VSL Script Template: The 9-Section Spine That Sells

The worst VSL script I ever reviewed had beautiful writing in it. Genuinely good sentences. The problem was structural: the offer arrived at minute 3 of a 22-minute video, the proof arrived at minute 17, and the mechanism - the reason anyone should believe the claim - never arrived at all. The writer had written scenes. He hadn't built a spine.

A VSL script is closer to load-bearing architecture than to prose. Sections carry weight in a specific order: each one earns the viewer's permission for the next, and a missing or misplaced section shows up as a cliff in the retention curve at a predictable timestamp. This is the template I use - nine sections, with word budgets, two runtime cuts, the genealogy of where each piece comes from, and the number in your analytics that tells you when a section is failing. Fill it in top to bottom and you'll have a shippable script today.

The 9-section VSL script template: hook (earn 30 more seconds), problem (name it precisely), agitation (price of inaction), mechanism story (why this works when others failed), proof (stack it), offer (value before price), objections and risk reversal, honest urgency, and the close loop. Budget roughly 150 words per minute, map sections to timestamps, and let the retention curve grade each section individually.

Where does this structure come from?

Three formulas, stacked. AIDA - usually credited to E. St. Elmo Lewis's late-1890s hierarchy, though the acronym itself dates to 1921 - supplies the attention-to-action arc. PAS, championed by Dan Kennedy as "the most reliable copywriting formula for sales ever invented," powers sections 2-4. And Star Story Solution, which Russell Brunson popularized and credits to Vince James, drives the mechanism story.

The genealogy matters because it tells you what's negotiable. The formulas survived a century of print, radio, TV, and now video because they describe how persuasion sequences, not how any medium works. What changed for VSLs is the packaging: Jon Benson is widely credited in direct-response circles with putting the sales letter on screen as text around 2007 (a credit that traces to Benson himself, so hold it loosely), and Jeff Walker's "sideways sales letter" showed the same spine could stretch across days of launch content. Same skeleton, different costumes.

Current practitioner guides converge on the same shape - Thrive Themes' 2025 guide runs nine stages, Green Frog Labs' 2026 version runs eight timed sections, Copy Posse compresses it to five questions - and the shared skeleton across all of them is exactly PAS plus proof plus offer mechanics. Nobody owns the spine. What differs is discipline in the joints, which is what the rest of this template is about.

The formulas are a century old because they describe persuasion, not the medium. Your script fails in the joints, not the bones.

What goes in each of the nine sections?

Each section below has one job, a word budget for a 12-minute cut (roughly 1,800 words total at 150 words per minute), and a fill-in prompt. Write them in order; every section assumes the viewer granted the permission the previous one asked for.

How long should each section run?

Budget roughly 150 spoken words per minute. The 12-minute cut above totals about 1,800 words; a 30-minute cut roughly triples the mechanism story, doubles proof and objections, and leaves the hook, urgency, and close nearly unchanged. The sections that grow with runtime are story and proof; the sections that never grow are hook and close.

That asymmetry is the practical difference between a short and long VSL, and it's why the length question is a scoping question rather than a style war. A high-ticket offer with a novel mechanism earns a long section 4 and a deep section 5; a simple offer to warm traffic doesn't. The full decision framework - traffic temperature, ticket, complexity, and the feedback loop that corrects your first guess - is in how long should a VSL be.

Runtime grows in the story and the proof. If your hook or your close got longer, you didn't scale the script - you padded it.

How do you know which section is failing?

Map each section to its timestamps before you publish, then read the retention curve against the map. Every section has one number that judges it: the hook is judged at second 30, the story by mid-video slope, the offer by the drop at price reveal, the close by CTA clicks per viewer who reached it.

This mapping is the entire reason to script in sections rather than in vibes. When the curve shows a cliff at 2:40, you don't have a "video problem," you have an agitation problem, and you know which 150 words to rewrite. The four launch numbers that anchor this - 30-second retention, midpoint retention, revenue per viewer, cost per qualified view - are covered in the metrics guide, and reading the section-level story out of watch density is the subject of the heatmaps guide.

Two habits that make the checkpoints work:

• Timestamp the script before recording. Write the planned timestamp next to each section header in the doc. After publishing, you're comparing plan to curve instead of scrubbing and guessing.
• Rewrite one section at a time. When a checkpoint fails, change that section and only that section, then re-measure. Rewriting three sections at once tells you nothing about which fix worked - the same discipline honest split testing demands.