Which hook and opening styles hold attention now in Reels versus Feed?
Our answer
Build hooks per concept. Design Reels openings for sound and make every opening read without it, name the buyer's situation and problem, and judge hook rate against each placement's norm: about 30% in Reels, 23% in Feed.
- Confidence
- Low
- Strongest proof
- 3 Meta rules
- 1 dataset, 9 practitioners, 10 opinions or claims
- Sources
- 15 sources
- 23 statements
- MLP
EZ
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentSound matters in Reels but not everywhere. Meta's Ads Guide strongly recommends sound for Instagram Reels ads. Its Instagram video ads page says most users watch with sound off and advises subtitles and on-screen text.
Each placement has its own normal hook rate. Median 3-second hook rate on cold traffic sits at about 23% in Feed and about 30% in Reels (AdSights, reported by Jetfuel Agency, a consumer marketing agency, in its compiled 2026 benchmarks).
Name the buyer's situation, not a personal trait. Common Thread Collective, an ecommerce ad agency, advises call-out hooks that name the person as specifically as possible. Meta's personal attributes rule bars implying traits such as a health condition or age.
Our judgment
- Design Reels openings for sound and make every opening clear without it. Meta's Ads Guide strongly recommends sound for Reels ads, the narrower statement, so it sets the Reels opening. Its sound-off advice, and its note that music or voiceover helps those who listen, set a floor for every placement.
- Keep one hook set per concept. Split hooks by placement only if your own breakdown shows a style winning in one, extending Taylor Sicard's advice to benchmark within placement.
- Put the offer in the opening during sales. This follows Motion's hit rates for promotional hooks.
- Judge hook rate against AdSights' norm for each placement, not Taylor Sicard's reading that Reels runs lower. AdSights gives a range for each placement and Sicard gives no placement figures. Neither discloses a sample.
Our judgment, built from Meta, Taylor Sicard, Motion and Jetfuel Agency: our reading of these sources, not a finding from any one of them.
What the sources say
23 statements from 15 sourcesSavannah Sanchez, who runs a UGC agency, says not to show the product until at least 3 seconds in. Open with a creator talking to camera so viewers don't swipe away. No data.
“Don't show your product until at least 3 seconds into your ad!”
Phil Kiel briefs creators monthly with a 20-odd page brief per brand built from account data, and asks for separate 3 to 5 second visual shots to pair with spoken hooks in the edit.
“The first 2 seconds of an ad are a picture, not a sentence, so we ask for the pictures separately and pair them with the spoken hooks in the edit.”
An operator had Claude review 100 top ads from 10 DTC brands: half were statics, 83 opened with words on the first frame, and only 3 spoke to the unaware. How top 1% was judged is not stated.
“Only 3 of 100 talk to the unaware.”
Nate Lagos says to write ad angles around acute pain, not chronic pain, and test them against existing copy while keeping the rest of the ad consistent. He says some DTC brands are crushing this but gives no figures.
“Humans are not motivated by chronic pain. They are motivated by acute pain.”
Meta's Instagram video advice: videos under 15 seconds tend to do best in feed and under 10 seconds in Stories, with the brand shown in the first 3 seconds. It cites no study.
“Feed video ads: Can be up to 60 minutes, but shorter videos (under 15 seconds) tend to perform best in feed. Stories ads perform best under 10 seconds.”
Reading AG1's Ad Library, Lorenzo Pravata notes its most-shown ad, a plain free Welcome Kit static, has run seven months; he advises one evergreen offer card and problem-first hooks. Not results.
“tells you nothing about spend or ROAS”
Meta's specs for Instagram Reels video ads: 9:16 ratio, sound optional but strongly recommended, and keep 14% of the top, 35% of the bottom and 6% of each side free of text and logos.
“Consider leaving at least 14% of the top, 35% of the bottom, and 6% on each side of your asset free from text, logos, or other important creative elements.”
An operator says usually 1 of 3 hooks wins. Putting a hook from a 58% hook-rate ad in front of a 25% ad helped an app account go from $6K ad spend in April to $392K in July.
“Hook rate is the leading indicator”
Ridge CEO Sean Frank argues that instead of obsessing over ad hooks and funnels, advertisers should make the product 10% better at the same price, or 10% cheaper at the same quality.
“Stop obsessing over ad hooks and funnels and start obsessing over how to make your product 10% better at the same price.”
Jetfuel relays benchmarks with no sample: AdSights puts the median 3-second hook rate near 23% for Meta feed cold traffic and 30% for Reels, while Taylor Sicard calls under 25% the kill zone.
“Notice the two sources disagree on where good starts, which is exactly why a single benchmark is dangerous.”
Meta's Health and Wellness policy (rewritten July 23, 2026) requires 18+ targeting for diet, weight and cosmetic ads and bans cure claims, but drops its 2024 ban on side-by-side weight loss comparisons.
“promises of specific outcomes within a set timeframe without disclaimers or qualifiers”
Opascope, an agency, says across 200+ audited accounts story-led visual hooks tended to beat talking-head openers, and hook-only tests no longer give clear signal on Meta. It gives no method or numbers.
“Hook-only tests no longer produce differentiated signal on Meta”
At Motion's July 2026 summit, Harry's said UGC cuts are not what they were and polished statics with B-roll are winning. Arrae said small creators in a yapper style work through partnership ads. No figures.
“putting reviews on things is suddenly working for us again”
Emily Huo says social ads should put the most relevant message within the first three seconds.
“make sure you have your message, most relevant message within the first three seconds of your ad.”
A benchmark guide says a good Meta hook rate is about 30% or higher and a good hold rate about 25%. It says to compare placements separately, since Reels and Stories run lower hook rates than in-feed video.
“Benchmark within placement, not across it.”
Savannah Sanchez says brands spending under $100K a month should test two written and two visual hook variations per concept, plus the same concept with different creators, and avoid repetitive AI scripts.
In one week of Wispr's video ads, the highest hook rate (39.5% watched 3 seconds) kept only 14.3% to the 25% mark and 3.6% to the end, and cost over $30 per download, against a winner about 40% cheaper.
“Wispr's highest hook rate of one week (39.5% watched 3 seconds) kept only 14.3% to the 25% mark and 3.6% to the end”
Josh Levine says Manychat builds each ad to hold viewers from second 3 to second 15 before working on the hook, since a weak hook is easier to fix once the body works.
“To Josh's point, hold beats hook with receipts here.”
Wispr scores video ads with a Banger Score: 30% hook, 30% hold, 25% CTR and 15% CPM efficiency, each against the median ad in the concept. A score of 50 is the median and 75 or more is a winner.
“We think 30% uh, towards hook, 30% towards hold, 25% click-through rate, and 15% CPM efficiency times by 50.”
CTC's podcast host, relaying a colleague, says each hook should call out the viewer as specifically as possible, like a mom of young children who jogs in the morning, then add the problem.
“Are you a mom of young children who jogs in the morning?”
A CTC podcast speaker says you can reach different audiences without targeting by changing the product name, landing page and hooks, such as athletic versus arthritis sleeve. Meta's category limits go unaddressed.
Motion's benchmark of 578,750 ads found promotional hooks (newness, sales, price anchors, urgency) had the highest hit rates, 6-11%, though Motion says its holiday data window inflates them.
“Promotional hooks are over-represented.”
Meta's ad rules ban copy that asserts or implies personal attributes like health, age or finances, such as "Do you have diabetes?", but allow passing references and "you" language without an attribute.
“ads must not contain content that asserts or implies personal attributes”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it2 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer