Pick episode one of your Instagram Series from Edits comparison
Compare up to three Reels in Edits, pick the one with the best watch time, then re-cut it as episode one with a clean opener, captions and a next-episode cue.

Compare your three best candidate Reels in Edits and pick the one with the strongest watch time as the pilot of your Series, then re-cut it so it works as episode one. Search coverage of the August Edits update says Insights can now compare up to three Reels side by side on views, watch time, follows and interactions.
That coverage is summarised by NapoleonCat (NapoleonCat, read 2026-10-07). Meta's Series test lets select creators group Reels, new and old, into a run with its own profile page (TechCrunch, read 2026-10-07), so an old winner can become episode one without a re-upload of the others.
Choosing the pilot
Do not pick by views alone. A pilot has to make a stranger want episode two.
| Check | Source | Pass if |
|---|---|---|
| Watch time | Edits comparison | Highest of the three |
| Follows | Edits comparison | Not lowest |
| Stands alone | Your own viewing | A new viewer understands it |
| Leaves a question open | Your own viewing | There is a reason to watch next |
Re-cutting it as episode one
Import the file to your workspace and look at the first frame with video frames at at: [0]. If the opener is slow, cut it with video trim ($0.02 per job). The default exact precision is frame-accurate.
Add a short end card or a spoken line pointing to the next episode. You can burn a line with cues in video captions ($0.20 per job up to 60 seconds): send text, start and end in seconds for the last two seconds. Keep it a cue, not a link, because a link inside the video is not clickable.
What to prepare for episode two
- Plan episode two before you post episode one, so the question you open has an answer.
- Keep one caption style for the whole Series by reusing the same
styleanddesignon every caption job. - Number your files. Whatever order Instagram shows, you should know yours.
What if the best Reel is not the best pilot
Sometimes the highest watch time belongs to a Reel that is a one-off joke, not a doorway to a series. If so, take the second-best and make it the pilot, then use the winner as a standalone. The numbers tell you what people watched, not what they would watch next.
A small test helps: describe the Series in one sentence and check that the pilot fits it. If it does not, either change the sentence or change the pilot.
A note on sample size
Two Reels posted a week apart reach different audiences, so a comparison is a hint, not a verdict. If the gap is large and the Reels are similar in length, trust it. If the gap is small, choose by fit with the Series and move on. The re-cut is cheap either way, and episode two will tell you more than any chart.
Sources
Related posts
More in Use cases
- Pinterest Collections video: standard 4 s to 15 min spec, 2:3 hero
Pinterest says Collections follow its standard video spec: 4 seconds to 15 minutes, 2 GB, with a 2:3 ratio. Here is what a Sume render has to match.
- Plant shop care-tip loop: Seedance 2.5 with one image as both frames
Send one photo as first_frame and last_frame to Seedance 2.5 on Sume for an 8-second 1:1 loop-style clip: the request body, $4.63 at 720p, and the limits.
- Podcast episode art for 50 episodes on GPT Image 2.5: medium vs high
Fifty square episode covers on ChatGPT Image 2.5 cost $0.83 at medium or $3.29 at high from text, as read from Sume's estimates.
- How do I turn one podcast episode into five quote clips for social?
Transcribe the episode in 10-minute chunks, split five quotes out, put the cover still under each and burn captions: $1.80 for a 28-minute episode on Sume.
Written by Sume