Sends per reach is the share of people who saw your Instagram post and sent it to someone in a DM, and the head of Instagram has named it as one of the three signals that matter most for ranking. The other two are watch time and likes per reach. That is the closest thing to a published formula Instagram has ever given, and it comes with one detail most summaries drop: which of the three matters depends on who you are trying to reach.
We sell Instagram followers, likes and views, so weigh the last section accordingly. The rest is what Instagram has said on the record, and where the record stops.
What Instagram has actually said
Three statements, in the order they were made.
In 2023 Instagram published its own explanation of ranking. It lists signals for each surface separately. For Reels it says the majority of what you see is from accounts you do not follow. For Explore it says popularity signals matter much more than they do in Feed. It names likes, saves, shares and comments, and it ranks none of them against the others.
In mid 2024 Adam Mosseri, who runs Instagram, got more specific. As Search Engine Journal reported, he said: “One of the most important signals we use in ranking is sends per reach. So out of all the people who saw your video or photo, how many of them sent it to a friend in a DM?”
In January 2025 he narrowed it to a list. Social Media Today quoted him: “The top three signals that matter most for ranking are watch time, likes and sends.” He told creators to look at average watch time, likes per reach and sends per reach in their own insights.
Then the detail. Likes count slightly more for connected reach, meaning people who already follow you. Sends count slightly more for unconnected reach, meaning everyone else.
Why it is a rate and not a count
All three signals are stated as ratios. That is the part worth sitting with.
A post seen by 1,000 people and sent by 30 has a sends per reach of 3%. A post seen by 100,000 and sent by 500 has 0.5%. The second has far more sends. The first is the stronger post, and by Mosseri’s description it is the one the system reads as worth showing further.
This is why a big account and a small one compete more evenly than follower counts suggest. Reach is the denominator. A small account is judged on what a small audience did.
It is also why chasing raw totals misleads. A post can collect an impressive number of likes simply by being shown widely, and still be a weak post per viewer.
How to find yours
Instagram shows the two numbers you need on any post.
- Open the post and tap View insights.
- Take the number of shares, shown with the paper plane icon.
- Take accounts reached.
- Divide shares by accounts reached. 24 shares on 1,600 reached is 1.5%.
Do the same with likes for likes per reach, and note average watch time on Reels.
We are not going to give you a benchmark for a good sends per reach, because Instagram has not published one and the figures in circulation come from tools measuring their own customers. The comparison that means something is your own posts against each other. Sort your last twenty by sends per reach and look at what the top five have in common.
Why a send is worth more than a like
Instagram has not published weights, so treat this as reasoning from what it has said.
A like costs nothing and is seen by nobody. A send costs a small piece of the sender’s reputation. It says to one particular person: this is worth your time. That is a much harder thing to get, which is what makes it informative.
A send also does something a like cannot. It puts the post in front of a new person who was picked by someone who knows them. For a system trying to work out who else would want to see a post, that is close to a direct answer.
The same logic sits behind what we wrote about saves versus likes on TikTok. Cheap signals are common and say little. Costly ones are rare and say a lot.
What gets sent
Mosseri’s own advice was not to force it, and he is right that “send this to a friend” as a caption works about as well as “please like”. What gets sent is content that is about the recipient more than the creator.
- The “this is you” post. A behaviour, a habit, a type of person. The viewer thinks of someone at once.
- The useful thing for a specific person. A recipe for the friend who cooks, a route for the friend who runs. Narrow beats general here, because narrow suggests a name.
- The plan. A place to eat, a trip, an event. A send is how two people decide to go.
- The shared reference. A joke that only works if you both know the thing.
- The news inside a niche. Something a group chat would want to know before everyone else does.
What rarely gets sent: announcements, product shots and anything whose subject is the account that posted it.
Watch time and likes still count
Sends get the attention because they are the newest of the three to be named. They are one of three.
Watch time was listed first. Metricool’s 2026 study of 24.4 million posts found average watch time on Reels more than doubled in a year to 8.5 seconds, so the bar is moving. A Reel that loses people in the first two seconds never reaches the point where anyone would send it.
Likes per reach matter slightly more with your existing followers. If your problem is that followers have stopped seeing your posts, likes are the signal to work on. If your problem is finding new people, it is sends.
Format plays into this as well. The same study found carousels earn nine times the saves of single images. We compared the formats for the other platform in TikTok carousels vs video.
What the list leaves out
Three things, and each is a gap in the record. None of them is a finding.
Comments were not in the top three. That does not make them worthless. It means Mosseri did not name them.
Saves were not in the top three either, despite years of advice that saves are the signal that counts on Instagram. The 2023 explanation lists them. The 2025 list does not.
And there are no weights. You will see claims that a send is worth some exact multiple of a like. We could not find Instagram saying that anywhere, so we have left the number out.
Where buying likes and followers fits
We sell Instagram services, so read this with the appropriate suspicion.
Run the arithmetic from above. Bought likes raise the like count on a post. They do not come with a send, and they do not come with watch time. On the two signals that reach new people, they add nothing.
Bought followers are worse on this measure. A post goes to a sample of your followers first, and followers who never respond add to reach without adding likes or sends. Every rate on this page has reach as its denominator. We go through that in does buying Instagram followers hurt your account.
What bought numbers do is change how a profile or a post looks to the next visitor. That is cosmetic, and Meta’s rules prohibit the transaction, as set out in the risks of fake engagement.
The lever that moves the signal is the content. A post about the viewer’s friend gets sent. A post about you does not.
Frequently asked questions
What is sends per reach on Instagram?
The number of times a post was shared in a DM, divided by the number of accounts that saw it. Adam Mosseri has described it as one of the most important signals Instagram uses in ranking.
What are the top Instagram ranking signals?
Watch time, likes and sends, according to Mosseri in January 2025. He told creators to track average watch time, likes per reach and sends per reach.
What is a good sends per reach?
Instagram has not published a benchmark. Compare your own posts with each other instead, and study the ones at the top of the list.
Do shares to Stories count as sends?
Mosseri’s description was specifically about sending a post to a friend in a DM. Instagram has not said how a share to a Story is counted, so do not assume the two are equal.
Are saves or sends more important?
Sends are in the top three Mosseri named and saves are not. Instagram’s 2023 explanation lists both without ranking them. On what has been said most recently, sends.
Does asking people to share a post work?
Mosseri advised against forcing it. A prompt works when it is specific and the content fits, such as “send this to whoever is booking the trip”. A general request to share does little.
The short version
Instagram’s head named three ranking signals in January 2025: watch time, likes and sends. He told creators to track each as a rate against reach.
Likes count slightly more with followers. Sends count slightly more with people who do not follow you, which makes sends per reach the number to watch if you want to grow.
It is a rate. Thirty sends on a thousand reached beats five hundred on a hundred thousand.
Instagram has not published weights or benchmarks, and it did not put saves or comments in the top three. Anything more exact than that comes from somebody other than Instagram.
People send posts that are about the person they send them to. Make those, and measure your own posts against each other.
