The number labelled forward in story insights counts people who tapped to jump to your next frame. They stayed with you. Nobody left. A forward tap is a speed setting, and a very large share of people watch stories exactly that way, thumb hammering the right hand side of the screen from your first frame through to your last. The two figures that record a real loss are next story, meaning somebody swiped sideways into a different account, and exited, meaning they closed the story screen. If a high forward count has had you deleting frames and rewriting your plan, you have been reading the wrong column of the right table.
Table of Contents
Each of the four figures records a different finger movement
Instagram groups four of these together, and the only reliable way to tell them apart is to picture the hand holding the phone.
| Figure | What the viewer physically did | What it tells you |
|---|---|---|
| Forward | Tapped the right hand side of the screen to reach your next frame | They are moving through your story at speed and are still in it |
| Back | Tapped the left hand side to return to the frame before | Something on the previous frame was worth a second look, or went past too quickly |
| Next story | Swiped sideways, which skips whatever you had left and opens another account | They chose somebody else over the rest of your frames |
| Exited | Closed the story screen, by swiping down, tapping the cross or leaving the app | They stopped watching stories altogether at that point |
Now the part almost nobody gets told. Every one of those four is credited to the frame that was on screen at the moment it happened. An exit logged on frame three means frame three is where the person gave up. A back tap logged on frame four is really a comment about frame three, because frame three is where that tap sent them. Read the panel with that offset in mind and it starts making sense.
One caution about your final frame. A forward tap there has nowhere left to go inside your story, so the last frame behaves differently from every frame before it. Compare frames one through five against each other and treat the closing frame as its own thing.
The forward count is the one almost everybody misreads
Here is the reasoning I hear constantly. Forward means skip. Skip means bored. Bored means your stories are too long, so cut them down to a single frame and stop testing everyone’s patience. You were expecting me to nod along to all of that, and it sounds sensible right up until you look at what the number is counting.
Somebody who taps forward through nine frames in eleven seconds saw nine frames. They read the headline on each one, they got the point, and they will open your story again tomorrow. That is a viewer. Somebody who exits on frame two saw two frames and went back to their feed. Those two behaviours produce completely different outcomes for you, and only one of them is a problem you need to fix.
Back taps deserve more attention than they get. A back tap usually means a frame moved on before somebody had finished with it. Text sized too small, a price they wanted to check, a name they meant to catch, a link sticker missed by half a second. If one frame keeps generating back taps from the frame after it, that frame is carrying too much and wants splitting in two.
Lay the frames out in order and the losing frame identifies itself
Single figures tell you very little. The shape across the whole story tells you almost everything. Write down the views for each frame in order, then work out how much each frame lost compared with the one before it.
Here is a made up example so the arithmetic stays easy.
| Frame | Views | Lost against previous frame | Exits |
|---|---|---|---|
| 1 | 900 | Low | |
| 2 | 840 | 7% | Low |
| 3 | 800 | 5% | Low |
| 4 | 560 | 30% | High |
| 5 | 540 | 4% | Low |
| 6 | 520 | 4% | Low |
Frame four is the answer. Everything before it holds and everything after it holds, so nothing is wrong with the length of the story or with the subject. Something specific on frame four sent people away. Open it and look at it with fresh eyes. A wall of small text, a video that opened at full volume, a sales ask that arrived earlier than people were ready for, a frame that simply sat there with nothing happening on it.
The way people left tells you which repair to try.
- Exits climb on one frame: that frame gave somebody a reason to put the phone down. Usually effort, or an ask.
- Next story climbs on one frame: that frame felt like an ending, so somebody else’s story looked like the natural next thing.
- Forward climbs while views hold steady: there is nothing to repair. People are reading you quickly.
- Back taps cluster after one frame: the frame before it was too dense or moved too fast.
Judge the story against your own account, because published benchmarks belong to somebody else
Plenty of articles quote an ideal completion percentage. Ignore them. That figure came from accounts with different follower counts, different posting frequency, different countries and a different relationship with their audience, and it will tell you nothing at all about yours.
Two comparisons are worth making instead. The first is frame against frame inside one story, which is the curve above. The second is story against story across your own history, similar day of the week, similar length, over several weeks. A healthy story loses people gently and evenly. An unhealthy one has a cliff in it.
Length changes the reading too. A fifteen frame story that has lost half its audience by frame ten is behaving normally. A four frame story that loses half by frame two is not behaving normally at all. Compare like with like or you will draw the wrong conclusion twice.
Check the first frame against your follower count before you blame anything inside the story. If frame one is reaching only a small slice of your followers, the frames are not the issue. People are not opening your stories in the first place, which is a question about how often you post and whether your ring looks worth tapping, and no amount of frame surgery will fix it.
This is where I stop joking
Small numbers lie. If a story reached sixty people, one person leaving moves your percentage by nearly two points, and you can burn a month chasing what was really one person answering the door. Wait until you have several stories of the same shape in front of you before you conclude anything.
This matters because of what people do next. They kill a series their audience actually liked. They stop posting the format that was quietly bringing in enquiries. They spend money trying to repair something the figures never showed was broken. Anything that costs money or cannot be undone deserves more evidence than one story’s panel. If you find yourself making a permanent decision on a Tuesday because of a Monday story, sleep on it.
Where the numbers live now, and how long you get to keep them
While your story is still live, open it yourself and pull up the viewer panel from the bottom of the screen. Viewers sit at the top and the figures sit underneath them.
After twenty four hours the story leaves the ring and the data goes with it, unless you have Save to archive switched on. With archiving on, the story sits in your archive and the same panel opens from there. Switch it on now if it is off, because a story you never saved cannot be brought back later.
Your professional dashboard also lists stories as content with a date range on it, which is a far easier place to compare many stories at once rather than opening them one at a time.
Two practical warnings. Instagram has moved these buttons more than once and renamed things along the way, so look for the small chart icon rather than memorising a menu path from a tutorial that may already be out of date. Views also replaced impressions as the headline figure, so any guide still telling you to find your impressions is describing a screen that no longer exists. And the app keeps this history for a limited window, so if you want to compare today against three months from now, screenshot the panel or type the frame by frame views into a note. Thirty seconds of typing, and you own the record instead of renting it.
Run this on tonight’s story
- Write down the views for every frame, in order.
- Work out the drop from each frame to the next as a percentage.
- Find the biggest single drop. That frame is your suspect.
- Check exits and next story on that frame to see how people left it.
- Change one thing about that frame, post the same kind of story again, and compare.
- Leave the forward count alone.
Do that four or five times and you will know more about your own audience than any benchmark article can tell you. The forward count was never the villain here. It was just standing closest to the door when the audience walked out.