A homepage teardown is a structured audit of the first screen and the path below it: who the page is for, what it promises, what proof it shows, and what it asks the visitor to do. Ten checks cover most of it, and you can complete them in under an hour. Do this before you buy a redesign, because half of what looks like a design problem is a message problem, and a new layout will not fix a message problem.
What does a teardown actually look like?
A written review of the page as it exists today, section by section, with specific findings and the change each finding implies. Not a mood board. Not "make it more modern". If a finding cannot be written as an instruction someone could execute, it is not a finding.
The 10-point homepage teardown checklist
- Is the audience named in the first line? A visitor should know within five seconds whether this page is for them.
- Does the headline state an outcome or a capability, not a slogan? "Bookkeeping for restaurants with one location" beats "Reimagining finance".
- Is there one primary action? Count the buttons in the hero. More than one competing action means no action.
- Is the offer concrete? What is being sold, what does it cost, what happens after the click. Vagueness reads as risk.
- Is there proof above the fold? Real output, a customer quote, a number you can defend, a screenshot. Placeholders do not count.
- Does the page answer the obvious objection? Usually price, trust, effort or time to value. Objections do not disappear because the page ignores them.
- Is the structure scannable? Headings that make sense on their own, short paragraphs, no wall of text.
- Are the visual choices consistent? Type scale, spacing, colour use, button style. Inconsistency reads as unfinished.
- Is the language free of internal jargon? Words your team uses in meetings are often words your buyer has never seen.
- Do the basics of accessibility hold? Contrast, focus states, alt text, headings in order. Useful reference points are the WAI guidance at w3.org/WAI and the research library at nngroup.com.
Write each check as pass, weak or fail, then rank the failures by how much they cost you in lost clarity. Fix the top three before anything cosmetic.
Fix or redesign: which one do you need?
| What you see | Likely fix |
|---|---|
| Traffic arrives but nobody clicks the main button | Rewrite the hero and the offer |
| Visitors ask the same question in email | Add a section that answers it plainly |
| Page looks dated but the message is clear | Visual refresh |
| Message is unclear and the page also looks dated | Message first, visuals second |
| One section is weak, the rest works | Targeted rewrite of that section |
A redesign that starts with layout tends to reproduce the original message problem with nicer type. Message work is cheaper and shows up faster in behaviour.
How do you review a homepage you wrote yourself?
Badly, usually, and that is normal. You know what you meant. The remedy is to read the page as a stranger would:
- Read only the headlines in order. If they do not tell a coherent story on their own, neither does the page.
- Read the first line out loud to someone who does not work with you and watch their face.
- Remove your logo and product name from a screenshot. Does it still make sense?
- Ask three people in your target audience what the page sells and what it costs.
What should a teardown deliverable contain?
A useful one has four parts: the findings in priority order, the specific replacement copy or layout change for each, a short list of things to leave alone, and a clear note about what the reviewer could not assess from the screen alone. That last part matters. Nobody can judge page speed, mobile behaviour or analytics from a screenshot, and a review that pretends otherwise is guessing.
Designful's homepage teardown feature is one of five fixed jobs sold at $49 one-time each. Every feature has a free preview of real output, watermarked where it includes images, so you can read the actual findings before paying. Paid deliverables arrive clean and at full resolution, and you own them outright with full commercial rights. One honest caveat: this is model-generated work built on structured recipes, not a human designer studying your analytics. If you need someone to instrument your funnel, that is a different purchase.
Other single jobs in the catalog include a logo refresh, a hero rewrite, a social launch kit and an ad creative pack. Any three are $99, or all five for $149, all one-time with no subscription. The catalog and sample output are at /.
How much should this cost?
Market comparisons are rough, and you should treat them as ranges rather than facts: agencies commonly quote $2,000 or more and multi-week timelines for brand or site work, freelancers often land around $300 and about a week. A one-time job at a fixed price is a different product from a retainer relationship, and cheap does not always mean better. If your homepage drives most of your revenue, a person who studies your data may be worth the money.
When is a teardown the wrong purchase?
Skip it if any of these are true:
- You have not launched yet and have no real page to review.
- You already know the problem and just need it executed.
- You need legal, medical or regulated claims reviewed by a professional.
- You expect a conversion outcome rather than a clear improvement in message.
For the first case, write the page first. Designful's hero rewrite feature exists for pages that are live but unclear, and a teardown is for pages that are live and unclear about why.
FAQ
How long does a homepage teardown take to receive? Deliverables are generated and delivered through a private link after purchase. Timing depends on the job, so check the feature page for the current turnaround.
Can I use the output commercially? Yes. Full commercial rights on everything produced.
Do I get a refund if I do not like it? All sales are final, which is why the free preview exists. Read the preview before you pay.
Does the teardown check mobile or page speed? Those cannot be assessed from the inputs used, so the output says so rather than guessing.

