Landing Page Value Review
Use this skill when
A buyer needs to know whether a stranger arriving from an ad can tell who the page serves, what it does, and what happens next. Each fix is returned as a hypothesis to test.
Common requests:
- "Review our landing page."
- "Why does the page not convert?"
- "Is the value clear?"
- "What is missing above the fold?"
Send these jobs to another skill:
- Ad wording versus page wording:
ad-to-page-promise-match-saas-ads. - Form fields and signup steps:
signup-form-friction-review-saas-ads. - Plans and prices:
pricing-page-review-saas-ads. - Choosing between a trial and a demo:
trial-or-demo-path-check-saas-ads.
Required input
Work from pasted exports and screenshots. No live account access is needed. If an input is missing, stop that finding, name the missing input, return needs_data, and do not estimate. A count below a volume floor returns needs_data (T01).
Required
- Full-page screenshots on desktop and on mobile.
- Page text in page order, with button labels.
- Ad that sends the traffic.
- Product described in one sentence.
- Proof the business can show: customer names with permission, numbers with a source.
- Main objection that sales hears.
Optional
- Weekly signups and paid customers for the page, from analytics and billing exports.
- Trial length, if the page leads to a trial.
Export route:
- Copy the final URL from the ad in Google Ads or Meta Ads.
- Open the URL and capture full-page screenshots at desktop width and at phone width.
- Copy the page text and the ad text.
- See
../saas-ads-shared/references/getting-your-data.md.
Analysis workflow
Read ../saas-ads-shared/references/thresholds.md before step 1.
- Read the ad, then read only the mobile first screen. As a stranger, answer three questions: who is it for, what job does it do, and what happens after the click?
- Repeat on desktop and note anything the mobile screen lacks.
- Match every claim on the page to supplied proof. Mark each unsourced claim as unsupported.
- Check whether the page answers the main objection, and where.
- Check each call to action. Does it name the next step, repeat after each major section, and keep one label?
- Compare page length with the size of the decision. A self-serve trial and an annual contract need different amounts of reading.
- Sort each finding into clarity, proof, or friction. State each finding as a hypothesis. Measure it on signups first, then on paid rate in cohorts matured under T02.
- Rewrite page lines only from supplied facts. Mark any line that needs a missing fact as
needs_data.
Decision rules
Thresholds: ../saas-ads-shared/references/thresholds.md. Labels and tags: ../saas-ads-shared/references/output-standard.md. What a screenshot shows is confirmed. Any effect on signups stays possible until a test measures it.
| Pattern | Fix type | Verdict | Next step |
|---|---|---|---|
| Mobile first screen lacks audience, job, or next step | clarity | possible | Rewrite headline from supplied facts |
| Claim without a supplied source | proof | needs_data | Remove it, or supply source and permission |
| Supplied proof absent or buried | proof | possible | Place it near the first call to action |
| Main objection unanswered | proof | possible | Answer it in the buyer's words |
| Call to action labels conflict | friction | possible | Use one label that names the next step |
| Signups rose, paid cohorts immature under T02 | test | monitor | Wait for matured cohorts |
| Signups rose, matured paid rate fell | test | hold | Recheck audience and promise |
| Signups and matured paid rate both rose | test | approval_needed | Present to the page owner |
Output format
## Landing page review: [URL], [traffic source]
Ad: [headline] | Product: [sentence] | Proof supplied: [list]
Mobile first screen: who [yes/no], job [yes/no], next step [yes/no]
### Decision table
| Finding | Evidence | Verdict | Severity | Confidence | Business impact | Next step |
|---|---|---|---|---|---|---|
### Rewrites from supplied facts
Current: [line] | Proposed: [line] | Source: [fact]
### Test plan
[hypothesis, fix type, metric: signups then paid rate per T02]
### Data gaps
[missing input and the verdict it would firm up]
Practical example
Illustrative only: the page, ad, and facts below are made up. Product: "Shift scheduling software for clinics with 5 to 50 staff." Proof supplied: Clinic A, named with permission, reports that weekly schedule building fell from 6 hours to 2.
- Clarity: the mobile first screen shows "Work smarter, together" and a photo, with no button.
screenshot,possible. - Proof: "Trusted by thousands of teams" has no supplied source.
page_text,needs_data. - Friction: the buttons read "Get started" twice and "Talk to us" once.
page_text,possible.
Headline rewrite, using supplied facts only: "Shift scheduling for clinics with 5 to 50 staff. Clinic A cut weekly schedule building from 6 hours to 2." That is 4 hours saved per week. Test the rewrite on signups, then on paid rate in cohorts matured under T02.
Common failure modes
- Quoting conversion rates or benchmarks that nobody supplied.
- Writing proof the business cannot show.
- Judging the mobile page from a desktop screenshot.
- Reviewing the page without the ad that sends the traffic.
- Calling a rise in signups a win before paid cohorts mature under T02.
Route to next
Ad and page mismatch: ad-to-page-promise-match-saas-ads. Form friction: signup-form-friction-review-saas-ads. Price questions: pricing-page-review-saas-ads. Trial or demo path: trial-or-demo-path-check-saas-ads.
Guardrails
- Never change a live account, budget, bid, tag, page, price, or form.
- Never invent numbers, customer names, quotes, or proof.
- End every review in an approval-ready recommendation for the page owner.
- Never quote conversion statistics, benchmarks, or expected lifts.
- Never use a customer name without supplied permission, or a number without a supplied source.
- Never call a page finding a cause until a test has measured signups and matured paid cohorts (T02).
- Never request or echo names, emails, or phone numbers. Ask for them removed before per-customer rows are pasted.
- If the account is ecommerce or lead-gen rather than SaaS, say so and stop.