Instagram Watch Time in 2026: Not a Solo #1

Is Watch Time Instagram’s #1 Ranking Signal?

The sentence people shorten into “#1” is a list of three. Social Samosa, dated 23 Jan 2025, quotes Mosseri from the first video in his video series: “The top three signals that matter most for ranking are watch time, likes and sends.” Watch time is named first in that list. The sentence still says top three.

Hootsuite’s July 15, 2026 guide (Alyssa Hirose) restates the same idea for 2026, with Mosseri as the source line: “For both, the same signals matter most: watch time, likes, and shares (or sends).” “Both” on that page is connected reach (people who follow you) and unconnected reach (people who don’t). Hootsuite then tells you to watch engagement rate, average watch time, likes per reach, and sends per reach.

What Did Mosseri Name, and What Do 2026 Recaps Add?

The January 2025 video is the sentence the recaps keep citing. Social Samosa writes that users should focus on average watch time, likes per reach, and sends per reach. On connected vs unconnected, that page prints two lines. First: “For connected content, likes have slightly more influence, while sends carry more weight for unconnected content.” Next: “If users aim to increase their visibility with current followers, likes are the most important.” Slightly more influence and “the most important” are both on Social Samosa. This article cites both. It does not pick one.

Eclincher, dated July 24, 2026, writes: “In January 2025, Mosseri publicly confirmed the three ranking factors Instagram weighs most heavily, and they still define 2026.” The next sentence on that page is Eclincher’s ranking: “Watch time is the single most important signal across surfaces.” The same page also writes that “Mosseri has said sends carry roughly three to five times more weight than likes when deciding whether to push content to non-followers.” Social Samosa’s quote of the January video is the top-three sentence: watch time, likes, and sends. The multiplier lives on Eclincher. The list of three lives on Social Samosa.

Sprout Social, published January 28, 2026, by Nader Nekvasil, leads its “top ranking signal” section with a different emphasis: “Sends Per Reach: Private sharing (DMs) is now arguably the strongest signal of value.” The same section says a View “is counted for every instance a post appears on a screen, including repeat views.” Elsewhere on that page: “Instagram now rewards content that is shared privately (Sends) more than content that is simply Liked.”

A sibling page already walks Mosseri’s comments on the ranked feed. That Mosseri watch page is the place for the chronological-feed argument. For the publisher-by-publisher ranking of the signal lists themselves — Hootsuite’s five-item line vs Sprout’s Sends-above-Likes framing — see the Instagram algorithm tips 2026 page. This page stays on watch time, the claim this URL used to lead with.

What Does Meta’s Ranking Card Measure?

The recap pages argue about a #1. Meta’s own Feed card argues in predictions. Instagram Feed, updated June 29, 2026, lists “How much time you spend on a post” among the signals for whether you are likely to enjoy a post. It also predicts “How likely you are to spend more than 10 seconds on the first post,” and “How likely you are to share a post with someone in a direct message.” Those are the nouns on the card. Hootsuite’s “spend more than 10 seconds on the first post” line matches that prediction. It is a first-post prediction, not a network-wide watch-time score.

Explore uses different clocks. Instagram Explore, updated June 22, 2026, predicts “How likely you are to spend more than five seconds viewing a post” and “How likely you are to watch more than 95% of a video.” The same card ranks candidates on “likelihood of user engagement such as like and save.” Five seconds on Explore is not the Feed card’s 10 seconds on the first post. 95% of a video is not average watch time in Insights. If you paste one “watch time” target onto every surface, you are mixing three fetched sentences.

How Do You Read Average Watch Time Across Accounts?

Hootsuite’s Insights line is per account: average watch time, likes per reach, sends per reach. A network dashboard that averages those tiles hides the login that is actually holding attention. Read the tile on the login that posted. If one account’s average watch time is the number you are managing, say so. Do not publish a blended “network watch time” as if Instagram scored the group.

Hootsuite also writes that the Feed algorithm “limits back-to-back posts from the same creator and blends content from followed and recommended accounts.” The noun in that sentence is one creator’s posts, stacked in a feed. A variant per login is how you get a separate average-watch-time read, instead of pasting account A’s Insights onto account B.

If distribution dies, Hootsuite’s Mosseri Reels basics say “Check Account Status to make sure your account is in good standing.” That is a restriction check, not a watch-time diagnosis. Start with Account Status, then the free Instagram shadowban checker, then the shadowban avoidance guide if you are trying to tell a ranking miss from a recommendation limit. Reels-specific ranking changes live on the Reels algorithm update.

1. 1

   Read the three Insights tiles Hootsuite names

   Average watch time, likes per reach, and sends per reach. Social Samosa’s January 2025 recap names the same three. Do this on the login that published the post.
2. 2

   Do not crown a winner the recaps do not share

   Eclincher says watch time is the single most important signal. Sprout says DM sends are arguably the strongest signal of value. Hootsuite keeps all three in one sentence. Write down which page you are using before you kill a format.
3. 3

   Keep Views off the watch-time tile

   Sprout’s View counts every time a post appears on a screen, including repeats. That is the primary metric Sprout describes. It is not Hootsuite’s average watch time tile. The Views-first page is the counting explainer.
4. 4

   Use the clock that matches the surface

   Feed card (June 29, 2026): more than 10 seconds on the first post, plus time spent on the post. Explore card (June 22, 2026): more than five seconds, and more than 95% of a video. Do not average 10 and 5 into a house target.
5. 5

   Check Account Status before you rewrite the hook

   Hootsuite’s Mosseri list says check Account Status so the account is in good standing. A restriction and a weak average watch time are different problems. The checker is the outside-in read. Account Status is the in-app one.

Edit Tests That Match the Fetched Sentences

Hootsuite’s Reels tip on the same July 15 page says keep them short (under 90 seconds), “open with a strong hook in the first three seconds, and loop them to stretch viewing time.” That 3-second hook and the loop are Hootsuite’s advice on that page. Judge the test by the average watch time tile Hootsuite names, not by whether the Reel matched the under-90 tip or the “3 minutes or less” line in Mosseri’s list on the same page.

Sprout’s sends line is the other edit test: private sharing is the signal that page calls arguably the strongest. A caption that asks for a like trains the like tile. A post someone would actually forward trains the send tile Sprout is describing. Hootsuite still keeps likes in the top three. Social Samosa says likes have slightly more influence for connected content, and also that likes are the most important if you want visibility with current followers. If the login is talking to people who already follow it, keep the like-worthy post. A sends headline is not a reason to delete it.

Operators who need the same test on several logins are not looking for a shared watch-time hack. They need each account’s own device and its own audience graph so the Insights tile is actually that account’s. The Veridia case study is the public example of running that kind of network. The setup we use is dedicated real phones, isolated environments (own device and own IP per account), years of experience running accounts at scale, and proprietary sequences built on hours of experimentation. That is the ceiling on how the system is described here.

The ranking-card stamps, and what shipped beside them, are on the algorithm update page.