You have a Reel you’re not sure about. The topic is a step away from your usual, the hook is sharper, the format is new. Post it and you risk followers who came for something else. Don’t post it and you never find out whether it would have worked. The old workarounds were a second account or “testing in Stories,” which told you nothing.
Trial Reels close that gap: the video goes only to people who don’t follow you. Followers don’t see it in their feed or on your profile. After 24 hours Instagram shows you the numbers, and you decide whether to send it to followers.
Below: who gets the feature and how to turn it on, what the stats show and how to read them honestly, which hypotheses are worth testing, a one-week testing routine, and the cases where Trial Reels won’t help you.
How Trial Reels work
- Available to professional accounts (business or creator). A personal account has no toggle; switching the account type in settings is free and reversible.
- Shown to non-followers only, in the Reels tab and Explore. On your profile the video is visible only to you, with a “Trial” label.
- After 24 hours you get stats: views, likes, comments, shares. Instagram also tells you whether the trial is doing better or worse than your recent Reels.
- Two ways forward. “Share to followers” manually at any time, after which it becomes a regular Reel. Or automatic sharing: if you enable it in advance, Instagram sends the Reel to followers after roughly 72 hours when the trial is “performing well” by its own criteria. The criteria aren’t published.
- You can also never share it. A failed test stays a trial or gets deleted. Followers never know it existed.
One important limit: a Trial Reel can only be published from the Instagram app. The Content Publishing API, which every scheduler uses, including Repost21, has no trial option at the time of writing. The test is manual; the calendar is for everything else.
How to publish a Trial Reel
- Build the Reel as usual: video, cover, caption, audio.
- On the final screen before posting, find the Trial toggle and switch it on.
- On the same screen, decide what happens after the test: keep the decision for yourself or turn on automatic sharing to followers.
- Tap Share. The Reel appears in your Reels tab with a label only you can see.
- After 24 hours, open the Reel and check the stats. If you decide to publish, the “Share to followers” button is right on the Reel.
If there’s no toggle, check the account type and update the app. The feature rolled out in waves, and some accounts got it later than others.
How to read the stats
The biggest mistake is comparing a Trial Reel with regular ones. A regular Reel gets its first views from followers, which inflates the start. A trial goes to a cold audience from second one, so it can only be compared with other trials.
What to look at:
| Metric | What it tells you | Rule of thumb |
|---|---|---|
| Views in 24 h | Whether the algorithm liked the video on a cold audience | Compare trial to trial only |
| Shares / views | Whether strangers will send it to a friend | Above 1 per 100 means the hook and topic work |
| Saves / views | Whether there’s something useful to come back to | Higher for how-tos and lists |
| Follows from the Reel | Whether it brought your audience, not random viewers | Check the full Reels insights |
| Retention | Where people drop off | A cliff at 2–3 seconds is a hook problem, not a topic problem |
The thresholds in the table are working numbers from practice, not platform norms. Your own norm appears after 5–6 trials: log the numbers in a spreadsheet, and in a month you’ll know what “good” means for your account.
Sample size. At 300 views every metric is noise. Draw conclusions from 1,000 views and up; if a trial didn’t reach that in a day, that is itself an answer. The rest of the metrics are decoded in our guide to Instagram Insights.
What to test
A Trial Reel answers one question: will this hook a stranger? So test the things that affect the first seconds and the topic.
- The hook. The same video with three different opening lines: “You’re overpaying for hair color,” “Three mistakes that wash your color out in a week,” “Stylists, never say this to a client.” Three trials, one topic, three entrances.
- A topic outside your niche. A pastry chef wants to film a video about pricing. Followers may not care; a cold audience might.
- Format. Talking head versus voiceover on process footage. One script, two shoots.
- Length. 15 seconds versus 45 on the same material.
- Language. For bilingual markets, one version in English and one in Spanish. The trial shows which cold audience responds without annoying followers with duplicates.
- On-screen text and cover. No headline, a question headline, a number headline.
What’s pointless to test: posting time (followers aren’t in the test and recommendations run for days), music without changing the content, minor edits. Video ideas and hook scripts are in our Reels ideas by niche and the article on AI video scripts.
A one-week testing routine
One test a day, so the videos don’t compete for the same audience.
- Monday–Wednesday: three trials of one video with different hooks, one per day, same time.
- Thursday: read the 24-hour numbers for all three. The winner is picked by shares and retention, not likes.
- Friday: share the winner with followers. Delete the two losers or leave them as trials.
- Weekend: log the result in a spreadsheet: hook, topic, format, views, shares, decision.
After a month you have eight rows and the first patterns: number hooks consistently beat question hooks, say, or videos over 30 seconds lose half the audience by second five. That’s the answer you were after.
When you don’t need Trial Reels
- Under a thousand followers. Your regular Reels are already watched 80–90% by non-followers. A trial adds almost nothing and costs time.
- Local business. A salon in Brooklyn gets test viewers from Manila and Madrid. They won’t come in, and their reaction says nothing about your clients. For a local business the profile and booking flow matter more; see our guide to social media for salons.
- You want to know how followers will react. A trial checks a cold audience. It won’t tell you how your own people take it.
- Tiny samples. If trials get 200 views each, the differences between them are chance.
- You need to publish on a schedule. A Trial Reel can’t be scheduled in the app or via the API. Manual only, at the moment of the test.
- A reach drop. A trial doesn’t fix a slump; at best it diagnoses hooks. The causes of falling reach are covered separately.
Trials are for testing. After that the regular work begins: a calendar, consistent publishing, weekly analytics. In Repost21, Instagram posts and Reels go out on schedule in the best slots, and account analytics shows what your test winners do to reach and follows over time. Try it free.