Reels Is Not TikTok With a Different Logo
Brands that copy TikTok creative straight into Reels usually see worse numbers, then conclude Reels doesn't work.
The placement does work. It just rewards slightly different things, and the audience arrives in a different frame of mind.
Three differences drive most of the gap.
Audience intent. Instagram users are more likely to already follow brands and more comfortable with polished content. Rougher-than-rough footage that thrives on TikTok can read as low effort here.
Sound behaviour. A larger share of Instagram viewing happens muted. Captions aren't optional.
Profile context. Someone who likes your ad will often tap through to your profile before buying. A neglected grid kills conversions that the ad already won.
The Setup Mistakes That Waste Budget
Running Reels inside a broad Advantage+ placement. Automatic placements will serve your vertical video into feed and Stories as well. The creative that wins in Reels is rarely the one that wins in feed. Isolate Reels when you're testing creative so you can read the results.
Uploading 9:16 with burned-in UI. If your video has a fake TikTok interface baked in, Instagram's algorithm and its audience both notice. Meta has historically down-weighted visibly recycled competitor content.
Ignoring the safe zone. Instagram overlays the caption, profile name, and CTA across the bottom of the frame. Anything important in the lower 20% gets covered. Keep text and product placement above it.
Creative That Works on Reels
Open on the product in motion, not on a talking head. Instagram's scroll is faster than it looks. Showing the thing being sold inside the first second outperforms introductions.
Caption everything. Not auto-captions pasted on top — readable, well-positioned text that carries the message with the sound off.
Use a clear visual rhythm. Cuts every 1.5 to 3 seconds hold attention. Static shots lose it.
Close with a reason, not a command. "Shop now" is weak. "Free returns for 60 days" gives someone a reason to act.
The Format Split Worth Testing
Reels rewards two very different creative styles, and most brands only try one.
Style A: the demonstration. Product, in use, in context, cut tight. Works for anything with a visible before and after — cleaning products, skincare, tools, apparel fit.
Style B: the talking recommendation. A person explaining why they like the product. Works for anything where the benefit is felt rather than seen — supplements, software, comfort-led products.
Run both. The winner is rarely the one you'd predict, and it often differs by product rather than by brand.
Measurement: Read Hold Rate, Not Just CTR
Click-through rate tells you the ending worked. It doesn't tell you where the video lost people.
Pull the retention graph and look at two points.
The 3-second hold. If fewer than half the impressions make it past three seconds, the problem is the opening frame. Nothing after it matters yet.
The midpoint. A steep drop around the middle usually means the video explained instead of showed. Cut the explanation and demonstrate earlier.
Fixing a hook is cheap. Fixing a whole concept is not, and the retention graph tells you which one you're dealing with.
Volume Still Decides the Outcome
The same rule that governs TikTok applies here: the brands that win are testing more creatives, not better-guessing which single creative to make.
Meta's delivery system needs variation to optimise against. Three creatives give it almost nothing to work with. Twenty give it a real signal.
This is the practical argument for generating ad variations rather than filming each one. When a new angle costs a few dollars instead of a few hundred, running fifteen openings against the same offer becomes a normal Tuesday rather than a quarterly project.
A Four-Week Starting Plan
Week 1. Produce 10 creatives: 5 demonstration, 5 talking recommendation. Same offer across all of them.
Week 2. Kill anything below a 25% three-second hold. Build 5 new variations of the top two openings.
Week 3. Take the winning opening and test it against three different offers.
Week 4. Scale spend behind the winning opening plus offer combination. Start the next batch.
The brands that compound on Reels are the ones running this loop continuously, not the ones producing one brilliant video per quarter.