Best time to post on Instagram: what 2,149 posts from 47 US brands show

The table shows each 3-hour block in Eastern time, the same moment in Pacific time, how much of the brands' posting happens there, and how those posts did against each account's usual post. Use the row for the moment your audience is awake: if most of your followers are on the West Coast, 6–9 PM ET is 3–6 PM their time.

US brands publish most Instagram posts between 9 AM and 3 PM Eastern. Those blocks are at or slightly below normal. The blocks that beat it are the edges of the day, before work and after it, when fewer brands post and people check their phones.

Each cell combines one weekday with one 3-hour block. Many cells have only 20–60 posts, so read the pattern, not single cells: the 6–9 PM column is green from Tuesday to Thursday, and Thursday and Friday mornings are red.

Brands spread their posts evenly from Monday to Friday and post about half as much on weekends (16% of posts). Tuesday and Wednesday did best, Friday did worst.

Within each niche we compare only the blocks with enough posts. Fashion and tech did best in the evening, beauty in the afternoon and fitness in the morning; home brands were flat. Food & drink's 6–9 PM posts did very well but had too few posts to count.

Sample. 47 public US brand and creator accounts on Instagram, picked as 9 candidates in each of 6 niches (beauty, fashion, food & drink, fitness, tech, home & pets) before we looked at any posting times, so the list cannot favour an hour. Accounts that were private, did not exist, were clearly a different company or had fewer than 20 usable posts were dropped.

Posts. The latest posts of each account (up to 50 on TikTok, 48 on Instagram), read on 2026-10-19. We removed pinned posts, posts under 2 days old (still collecting engagement) and posts older than 12 months, which leaves 2,149 posts published 2025-10-31 to 2026-10-17.

Metric. Engagement = likes + comments, the two counts Instagram shows publicly (saves and shares are private). Posts with hidden like counts are dropped. Instagram's post data does not include a publish time, so we decode it from the post ID, whose first bits are the creation time; scheduled posts are created at the scheduled time. We divide each post's engagement by the median of that account's posts in the sample, so 1.00× is a typical post for that account and 1.50× is 50% more. This lets a brand with 200,000 followers and one with 10 million count the same, and it compares each account with itself rather than big accounts with small ones.

Time. Publish times come in UTC and are converted to US Eastern time (America/New_York, with daylight saving). For each 3-hour block, hour and weekday we report the median score, the share of posts above their account's usual, and a 90% interval from 2,000 resamples of whole accounts, so one account that posts a lot at one hour cannot carry the result on its own.

Reporting rule. A block needs at least 30 posts from 8 accounts to be called a result; smaller ones are shown and marked as small samples. Differences under about 10% are treated as noise. The collection and analysis scripts, the list of accounts and the summary tables are kept with the study; the summary tables are free to download below.

Other studies disagree with each other, mostly because they measure different things: some pool a scheduling tool's customers across countries, some use a survey of marketers, and most compare engagement across accounts of very different sizes. Our 6–9 AM result agrees with Later and SocialPilot's morning findings and contradicts HubSpot's survey, which ranks 6–9 AM as the worst window; our evening result is close to Metricool's.

Start with your own data: in the Instagram app, open Insights → Total followers → Most active times. If your followers are mostly in one time zone, convert our Eastern times to it.

Then test one change at a time. If you post mid-morning like most brands, move some posts to 6–9 AM or 6–9 PM for four weeks and compare each post with your own median, not with your best post. Keep the format the same, so you are testing the hour and not the format.

Do not skip Fridays because of a 10% difference. Consistency, the format and the post itself matter more; use the timing as a tiebreaker for the posts you can schedule.

This shows correlation, not causation. Brands choose when to post, and a slot can look good because better content (launches, collaborations, giveaways) is scheduled there. The early-morning block has fewer posts and a wider interval than the daytime ones.

The sample is 47 established brand and creator accounts with teams, not small businesses or new creators, whose followers may behave differently. Results are for US accounts in Eastern time; we did not check where each account's followers live.

We count likes and comments only. Reach, saves and shares are private, and Reels views are kept separately. Engagement was read once, so older posts had longer to collect it; that affects every hour the same way but adds noise.