Guides

What a homepage must state for the brief to work

Your homepage is the brief. Nine copy checks that let an automated site read nail your offer, buyer, and tone — plus the five phrases that break the read.

By 21 Nov 2026
What a homepage must state for the brief to work

You paste your URL into the tool, wait, and the brand brief returns describing "innovative solutions for modern teams" — a line that could belong to ten thousand companies, including none of yours.

That is rarely a parsing failure. It is the predictable output of the input. Take the same page, rewrite the hero as a named buyer plus a concrete outcome — invoice reconciliation for freelance designers who bill in three currencies — and the next read returns the offer instead of the category.

Getting homepage copy for automated short form video right is an input problem before it is a tooling problem, and most vendor blogs will not tell you that because it puts the work back on the page.

Here is the roadmap: the three fields a read can actually extract, a field map from page element to brief field with the failure mode for each, and a nine-step repair sequence with before-and-after rewrites.

One finding is worth flagging now: the highest-leverage line on the page is not the hero. It is a three-adjective block most founders never write. Step five covers it.

When you are ready to test the result, add your website URL and let Vlozy build the brief, then the videos, music, captions, and posts from it. The homepage is not a brochure here — it is the input document.

What does a homepage need to state for a website read to build an accurate brand brief?

A brand brief is the structured summary a site read produces from your public pages: what you sell, who buys it, and how you sound. For that read to build an accurate brief, the homepage must state three things explicitly — offer, buyer, tone. Everything else on the page is scaffolding around those three fields.

The three-field rule: offer, buyer, tone. Stated once, in plain nouns, in the first screenful and the paragraph beneath it.

Think of a site read like a new hire's first morning. Hand them a brochure of adjectives and they nod politely and produce generic work. Hand them one page with the buyer named and the result quantified and their first deliverable lands close to right. The read has no taste to apply, only text to lift.

Page elementBrief field it fillsFailure mode when thin
Hero headlineOfferReads as a category: "solutions for modern teams"
SubheadOutcomeBrief lists features instead of results
First paragraphBuyerBuyer defaults to "businesses" or "everyone"
Pricing and nav labelsCategory anchorProduct filed under the wrong category
About pageToneNeutral voice, captions with no point of view

Four prerequisites decide whether a page is readable at all: the page is public, the copy sits in the HTML rather than being injected after load, the language is English, and pricing is either visible or marked with an explicit "talk to us" line. A page that hides all four gives the read nothing to anchor to, and no amount of rewriting the hero will fix it.

Vlozy turns your website into Shorts, Reels, and TikToks from that brief, so the fields above are the whole input surface.

How to fix your homepage copy in nine ordered steps

How to fix your homepage copy in nine ordered steps

Work top to bottom — each step repairs one brief field, and later steps depend on earlier ones. You need edit access to your homepage template or CMS page editor, your sales inbox or analytics to confirm who actually buys, and roughly an hour. No design tool, no copywriter.

  1. Rewrite the offer sentence. Open your hero block and replace three stacked headlines with one sentence of 12 to 20 words. Before: "Built for the future of work." After: "We turn your website into daily short-form video." Success looks like a stranger reading it back to you in one breath. About 10 minutes.
  2. Name the buyer noun. Type the specific person or team into the first paragraph — freelance designers, seed-stage founders, two-person marketing teams. Never "teams" or "businesses," which the read discards. You should be able to point at three real customers who match the noun. 5 minutes.
  3. Replace the subhead with the outcome. Delete the mechanism and state the result. Before: "AI-powered content pipeline." After: "A new video goes out every day without you editing anything." The mechanism belongs in a feature list. 5 minutes.
  4. Anchor the category. Put your category word in the navigation label or the pricing line, phrased the way a buyer would search it. Check that a reader landing cold knows what shelf you sit on within five seconds. 5 minutes.
  5. Add the three-adjective tone block. In your about section, write three adjectives you would accept and one you would refuse: "plain, fast, unglamorous — never playful." This is the line most briefs are missing and the one that changes your captions most. 5 minutes.
  6. Write the refused-words list. Add a short block of words you will not use: solutions, empower, all-in-one, seamless, next-gen. Tone comes from what you decline to say. You should see the remaining copy get sharper as you delete them. 10 minutes.
  7. Split or drop the second audience. Two audiences in one sentence is a failure mode — the read averages them into a buyer nobody recognises. Open the section, move the second audience to its own page, or cut it. The first paragraph should now describe one reader. 10 minutes.
  8. Move the buried outcome above the fold. Find the sentence deep in your page that actually states the result, cut it, and paste it under the hero. Reload and confirm it is visible without scrolling on a laptop. 5 minutes.
  9. Keep the category keyword in the meta title. Type the category phrase into your SEO title field and take it out of the hero. The hero sells, the meta title files. Check that the two now say different things. 2 minutes.

With those nine done, the question stops being what to write and becomes whether the read picked it up.

How do you check whether the read actually worked?

How do you check whether the read actually worked?

Run a two-pass read-back test: save the brief your URL produced, rewrite the hero so it names one buyer and one offer, then re-add the same URL and diff the two briefs line by line. The pass condition is the three-companies rule.

If the brief's one-line offer could plausibly describe three other companies in your category, the homepage is at fault, not the tool. A read can only return what the page states, so a generic output is usually an accurate mirror of a generic input.

Brief fieldBefore the rewriteAfter the rewrite
Offer line"Seamless scheduling for modern teams" — fits dozens of products"Scheduling for solo consultants who book clients by DM" — fits one
Buyer line"Teams" — the category, substituted for a persona"Solo consultants, one to three clients a week"
Tone line"Professional, modern" — a default, not a voice"Direct, dry, allergic to hype" — a voice you can write from

The same test works whichever tool does the reading. The honest comparison is less about software and more about what each option needs before it can tell you anything.

Checking methodWhat it needsBest for
Vlozy URL read-backA public site with copy in the HTMLFounders who want a fast before/after on copy alone
Human brand strategistA call, a brief, a week of turnaroundTeams with budget who need positioning decisions, not just diagnosis
Video-first tools such as OpusClipExisting footage before anything is generatedCreators with a back catalogue of recorded video

Video-first tools are genuinely good at what they do — repurposing long recordings into clips — but they cannot diagnose homepage copy, because they never read the page. Vlozy's own comparison page makes the split plain: OpusClip needs video, Vlozy starts from your URL.

Some reads cannot be fixed by copy. Private pricing behind a login, text rendered only by script so it never reaches the parser, non-English pages, and a product that is genuinely still vague all limit what any tool can extract. Naming that is cheaper than rewriting the hero five times.

A read-back test also tells you which failure you have. If the offer line changes after the rewrite but the buyer line does not, the page still lacks a named customer. If neither moves, the problem sits below the hero — in product naming, or in a category so crowded that every sentence sounds like everyone else's.

Once the two briefs diverge, you have a copy you can trust as an input, and the next step is mechanical: keep the URL in the loop and let the pipeline run from the brief that passes the rule.

Common questions about homepage copy and the brand brief

Common questions about homepage copy and the brand brief

Why did the description of my brand come back wrong?

The page never names a buyer, so the read substitutes the category. "Modern teams" is not a persona the brief can bind to, so it returns "teams in general" — accurate to your words, useless as a target. A buyer noun in the hero stops the substitution.

Where does tone come from if my homepage is deliberately neutral?

Supply it in a three-adjective block, then list the words you refuse to use. "Direct, dry, calm" gives the brief a register; refusing "empower, seamless, revolutionary" gives it an edge. The refusals carry more signal than the adjectives.

Does homepage length change the read?

One 12 to 20 word offer sentence outperforms three stacked headlines. Stacked headlines split into competing claims, and the read keeps the loudest one rather than the most accurate. Short and specific beats long and layered.

Does the brief also produce captions and music?

Yes — captions, music, and post copy are generated from the same brief. That is why editing the hero is the highest-leverage change available: it moves every downstream asset at once, not just the offer line.

The three fields, in one sentence, and what to do next

Offer, buyer, tone — a page that names all three earns a brief that sounds like the brand; a page that names none earns one that sounds like its category.

Rank the work by leverage. The hero carries both the offer and the buyer line, so rewrite it first; product naming, tone rules, and section-level phrasing are second-order edits that only pay off once the hero is specific.

The strongest next action is the loop itself: re-add your website URL, compare the two briefs against the three-companies rule, and keep the version that passes. A weaker but still useful step is shipping one video from the current brief and watching where it drifts — drift maps back to the field that is missing.

When the brief holds, the output side is already wired: turn your website into Reels and let the same brief carry into the next 14-day cycle. Add the URL, rewrite the hero, compare the briefs — the page you already have is the input, and the edit you make this week is the only variable that matters.

Turn a website into short-form video

Vlozy drafts the hook, assembles the clip, and publishes to the accounts you connect. Video can go to Reels, Shorts, TikTok, Facebook, Threads, and Pinterest. LinkedIn and X get a text post.

Get Started