in column M, called local:

=DATEVALUE(LEFT(F2,10)) + TIMEVALUE(MID(F2,12,8)) + (-5/24)

Change -5 to your offset from UTC (for example 1 for Central European winter time, -3 for São Paulo). Format the column as date and time. Then two more helper columns:

  • N, day: =INT(M2), formatted as a date.
  • O, hour: =HOUR(M2).

Comments per day

Excel: Insert, PivotTable; Rows: day; Values: Count of comment_id. If Excel groups the dates into months, right-click a date in the pivot, Ungroup. Then PivotChart, Line.

Google Sheets:

=QUERY(A1:O, "select N, count(B) where N is not null group by N order by N label count(B) 'comments'", 1)

Select the result, Insert, Chart, line chart.

Comments per hour of the day

Same idea with hour in Rows. Use a column chart with hours 0 to 23 along the bottom. On a post that ran for a week, this shows your audience's daily rhythm; on a post that was mostly one evening, it mostly shows that evening.

Reading the chart

  • A spike days after posting usually means the post was shared, featured or mentioned somewhere. Check the comments on that day for a common source.
  • Replies peaking later than comments is normal: conversation follows the first wave.
  • A flat line that stops suddenly can be the end of a giveaway window, or comments being limited on the post.
  • A chart that only covers the first hours of a post that ran for days means the export did not read the whole thread. Count the rows against the post's count before drawing conclusions.

FAQ

Why not use the timestamps as they are?

They are UTC text. Sorting them works, but a pivot by day in UTC splits your evening across two dates if you are far from UTC.

Can I compare two posts?

Yes: stack both exports with a column naming the post, use days since posting (day minus the post date) in Rows and the post column in Columns.