The best time to post on Instagram on Saturday, from 741 posts

Buffer, Sprout Social, Later, and SocialPilot all put Saturday at or near the bottom of the week for Instagram. Sprout doesn't name a Saturday time at all: "Engagement flatlines on Saturdays." In this sample Saturday came second of seven, and, like Friday and unlike Monday to Wednesday, the hour you post on it made a difference that cleared the noise.

The crawl ran on September 22, 2026. Of 5,734 posts from 171 public Instagram accounts, 741 went out on a Saturday, from 162 of the accounts. Each post was scored as its likes divided by its own account's median likes, so a two-million-follower account and a two-thousand-follower account count the same. Scored that way, Saturday's curve starts high in the small hours, drops under normal through the morning, and climbs back above it from 1pm.

At a glance

What is the best time to post on Instagram on Saturday?

The best Saturday block was midnight to 4am ET, at 1.27x an account's normal likes, and it beat the rest of Saturday by a margin that clears. That's 96 posts from 46 accounts. Against the rest of Saturday it's 1.25x, interval 1.02 to 1.53, and against the same four hours on the other six days it's 1.29x, interval 1.05 to 1.60.

Midnight to 4am ET on a Saturday is still Friday night for a lot of people. It's 9pm to 1am in Los Angeles, and it follows the Friday study's strongest stretch, where Friday 8pm to midnight scored 1.20x. Read that way, the Friday night lift carries on past midnight. The raw midnight and 1am hours scored 1.51x and 1.53x on 31 and 15 posts.

What would make this wrong: a few big posts. Only 48% of the block's posts beat their own account's median, about the same share as Saturday as a whole, so the high score comes from a minority of posts that did very well. The block is also small, and US Eastern isn't every account's clock: midnight to 4am in New York is Saturday morning in Europe. Resampling by account keeps one prolific account from carrying it, and it still clears, but by two points at the low end.

Noon to 4pm is the safer Saturday slot. It scored 1.12x on 181 posts, with 52% beating their account's median, and it ran 1.20x the same block on other days, interval 0.99 to 1.40, a point short of clearing. 4pm to 8pm had the highest share of posts beating their median, 55%, but only 75 posts, and it's level with other days' 4pm to 8pm.

The late evening isn't special on Saturday. Saturday evenings (6pm to midnight) ran 1.09x other days' evenings, interval 0.90 to 1.32, on 101 posts. Buffer names 9pm, 10pm, and 8pm, all in the 8pm to midnight block at 1.08x, about where the same block sits on other days.

What is the worst time to post on Instagram on Saturday?

The worst Saturday block was 8am to noon ET, at 0.82x an account's normal likes, and it trailed the rest of Saturday by a margin that clears. That's 181 posts from 78 accounts, at 0.73x the rest of the day, interval 0.57 to 0.90. Only 44% of its posts beat their account's median, the lowest share of any Saturday block.

The smoothed curve bottoms out between 0.87x and 0.88x from 8am to 10am, and the raw 10am hour scored 0.63x on 55 posts. Accounts post a lot into that dip. 10am is the second-busiest Saturday hour, with 7% of the day's posts, and 30% of Saturday posts went out between 8am and 1pm.

Against other days' mornings the dip is smaller. Other days' 8am to noon scored 0.96x, and Saturday against it is 0.86x, interval 0.69 to 1.04, which doesn't clear. So Saturday morning is weak against the rest of Saturday, and mornings in general are a little weak, but this sample can't say Saturday morning is worse than a Tuesday morning.

This is the one place the big studies and this sample agree. Buffer says to "avoid posting on Friday lunchtime and Saturday morning," and SocialPilot says audiences "tend to stay off the gram on Saturday morning."

The early morning before it isn't part of the dip. Saturday posts from 5am to 9am scored 1.03x the 9am to 5pm stretch, interval 0.82 to 1.33, on 144 posts, and the 4am to 8am block scored 1.09x. Later names 5am, which falls there. Hootsuite says to avoid posting before 8am and names 11am, which falls in the weak block.

Read raw, the individual Saturday hours bounce between 0.63x and 1.67x on cells of 8 to 61 posts, which is why the hour curve smooths each hour over its two neighbours on either side. Every smoothed window holds at least 71 posts. The 9pm high of 1.67x raw comes from 8 posts.

Is Saturday a good day to post on Instagram?

Saturday was the second-best day to post on Instagram in this sample, at 1.04x an account's normal likes. Monday was 0.95x, Tuesday 0.98x, Wednesday 0.93x, Thursday 0.99x, Friday 0.98x, and Sunday 1.08x.

None of the day comparisons clears:

The bottom of each interval still matters. Against the weekdays, the interval stops at 0.97, so this sample doesn't fit Saturday being more than about 3% under a weekday for the typical account. That's the claim Buffer, Sprout, Later, and SocialPilot make.

The earlier Instagram study, on the same accounts and about a third as many Saturday posts, had Saturday as the best day, at 1.11x on 258 posts. Two samples put Saturday in the top two. In both, it's a few points clear of the weekdays, small enough that a third sample could reorder them.

Split by account size, the smallest accounts carry most of the lift:

• Under 10k followers: Saturday ran 1.16x the same accounts' other days, interval 1.06 to 1.28, on 342 Saturday posts against 2,011
• 10k to 100k: 1.03x, interval 0.86 to 1.22, on 255 against 1,899
• 100k to 1M: 0.88x, interval 0.58 to 1.26, on 116 against 895
• Over 1M: 0.98x, interval 0.55 to 2.19, on 28 against 188, from 7 accounts

The under-10k tier is the only cut that clears, and it's one of four, so it could be chance. If your account is that size, your Saturdays are the first thing to check.

What do Buffer, Sprout Social, Hootsuite, and Later say about Saturday?

Four of the five big Instagram timing studies call Saturday one of the worst days of the week, and the fifth leaves it out of its best three. Each row quotes the study's own page as fetched on October 2, 2026.

On the day, this sample and the big studies point opposite ways. They put Saturday at the bottom, and here it's second from the top. On the hour, they mostly miss the one block that cleared. None names a time between midnight and 4am, and the two that name a late-morning time, Hootsuite's 11am and the start of SocialPilot's 10am to 2pm, fall in Saturday's weakest block here. Hootsuite also gives 11am on Saturday as its best time for an Eastern audience.

The two methods can disagree without either being careless. The five studies pool engagement across many accounts, and Buffer, Sprout, Later, and SocialPilot say their times apply in the audience's own time zone. Pooled engagement can fall on a day when big or busy accounts post less or reach fewer people, even if the typical account's Saturday post does as well as its other posts. This study scores each post against its own account and reports every time in US Eastern. None of the five pages gives a post count or a margin of error for the hours it names.

How I measured it

This is the same crawl as the Friday, Monday, Tuesday, and Wednesday studies, cut for Saturday. It ran on September 22, 2026 through one Instagram endpoint on AnyAPI, instagram.user_posts, in 657 calls, and cost $0.53. The Saturday cut needed no new calls.

1. Start from the accounts already in the sample. The 211 accounts came from the earlier Instagram study, which found them through 40 creator-niche searches and picked them round-robin across four follower tiers.
2. Pull deeper. instagram.user_posts returns twelve posts per call with a timestamp and a like count, plus a cursor for the next page. Each account was paged up to four times, stopping early once a page reached back past the window. The median account contributed 38 posts.
3. Apply the cutoffs. Posts had to be published between 2025-09-22 and 2026-09-08, so every like count had at least 14 days to settle. Accounts needed five scorable posts. That dropped 1,969 posts as too recent or too old, 10 posts with zero likes, and 40 accounts that fell under five, leaving 171 accounts and 5,734 posts, published between 2025-09-23 and 2026-09-08. 162 of those accounts posted on a Saturday.
4. Score each post against its own account. Divide each post's likes by that account's median likes over the window, and cap the result at 6x so one viral post counts once. 433 posts hit the cap, 58 of them on a Saturday.
5. Bucket and average. Convert every timestamp to US Eastern. Group by weekday, by four-hour block, and by hour, and take the geometric mean of the scores in each group. Hourly values are smoothed over a five-hour window. Every interval in this post comes from resampling whole accounts 1,000 times.

Here's one page of the 2026-09-22 crawl, on one account in the sample:

curl -sS -X POST https://api.getanyapi.com/v1/run/instagram.user_posts \
  -H "Authorization: Bearer $ANYAPI_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"handle":"_garden.ing"}'

{"output":{"data":{"posts":[{"id":"3988678296405120718","createdUtc":1789707527,"likes":804,"comments":217}],"nextCursor":"..."}},"costUsd":0.0008}

The sample: 69 accounts under 10k followers, 63 between 10k and 100k, 32 between 100k and 1M, and 7 over 1M. The typical account's typical post had 112 likes.

What are the limits of this data?

This Instagram study measures likes only, in US Eastern time, on a convenience sample of 171 public accounts and 741 Saturday posts.

• Likes only. Nothing here speaks to reach, saves, shares, comments, or follower growth, and Instagram's ranking leans on sends and saves more than likes.
• US Eastern for everyone. Accounts were found globally and no account was mapped to its audience's time zone. During daylight time, Saturday 2am in New York is 8am in Berlin and Friday 11pm in Los Angeles. A block that clears in Eastern time could be several audiences' different hours added together.
• The best and worst blocks were picked after looking. With six blocks, testing the top one against the rest flatters it. Both cleared, in opposite directions, and the midnight block also beat the same hours on other days, but the low ends of their intervals are close to 1.00.
• A minority carries the midnight block. 48% of its posts beat their account's median, so its score rests on posts that did very well.
• Observational. Content, format, and promotion all confound this. An account that saves its best post for Friday night and goes out after midnight makes the small hours look good.
• Four tier cuts. One of the four clears, which is about what chance alone can produce. Treat the under-10k result as a lead to check on your own account.
• A convenience sample. The accounts came from 40 niche searches a year ago. A different set of niches could shift a result this size.

The Saturday hour gap, best block against worst, is about half again, and its interval runs from a 17% gain to roughly double. The day effect is smaller, about 8% against the weekdays, and it doesn't clear.

What should you change in your Saturday Instagram schedule?

In 741 Saturday Instagram posts, two hour effects cleared and the day effect didn't.

1. Don't drop Saturday because a big study calls it the worst day. Here it was second of seven, and the interval leaves little room for it being much worse than a weekday.
2. Move a Saturday morning post to the afternoon. 8am to noon was Saturday's one weak block, and noon to 4pm was steady at 1.12x, with more than half its posts beating their median.
3. If you want a Saturday slot to test, test just after midnight in your audience's time zone. It's the highest block here, and it clears against the rest of Saturday, but few posts carry it, so check it on your own account before you move anything.
4. Check your own account once it has thirty Saturdays. Pull your posts with instagram.user_posts, page past the newest twelve so you're not scoring posts that are still collecting likes, divide each post's likes by your median, and group by hour in your audience's time zone. Ignore any hour with fewer than about ten posts. Four calls cost $0.0032.

The Friday study found the evening lift that Saturday's small hours continue, the Monday, Tuesday, and Wednesday studies found flat hour curves on those days, and the Instagram study across all seven days uses the same method on the same accounts. The Instagram endpoints are what you'd use to rerun it on your own account.

Use data responsibly and follow AnyAPI's Acceptable Use Policy.