Product Video Script#

Introduction#

Short-form video is the highest-leverage sales channel in ecommerce, but the difference between a video that converts and one that gets scrolled past is decided in the first second. A script has to earn attention before it earns trust, demonstrate value before it asks for anything, and make the purchase feel like the obvious next step — all inside 15 to 60 seconds. This skill generates platform-optimized, shot-by-shot video scripts for TikTok, Instagram Reels, and YouTube Shorts using direct-response frameworks adapted to how each platform's audience actually behaves.

The output is not a vague creative brief. It is a production-ready script: a scroll-stopping hook, beat-by-beat spoken lines, on-screen text, visual direction for each shot, a caption, and a hashtag set — everything a creator, UGC partner, or in-house editor needs to shoot and publish without guessing.

Quick Reference#

DecisionStrongAcceptableWeakHook timingTension or payoff promised in 0–1s, on screen + spokenHook lands by 2–3sSlow brand intro / logo firstHook typePattern interrupt tied to a real viewer pain or desireGeneric "check this out""Hi guys, today I want to talk about…"Framework fitChosen for product + awareness stage (see references)Default hook-demo-offerSame template for every productPacingNew visual or idea every 2–3s; cuts on the beatOne scene change mid-videoSingle static talking-head 30s+ProofSpecific demo, result, or number shown on screenVerbal claim onlyNo proof, adjectives onlyCTAOne clear action matched to the platform's buy pathSoft "link in bio"Multiple competing asksLength vs platform15–34s TikTok/Reels, 20–45s ShortsSlightly over60s+ with no payoff reasonOn-screen textReinforces (not repeats) audio; readable 2+ sCaptions onlyWall of text / none

What this skill solves#

• The skipped hook — videos that lose 60%+ of viewers before the value lands because they open with a logo, a greeting, or a slow pan instead of a reason to stay.
• Framework mismatch — applying a testimonial structure to an impulse-buy gadget, or a fast unboxing to a considered, high-ticket purchase.
• Feature dumping — scripts that list specs the seller cares about instead of the one transformation the buyer cares about.
• Weak or buried CTAs — ending on "thanks for watching" instead of a single action wired to the platform's purchase flow (TikTok Shop cart, link sticker, pinned comment).
• Platform tone-deafness — reusing one cut across TikTok, Reels, and Shorts when each rewards different pacing, sound, and sell intensity.
• Creator ambiguity — handing a UGC partner a paragraph of vibes instead of exact lines and shots, producing inconsistent, off-brand footage.
• Low completion / weak retention signals — no pattern interrupts, curiosity loops, or open loops to hold watch time, so the algorithm never amplifies the video.

Workflow#

1. Clarify the brief. Confirm the required inputs: product_name and details, target_platform, target_audience, and video_length. Capture optional script_style and key_selling_points. If the product type or platform is missing, ask — they change everything downstream.

2. Diagnose awareness stage. Decide whether the audience is problem-unaware, problem-aware, solution-aware, or product-aware. This determines how much education the hook must do before the sell. See references/short-form-frameworks.md.

3. Select the framework. Map product + awareness + platform to one of: hook-demo-offer, problem-agitate-solve, before-after-bridge, unboxing-reaction, or testimonial-story. Default to hook-demo-offer for visual impulse products; use problem-agitate-solve for products that fix a specific frustration.

4. Engineer the hook. Write 3 hook options. Each must promise tension or payoff in the first second, work as spoken line AND on-screen text, and avoid brand-first openers. Pick the strongest; keep the other two as A/B alternates.

5. Build the body beat-by-beat. Structure the middle around the framework, changing the visual every 2–3 seconds. Translate each key_selling_point into a benefit the viewer feels, and attach a concrete proof shot (demo, result, before/after, or number).

6. Land the CTA. Choose one action matched to the platform's buy path and add a light urgency or curiosity trigger that is honest (genuine scarcity, a real bonus, or a "watch til the end" loop you actually pay off).

7. Format for production. Render the script in the shot-by-shot table from references/output-template.md: timecode, spoken line, on-screen text, visual/shot direction. Add caption + hashtags, and run assets/quality-checklist.md before delivery.

Inputs#

• product_name / details (required): What it is and the one thing it does best. Example: "Mello — memory-foam house slippers, machine washable, rated 4.8★ across 12,000 reviews."
• target_platform (required): "TikTok", "Instagram Reels", or "YouTube Shorts". Structure, pacing, and sell intensity differ by platform.
• target_audience (required): Demographics, the pain or desire, and the buying trigger. Example: "Women 25–40 who work from home and want comfortable but presentable footwear for video calls."
• video_length (required): Target seconds, e.g. "30" or "45".
• script_style (optional): hook-demo-offer · problem-agitate-solve · before-after-bridge · unboxing-reaction · testimonial-story. If omitted, the skill picks the best fit.
• key_selling_points (optional): Top 2–3 benefits to emphasize, in priority order.

Worked example 1 — TikTok, impulse product (hook-demo-offer, 30s)#

Brief: Mello washable memory-foam slippers; TikTok Shop; women 25–40 working from home; 30s; emphasize machine-washable + all-day comfort.

Hook (0–2s): Spoken: "The reason your house shoes smell after two weeks?" On-screen: "POV: you never washed your slippers 🤢" Visual: hand pulls a worn slipper toward camera, hard cut.

Body (2–22s): Quick reveal of Mello → toss both slippers straight into the washing machine (pattern interrupt, the "wait, you can do that?" moment) → cut to fluffy, dry, like-new pair → wear them on a mock video call, feet flexing on memory foam. On-screen text stacks the proof: "Machine washable" → "Memory foam" → "4.8★ · 12,000 reviews."

CTA (22–30s): Spoken: "Tap the orange cart before the restock sells out." On-screen: "🛒 Tap to grab a pair." Visual: finger points to cart icon; product flat-lay end card.

Caption: Slippers you can actually throw in the wash 🧺 #tiktokshopfinds #homefinds #cozyseason Why it works: the hook weaponizes a real disgust trigger, the wash shot is the curiosity payoff, and the CTA names the exact buy action.

Worked example 2 — YouTube Shorts, considered purchase (problem-agitate-solve, 45s)#

Brief: "FocusDesk" standing desk converter; YouTube Shorts; remote workers 30–50 with back pain; 45s; emphasize 30-second setup + sturdy at standing height.

Hook (0–2s): Spoken: "If your back hurts by 3pm, it's not your chair." On-screen: "The 3pm back-ache fix." Agitate (2–14s): show slumping at a desk, clock jumping 9am → 3pm, hand rubbing lower back — name the cost ("8 hours seated wrecks your posture"). Solve (14–34s): lift FocusDesk into standing position in one motion, count "thirty… seconds," tap the surface to show zero wobble with a monitor + laptop on it. On-screen proof: "No tools" → "Holds 33 lb" → "Sits + stands." Bridge (34–45s): Spoken: "Alternate sitting and standing and the 3pm crash disappears." CTA: "Link in the description — read the 30-day reviews first." Why it works: Shorts viewers tolerate a few more seconds of setup, so the agitate beat earns the solution; the CTA invites research, which suits a higher-consideration buy.

Common mistakes#

1. Brand-first opening. Logos, "welcome back," and slow pans burn the only second that matters. Lead with tension or payoff.
2. Hook that doesn't match the body. A shocking hook followed by a flat demo feels like bait. The payoff must deliver on the hook's promise.
3. One framework for everything. Impulse gadgets and high-ticket considered purchases need different structures and lengths.
4. Listing features, not benefits. "33 lb capacity" is a spec; "holds your monitor and laptop without wobbling" is a benefit. Show, then name.
5. No on-screen proof. Claims without a visible demo, result, or number read as ads and get scrolled.
6. Reusing the exact cut across platforms. Re-pace and re-tone for TikTok vs Reels vs Shorts instead of cross-posting one master.
7. Multiple CTAs. "Follow, like, comment, and buy" splits attention. Pick one action tied to the buy path.
8. Dead air / static talking head. No visual change for 5+ seconds tanks retention. Cut every 2–3s.
9. Fake urgency. Invented countdowns erode trust and risk platform policy. Use real scarcity or a real bonus, or none.
10. Unreadable on-screen text. Tiny, fast, or audio-duplicating captions waste the visual channel. Keep text on screen 2+ seconds and complementary to speech.

Resources#

• references/short-form-frameworks.md — the five frameworks, when to use each, and awareness-stage mapping.
• references/platform-playbook.md — TikTok vs Reels vs Shorts: pacing, length, sound, sell intensity, and CTA paths.
• references/output-template.md — the shot-by-shot script template and caption/hashtag block to deliver in.
• assets/quality-checklist.md — pre-delivery quality gate across hook, structure, proof, CTA, and platform fit.