Key Takeaways
- Instagram replaced “impressions” with “views” across the whole platform in April 2025. Any article still talking about impressions is out of date.
- Adam Mosseri named watch time, likes, and sends as the top three ranking signals in January 2025. Comments are no longer in that top three.
- Sends per reach is the signal that pushes your content to people who don’t follow you. Likes per reach moves it to people who already do.
- Every metric maps to a specific content fix. That is the part almost nobody teaches.
- The native Edits app is free, built by Instagram, and gives you every number you need. You don’t need a third-party tool.
- Instagram shows every new video to a tiny test group first, sometimes as few as five people. A low view count is not always a verdict on your content.
Instagram Insights Explained: Why Most Creators Read Them Wrong
Most creators open their Insights, see a number, and feel either good or bad about it. That is it. No action. No decision. Just a vague feeling and then they close the app.
That is the wrong approach.
Instagram Insights is not a report card. It is a diagnostic tool. Every metric tells you something specific about what went wrong or what went right. And once you know what each one means, you can fix it.
That is what this post is. Metric by metric. With the specific action you take after each one.
Let’s crack into it.
Do You Need a Special Account to See Instagram Insights?
Yes. Insights are only available on Creator and Business accounts. Personal accounts do not have access. Data collection starts the day you switch and is never backdated, so switching today means data starts today and not last month.
According to Instagram’s own Help Center, the professional dashboard is where all Insights live. If you are still on a personal account, switch now. You are flying blind.
What Tool Should You Use to Read Your Metrics?
Use the Edits app. It is built by Meta, native to Instagram, free, and lives on your phone. It shows you views, accounts reached, shares, saves, likes, comments, and more. All in one place. No subscription required.
Most articles on this topic assume you are paying for a third-party dashboard. You don’t need that.
Open the Edits app, tap Insights at the bottom right, and change the date window from 30 days to 7 days. That is your weekly snapshot. Instagram’s own creator resources walk you through exactly this workflow.
Set aside 30 minutes once a week to run this review. Monday morning is a solid choice. Check the last seven days, not the last 30. You want to react to what is happening now, not patterns from a month ago.
Anywho. Let’s get into the actual metrics.
What Happened to Impressions on Instagram?
Instagram removed impressions and replaced them with views across the entire platform in April 2025. Any tool still showing impressions is pulling stale data. Views is the only metric now, and it is unified across every content type.
The Meta Graph API v22.0 changelog confirmed that impressions was deprecated for all media and user insights effective April 21, 2025. Meta followed up in August 2025 to confirm the same change would hit the Page Insights API by November 2025. The migration is complete.
The new unified metric is views. Views count every time a video starts playing, including replays. If someone watches your Reel three times, that counts as three views.
For Reels specifically, Instagram defines views as the number of times a Reel starts to play or replay, including time spent on replays. Watch time is the total time a Reel was played across all of those plays.
Stop looking for impressions. They are gone.
What Does Watch Time Tell You?
Watch time is the most important metric in your Insights. It tells you whether your video held attention from beginning to end. Low watch time means one of two things: a hook that lost people in the first seconds, or something in the middle that caused them to scroll away.
Here is what most people miss. Watch time includes replays. If your video is genuinely interesting, people watch it again. Each replay adds to the total. So watch time is not just about video length. It is about whether you made something worth watching more than once.
The fix for low watch time is pretty straightforward. Make your video more interesting from beginning to end. Cut slow sections. Open on something that creates immediate curiosity. The retention chart, which Instagram added in August 2025, tells you exactly where the drop-off happens, so you are not guessing.
Videos that stall at a few hundred views often have decent hooks but loose middles. People started watching and had no reason to stay.
What Is Skip Rate and Why Does It Matter?
Skip rate shows the percentage of viewers who scrolled past your Reel within the first three seconds. Instagram added it in August 2025, replacing the older view rate metric. It is now the primary hook diagnostic for individual Reels.
If skip rate is high, people are not stopping for your hook. The video itself might be great. The opening three seconds are not doing the work.
Check skip rate on individual Reels, not just your overall account stats. A Reel with a high skip rate and a low view count is sending you one clear message: fix the hook. Start faster. Cut the preamble. Open on the most interesting moment. For a breakdown of how to structure a Reel opening that stops the scroll, see Instagram Reel editing tips for beginners.
What Does the Retention Chart Show You?
The retention chart maps viewer drop-off across the full length of your video. Instagram added it to Reels in August 2025. The line always slopes downward. That is normal. The question is how steep the slope is and whether there is a sharp drop at a specific timestamp.
A gradual, flat slope means high relative retention. People stayed. A sharp drop at a specific moment tells you something went wrong there. A slow section. A confusing transition. A pacing break.
Use the retention chart alongside watch time. Watch time gives you the average. The retention chart tells you exactly where that average broke down.
What Do Shares Tell You?
Shares tell you your video was valuable or entertaining enough that someone wanted another person to see it. That is a high bar. People only forward content that is freaking good or freaking relevant to someone specific they know. It is a deliberate action.
Low shares usually means one of two things. The video was interesting to the viewer but not useful enough to pass along. Or it was useful, but only to them, not to anyone they know.
The fix: ask whether your video solves a problem that exists for more than one person. If a viewer watches your video and immediately thinks of a friend who needs to hear this, that is a share. If they watch and think “that is just for me,” that is a like. Both matter, but shares have far more growth potential.
What Is Sends Per Reach and Why Does It Matter?
Sends per reach is the metric that pushes your content to people who have never heard of you. It is super duper key for growth, and it became the most underrated metric on the platform after one specific moment.
In January 2025, Adam Mosseri stated: “The top three signals that matter most for ranking are watch time, likes and sends. So when looking at your insights, pay close attention to average watch time, likes per reach, and sends per reach. Likes are slightly more important for connected content, and sends are slightly more important for unconnected content.”
When someone DMs your post to a friend who does not follow you, that is a vote for unconnected reach. The algorithm takes it seriously. It treats that send as evidence your content is worth showing to strangers.
Likes per reach, by contrast, moves your content in front of people who already follow you.
Hold up. What is connected vs. unconnected reach exactly?
Connected reach is your existing followers. Unconnected reach is everyone else. Every account that has never seen your content before. The algorithm uses different signals to decide how much of each you get. For a deep look at how it makes those decisions, see how the Instagram algorithm really works.
Do Comments Still Matter for Growth?
Wait, Max. Mosseri listed watch time, likes, and sends. Comments are not on that list. Are comments done?
Comments still matter. They are just not in the top three ranking signals anymore.
The old playbook from 2022 to 2024 said to optimize for comments above everything else. Comment pods. Comment bait. “Drop an X below for my free guide.” That worked in a different version of the algorithm.
Now, sends per reach carries more weight for growing your audience beyond your existing followers. Comments are still a signal that your content made people think. They help. But if you are focused on one lever, optimize for content people want to DM to someone else.
The clearest path to high sends: make your video so targeted and useful that a viewer immediately thinks of one specific person who needs to hear it. That is the creative brief for every piece of content.
What Do Saves Tell You?
Saves tell you that your video was valuable enough that someone bookmarked it to come back to later. The social component is smaller here. This is personal. When someone saves, they are saying: I am not done with this yet. That is a strong signal that you gave them something worth keeping.
Low saves usually means your content was interesting in the moment but not deep enough to revisit. The fix is to go deeper. Teach something. Give a checklist. Give a step-by-step breakdown. Something the viewer will want to return to, not just watch once.
Carousels tend to get more saves than Reels because they function more like a reference document. If saves are your goal, carousels are often the better format. For save-friendly formats that actually work, see Instagram carousel ideas that get saves and shares.
What Do Likes Tell You?
Likes are about emotional resonance. When someone double-taps, they are usually saying: I feel this. Or: I agree. Or: This is exactly what I needed to hear. Mosseri placed likes in the top three signals in January 2025, but specifically for connected content: content moving to people who already follow you.
Low likes often means the content is too informational and not relatable enough. The fix is to make the content feel more human. Share a real situation. Voice the objection your audience is already thinking. Make them feel seen.
That is different from chasing sentiment. It is about being specific enough that the right people see themselves in your content.
What Does Accounts Reached Tell You?
Accounts reached is the total number of unique accounts that saw your content. The breakdown inside it matters more than the headline number. Connected reach is followers. Unconnected reach is everyone else. That split tells you whether you are growing or just talking to people who already know you.
If your unconnected reach is low, your content is staying inside your existing audience. The lever to pull here is sends per reach. Get people DM-ing your content to friends who do not follow you. That is what pushes it into cold audiences.
If your unconnected reach is high but follower growth is not keeping up, you are reaching new people who are not converting. That usually means your hook is broad enough to attract people but your overall content is not specific enough to make them stay.
What Does Accounts Engaged Tell You?
Accounts engaged is the number of unique accounts that interacted with your content in any way: likes, comments, saves, shares, replies, you get the gist. It is a useful gut-check number but too broad to act on by itself.
High accounts engaged but low saves and sends? People liked the content but didn’t find it deeply valuable or shareable. Common with entertaining content that lacks a specific takeaway.
High saves and sends but low likes? Your content is genuinely useful but may not be landing emotionally. Add a moment of real relatability to the next version and see if likes pick up.
What Do Stories Metrics Tell You?
Stories have their own metrics and they work differently from feed posts. The main ones are views, accounts reached, and interactions.
Instagram’s Help Center for Stories defines interactions as replies, shares, sticker taps, and link taps. One thing almost nobody knows: Stories Insights remain accessible for up to two years after you post, even after the Story itself has expired. You can still pull that data long after it is gone from the feed.
For Stories, the metric to watch is replies and DMs. If a Story gets a lot of replies, your audience is not just watching passively. They are responding. For creators who want to turn attention into real conversations with potential clients, Stories replies are often where that starts.
Hook Problem or Content Problem? Here Is a Three-Question Check
You can diagnose most underperforming Reels with three questions. Each answer points to a different fix, so you know exactly what to change in the next video.
Is skip rate high? Hook problem. The content may be solid, but the first three seconds are not stopping the scroll. Fix the opening.
Is skip rate low but watch time also low? Content problem. People stopped to watch, but something in the video lost them. Check the retention chart for the specific timestamp where viewers drop off.
Are skip rate and watch time both solid, but shares and sends are low? The content held attention but was not valuable or compelling enough to pass along. Ask honestly: would you forward this video to a specific person you know? If not, why not?
Three questions. Five minutes. One fix at a time.
The Small Test Audience That Decides Your Video’s Fate
Here is something most platform guides skip over.
Every time you post, Instagram does not push your video to your full audience at once. It shows it to a small test group first. Sometimes as few as five people. Sometimes up to 50.
That first group decides what happens next. If they scroll past quickly, the algorithm assumes the video is not interesting and slows distribution. If they watch, save, and send it, the algorithm expands reach. The cycle scales from there.
This is very very key. A great video can stall at a handful of views. It may have hit the wrong first five people on the wrong day. Their behavior had nothing to do with your content quality.
Hold on, Max. Does that mean I should repost videos that did not get traction?
Yes. Repost them. Not a copy. The original. Give it a different test group and a second chance.
Before you do, check your skip rate and watch time on the small audience that did see it. If those numbers look solid, the content is probably fine. The distribution just didn’t break through on the first attempt. Reposting is a valid and underused response to that situation.
What you should not do is panic and assume the video was bad because the view count is low. Let the data tell you.
Instagram Insights Explained: The Full Metric-to-Action Map
Every metric points to one specific behavior change. Here is the complete map.
Watch time is low: Make the video more interesting from beginning to end. Tighten the hook. Cut slow sections. Use the retention chart to find the exact timestamp where viewers leave.
Skip rate is high: Fix the first three seconds. The hook is not stopping the scroll.
Shares are low: Make the content more valuable or more entertaining. Ask if a viewer would send this to a specific person they know. If not, find out why.
Sends per reach is low: Make the content more targeted and more useful. Sends happen when the video is so on-target that a viewer immediately thinks of someone who needs to see it.
Saves are low: Go deeper. Give something worth bookmarking. A checklist, a framework, a step-by-step breakdown the viewer will want to return to.
Likes are low: Make the content more relatable. Share a real situation. Voice the objection your audience is already thinking. Make them feel seen.
Comments are low: Make the content more provocative. Ask a direct question. Add a CTA that invites a specific reply.
Unconnected reach is low: Focus on sends. Get people DM-ing your content to friends who don’t follow you.
That map is the whole weekly review. Thirty minutes. The Edits app. One metric furthest off. One fix. Repeat.
How the Building Blocks Principle Connects to Your Weekly Review
Boom. Now you have the map. Here is how to use it without burning out.
Max teaches something called the building blocks principle. Any video can be broken down into its parts: the hook, the format, the topic, the call to action. Each one is a building block.
When a video underperforms, you do not scrap the whole thing. You swap one building block and test the result. That is the whole game.
It connects directly to reading Insights. Skip rate high? Test a hook swap. Saves low? Test a deeper content swap. One variable per iteration. One metric at a time. That is how you get better without guessing and without burning yourself out trying to change everything at once.
For a full breakdown of how the building blocks work together, see the 7 building blocks of a viral video.
You can also see how other creators have put this system into practice at Max’s creator reviews page.
Want the templates behind content that actually gets watched? Grab the templates here and start with the first one today.