Question hooks — open the loop, earn the next 3 seconds

A question hook doesn't work because it ends in a question mark. It works because the human brain auto-completes questions before the conscious mind makes a decision. By the time your viewer's thumb registers a swipe, they've already answered the question internally — and that tiny moment of participation is what earns the next three seconds.

Most operators think they understand this archetype. Most of them are writing the wrong questions.

"Want to lose weight?" is a question. It's also the fastest way to get skipped by people who do and don't alike. The question hook archetype isn't about asking something — it's about opening a curiosity gap the viewer can't close without your ad. That's a much narrower brief.

Question hook is one of the 5 hook archetypes that govern every winning DTC ad. This is the full-depth treatment: sub-types, failure modes, the platform where questions die, and how AdRevila's analyzer grades the archetype in your reports.

TL;DR

• Question hooks work by opening a loop the viewer must close — curiosity gap, not grammar.
• Four sub-types: Diagnostic, Rhetorical, Provocative ("why does nobody tell you…"), and Yes-Ladder. Each fails differently.
• The most common failure modes: questions the viewer can answer "no" to and scroll; questions with obvious answers; low-stakes questions; questions that don't match the offer.
• Questions often die on TikTok — the platform's contract is "tell me something," not "ask me something." Convert to an implied-question statement instead.
• Strongest Schwartz fit is Problem-Aware and Solution-Aware cold traffic. Wrong archetype for Most-Aware retargeting.

Two questions — one opens a loop, one closes it instantly

AG1, Meta, 2025. Clean white background, no music, creator looking straight at camera: "Why do you feel foggy every morning even after 8 hours of sleep?" Pause. The viewer's internal answer is: I don't know. I've wondered about that. The loop is open.

Generic supplement brand, Meta, 2025. Creator in a kitchen: "Want more energy?" The viewer's internal answer is: sure, or whatever. Loop closed before the body even starts.

The first question names a specific, falsifiable state — the viewer either experiences it or doesn't. If they experience it, they need the answer and they wait for it. If they don't, they self-select out immediately. Both responses are right.

The second question is technically a question, but it's stakes-free. Nobody feels the cognitive pull to answer "Want more energy?" — it's been on every supplement ad for 20 years. The viewer's pattern-recognition marks it as spam and the thumb moves.

The question hook archetype lives entirely in the specificity of the problem named and the urgency of the gap it opens. Write questions the viewer already has but hasn't said out loud. That's the job.

What the curiosity gap actually is

The term comes from behavioral economics — the "information gap" model from George Loewenstein. The idea: curiosity isn't triggered by what you don't know in general. It's triggered by knowing there's something specific you don't know, and sensing you're close to knowing it.

"Why does your stomach still bloat after lunch?" works not because bloating is novel, but because it frames the problem as having a specific, nearby answer the viewer is missing. The "still" is load-bearing — it implies they've tried things and failed, and there's a reason why. That's the gap. The ad promises to close it.

"Why does soda cause inflammation?" is a worse question hook for the same product. The viewer doesn't know they have that problem. No gap, no pull.

Senior strategists write question hooks backward: find the gap the Problem-Aware buyer has been living with, then write the question that names it exactly. The question is the symptom; the offer is the diagnosis.

The 4 sub-types of question hooks

The archetype splits into four types that operators often conflate. They have different mechanisms, different failure modes, and different stage fits.

Diagnostic — "Are you still…"

The diagnostic question implies the viewer has an ongoing, unresolved problem. The "are you still" or "do you still" framing is the tell.

Examples. "Are you still washing your face with a cleanser that strips your barrier?" (skincare). "Do you still feel tired three hours after waking up?" (AG1-adjacent). "Are you still buying supplements without checking the third-party cert?" (supplement category).

Why it works. The word "still" implies the viewer should have solved this by now — it carries a soft accusation that creates mild urgency. The viewer is either nodding (in) or scrolling (correctly out). For Problem-Aware cold traffic, this is the highest-hit-rate sub-type.

Fails when the "still" is obviously untrue for most viewers. "Are you still using a bar of soap from 2004?" — the viewer isn't and knows it. The false premise kills the diagnostic before it opens.

Rhetorical — "Have you ever noticed…"

A softer entry. The question doesn't name a problem directly — it asks the viewer to observe something, implying the observation is the path to the answer.

Examples. "Have you ever noticed your skin looks worse when you're stressed?" (skincare — Krave Beauty's register). "Have you ever tracked how your energy changes hour by hour?" (wearables/health trackers). "Have you ever read the ingredients on your protein powder?" (Ridge/Magic Spoon adjacent).

Why it works. Lower accusation, higher curiosity. The viewer can engage without admitting a problem — they're just observing. Better for audiences who aren't sure they have the problem yet; good bridge from Unaware to Problem-Aware.

Fails when the observation is so obvious the question reads as patronizing. "Have you ever noticed you get tired when you don't sleep?" — yes. Thanks. Sub-type works on novel observations only.

Provocative — "Why does nobody tell you…"

The question frames the answer as suppressed, hidden, or counterintuitive. The viewer isn't expected to have the answer; they're being invited into a reveal.

Examples. "Why does nobody tell you that zinc depletes before magnesium?" (supplement). "Why do every 'clean' skincare brand still use these three ingredients?" (clean beauty). "Why did I get worse results at the gym after hiring a coach?" (fitness info-product framing).

Why it works. The "nobody tells you" framing creates an in-group/conspiracy lean — the brand is about to give the viewer information that others are withholding. Strong on Hims, Ritual, Seed, and any brand playing the "most brands lie about X" angle. Curiosity gap is wide because the viewer genuinely doesn't know the answer.

Fails when the answer is, in fact, widely known. If the viewer's answer is "everyone knows that," the provocative frame reads as dishonest. Test: would a normally-informed person in your category find the answer surprising? If no, don't use this sub-type.

This sub-type travels well to Meta and stays in its lane. It's the one that gets misused on TikTok most often — see the TikTok section below.

Yes-Ladder — sequential micro-agreements

The yes-ladder isn't usually framed as a single question. It's a rapid sequence of questions (or implied questions) that the viewer answers "yes" to — each one builds commitment before the offer frame.

Examples. "Do you exercise at least four days a week? Do you watch what you eat? Do you still feel like you're behind on your goals?" Sequence of three; the viewer answers yes, yes, yes. By question three, the implicit contract is: you're doing the work, the problem isn't you, there's a missing piece. Now the product is the missing piece.

Why it works. Cialdini's commitment-consistency principle, applied to hooks. Once someone has said yes twice, the third yes — to the offer frame — feels consistent. Ridge, Liquid IV, and Hims have all run this format.

Fails when the ladder runs too long. Three questions is the ceiling for a cold-traffic hook. Four questions reads as a quiz, not a hook, and the viewer opts out of the work. Also fails when any question in the sequence is one they'd answer "no" to — the whole ladder collapses.

The failure modes (4 ways question hooks die)

Question hooks have a specific set of failure modes that don't apply to other archetypes. These are the ones operators miss most.

1. The answerable-no question.

The easiest scroll escape. "Are you tired of feeling bloated?" — the viewer answers "I don't have that problem" and swipes. "Do you want to feel your best every day?" — the viewer answers "sure, doesn't everyone" and swipes. Any question the viewer can opt out of with a casual "no" or "not really" loses them instantly.

Diagnostic: if your question has a population-level affirmative rate below ~60–70% for your actual targeting, it's answerable-no. The question needs to describe something specific enough that the people who have it are instantly, viscerally engaged.

2. The obvious-answer question.

"Want to lose weight?" "Tired of bad skin?" "Want to save money?" These register as spam before the viewer finishes reading them. The brain pattern-matches to 15 years of internet ads and marks the whole ad as low-trust within a second.

The fix: add specificity until the question doesn't sound like every other ad in the category. "Want to lose weight?" is obvious. "Why do you lose three pounds in week one and nothing in weeks two through four?" is specific. Same offer. Different attention-pull.

3. The low-stakes question.

The viewer needs to feel like the answer matters to their life. "What's the best wallet material?" is a question, but most wallets work fine — the stakes are low. "Are you still carrying a wallet that bends your cards and bulks your pocket every day?" names a specific physical cost. Stakes are present. Low-stakes questions earn low-stakes attention.

4. The question that doesn't match the offer.

The question opens a gap; the body has to close it with the offer. When the question is about one thing and the ad is about another, the viewer feels bait-and-switched and the trust is gone in three seconds.

"Why does everyone in your gym look better than you after six months?" opens a question about training or nutrition — the viewer expects an answer about form, programming, recovery. If the body pivots to a white-label supplement offer, the contract breaks. The question has to be the problem the product solves, not a related problem in the same zip code.

The TikTok caveat — why questions often die there

Meta feeds are text-friendly. The viewer is used to reading questions in ad copy. The format works.

TikTok's contract is different. TikTok's organic content signals are: tell me something, show me something, take me somewhere. The viewer's implicit expectation is delivery, not solicitation. A direct question — "Why does your skin still break out at 32?" — can feel like a Meta ad accidentally dropped into the wrong app. It reads as off-pattern. Hook rate suffers, especially with the under-25 cohort.

The fix is the implied-question statement: convert the question into a statement that contains the same curiosity gap without the question mark.

• "Why does your skin still break out at 32?" → "Your skin still breaks out at 32. Here's why."
• "Are you still washing your face wrong?" → "Most people are washing their face wrong. You probably are too."
• "Why do you lose weight in week one and plateau in week two?" → "Week two plateaus aren't about effort. They're about cortisol."

The gap is the same. The TikTok viewer doesn't have to answer a question — they're being told something. That's native. That's the feed contract.

For the full platform translation guide, see TikTok vs Meta hook archetypes.

Matching question hooks to Schwartz stage

Question hooks are stage-sensitive in a way operators underestimate. The same question lands differently depending on where the viewer is in their awareness.

• Unaware — weak fit. The viewer doesn't know they have the problem. A question that names the problem reads as irrelevant. They'll self-select out correctly — but you're wasting cold traffic budget. Better to lead with a Pattern Interrupt that introduces the problem visually before naming it.
• Problem-Aware — strongest fit. The viewer knows the problem, hasn't picked a solution category. A Diagnostic or Provocative question that names their exact problem in their language gets a "that's me" response. This is where the archetype earns its strongest hook rates.
• Solution-Aware — strong fit. The viewer is comparing options. A Provocative question ("Why does every [category] product still contain X?") that reframes the category can differentiate one brand in a crowded comparison set. Rhetorical questions that question the viewer's current solution also work here.
• Product-Aware — weak fit. The viewer knows your brand. Asking a question they've already answered by using your product reads as thin. Move to Bold Statement with new evidence, or lead with new social proof.
• Most-Aware — wrong fit. Retargeting visitors who've seen your landing page or carted don't need curiosity. They need friction reduction. Hit them with the guarantee, the offer, or an objection directly addressed.

The full awareness ladder is in Schwartz's 5 stages of awareness — a 2026 DTC operator's guide. The short version for question hooks: write question hooks for buyers who know they have a problem but don't know the answer yet. Anyone earlier or later is the wrong audience for the archetype.

How AdRevila scores this in your reports

When you paste a Meta or TikTok ad URL into AdRevila, a single Gemini model call watches the full video and returns a structured report. If the analyzer tags the hook as question, the report scores three specific signals:

1. Gap quality (1–5). Is the question specific enough that a "yes" response creates genuine tension? A 5 is "Why do you still feel foggy after 8 hours of sleep?" — specific symptom, implied nearby answer. A 1 is "Want more energy?" — effectively no gap.

2. Answerable-no risk (pass/fail). Can the target audience answer "no" and scroll? The analyzer cross-references the question against the inferred audience and flags this if the question likely has a self-selection escape route.

3. Stage match (pass/fail). The report cross-references the inferred Schwartz stage of the traffic with the question sub-type. A Diagnostic question on Unaware cold traffic flags as "stage mismatch — viewer lacks problem frame, convert to Pattern Interrupt or rhetorical entry." A Diagnostic question on Problem-Aware traffic passes.

The report also notes if the ad is running a question on TikTok and flags whether the statement-conversion would read more native — that's a one-edit fix most operators don't see until the platform data comes in.

<!-- TODO(embed): public AdRevila report with a question hook -->

See it in action: View the AdRevila report →

The embedded report shows a live question-hook ad scored on gap quality, answerable-no risk, and stage match — the same frame this article teaches, applied to a real ad in two minutes.

What to do this week

Five moves, one per day.

1. Monday. Pull five question-hook ads from the Meta Ad Library across the categories you care about (skincare, gut health, sleep, fitness, supplements). Transcribe the opening line. Sort each into Diagnostic / Rhetorical / Provocative / Yes-Ladder.

2. Tuesday. Run the answerable-no test on each. Could the typical viewer answer "no, not me" and scroll? Mark pass/fail. Note what made the failing hooks answerable-no — usually low specificity or a false universal assumption.

3. Wednesday. Take the two strongest and rewrite each for TikTok using the implied-question statement conversion. Read both aloud. Which reads like a native take versus a pasted Meta ad?

4. Thursday. Write one question hook for your current offer in each sub-type. Four hooks total. Run the gap-quality check on each — would a Problem-Aware buyer feel the pull to answer it?

5. Friday. Ship the strongest into a new ad set against Problem-Aware or Solution-Aware cold traffic. If hook rate is up but CVR is flat, check whether the body closes the gap the question opened. That's the most common post-hook failure for this archetype.

Related reading

• The 5 hook archetypes that govern every winning DTC ad — where question hooks sit in the full archetype vocabulary.
• How to analyze a winning Meta ad — the diagnostic rubric this archetype sits inside.
• Bold Statement hooks — the 3-second trust contract — the archetype most often confused with Provocative question hooks; different failure modes.
• Pattern Interrupt hooks — when visual disruption wins and when it fizzles — the archetype to use instead of question hooks on Unaware cold traffic.
• TikTok vs Meta hook archetypes — the platform translation guide, including the question-to-statement conversion rules.
• Schwartz's 5 stages of awareness — a 2026 DTC operator's guide — the full awareness ladder for stage-match diagnosis.
• Question hook (glossary) — the one-minute definition.

---

Run this on your own ad. Paste a Meta or TikTok ad URL into AdRevila, get the question-hook gap score and stage-fit verdict in two minutes, ship a sharper hook this week. Start free →