Landing page audits
Find out what’s costing your landing page conversions.
Crobi audits a landing page in about three minutes and gives you a ranked list of what to change. Every finding comes with the evidence, the reasoning and the rewritten line, ready to brief or test.
Bring a URL. We’ll audit one of your pages on the call.
- Claim
“The all-in-one workforce platform for hospitality” tells an operations manager what Ferrow is, but not what changes for them. Visitors arriving from a “rota software” ad have to work out the benefit themselves.
Evidence AnalysisThe headline and subheadline contain no outcome, number or time frame.
Proposed change“Build next week’s rota in 20 minutes, and stop covering no-shows by phone.”
Confidence High- Hero value proposition
- Copy specificity
- Claim
On phones the only call to action sits below a full-height hero image.
Evidence MeasuredOn a 390 × 844 viewport the button starts at 1,140 px, 296 px below the first view.
Proposed changePlace the button directly under the subheadline, and crop the hero image to 40% of the viewport height on mobile.
Confidence High- Mobile first view
- Primary CTA clarity
- Evidence Measured
Nine required fields, including number of employees, current software and annual revenue.
Proposed changeAsk for name, work email and number of locations. Collect the rest on the call.
Confidence High- Lead form fields
- Commitment level of the ask
- Evidence Measured
Grey #9AA0A6 on white measures 2.6:1; WCAG AA requires 4.5:1 for text this size.
Proposed changeDarken the subheadline to at least #5F6368 (6.0:1).
Confidence High- Visual hierarchy and scannability
- Claim
The logos show who uses Ferrow but not what changed for them. A sceptical buyer comparing three vendors learns nothing they can repeat to their boss.
Proposed changeAdd one quote that names a result, attributed with name, role and company, for example a measured cut in weekly planning time.
Confidence Medium- Specific social proof
- Claim
There is no price, price range or pricing model anywhere on the page, so visitors can’t qualify themselves before booking.
Proposed change“From €4 per employee per month. No setup fee.”
Confidence Medium- Offer and pricing clarity
- Evidence Measured
Largest Contentful Paint is 4.6 s on an emulated mid-range phone. The hero image is a 2.8 MB JPEG.
Proposed changeServe the image as AVIF at mobile size and preload it.
Confidence High- Page speed
- Claim
A paid landing page with full site navigation gives visitors seven exits before the one action that matters.
Proposed changeKeep the logo and the “Book a demo” button; remove the other links on paid-traffic pages.
Confidence Medium- Landing page focus
- per audit
- ~3 min
- ranked findings per page
- 8–20
- conversion areas checked
- 25
- viewports: desktop and mobile
- 2
Your campaigns get optimised every day. The page they land on gets a guess.
Bids, audiences and creatives are tested constantly. The landing page, where the spend either converts or doesn’t, gets reviewed when someone finds the time. Agency audits take weeks. Generic checklists don’t know your page. Crobi looks at your actual page, on desktop and mobile, and tells you what to change first.
This is what you get.
A complete Crobi report on a demo-booking page. Click a marker or a finding to see how they connect.
ferrow.com/demo
Captured 6 Oct 2026, 10:12- Purpose
- Book a demo
- Primary conversion
- Demo request form
8 findings
1 critical2 high4 medium1 lowferrow.com/demo
8 findings · Book a demo- Claim
“The all-in-one workforce platform for hospitality” tells an operations manager what Ferrow is, but not what changes for them. Visitors arriving from a “rota software” ad have to work out the benefit themselves.
Evidence AnalysisThe headline and subheadline contain no outcome, number or time frame.
Proposed change“Build next week’s rota in 20 minutes, and stop covering no-shows by phone.”
Confidence High- Hero value proposition
- Copy specificity
- Claim
On phones the only call to action sits below a full-height hero image.
Evidence MeasuredOn a 390 × 844 viewport the button starts at 1,140 px, 296 px below the first view.
Proposed changePlace the button directly under the subheadline, and crop the hero image to 40% of the viewport height on mobile.
Confidence High- Mobile first view
- Primary CTA clarity
- Evidence Measured
Nine required fields, including number of employees, current software and annual revenue.
Proposed changeAsk for name, work email and number of locations. Collect the rest on the call.
Confidence High- Lead form fields
- Commitment level of the ask
- Evidence Measured
Grey #9AA0A6 on white measures 2.6:1; WCAG AA requires 4.5:1 for text this size.
Proposed changeDarken the subheadline to at least #5F6368 (6.0:1).
Confidence High- Visual hierarchy and scannability
- Claim
The logos show who uses Ferrow but not what changed for them. A sceptical buyer comparing three vendors learns nothing they can repeat to their boss.
Proposed changeAdd one quote that names a result, attributed with name, role and company, for example a measured cut in weekly planning time.
Confidence Medium- Specific social proof
- Claim
There is no price, price range or pricing model anywhere on the page, so visitors can’t qualify themselves before booking.
Proposed change“From €4 per employee per month. No setup fee.”
Confidence Medium- Offer and pricing clarity
Applies to the whole page - Evidence Measured
Largest Contentful Paint is 4.6 s on an emulated mid-range phone. The hero image is a 2.8 MB JPEG.
Proposed changeServe the image as AVIF at mobile size and preload it.
Confidence High- Page speed
- Claim
A paid landing page with full site navigation gives visitors seven exits before the one action that matters.
Proposed changeKeep the logo and the “Book a demo” button; remove the other links on paid-traffic pages.
Confidence Medium- Landing page focus
Ferrow is a fictional company created for this example. Its page was designed to show typical issues.
Every finding is specific enough to act on.
Ranked. Ordered by severity and expected impact, so the top of the list is where you start.
Severity. Critical, High, Medium or Low, always shown with a label as well as colour.
What’s wrong. One to three sentences, about this page, not pages in general.
Confidence. How sure the analysis is. Low-confidence findings say so.
“The all-in-one workforce platform for hospitality” tells an operations manager what Ferrow is, but not what changes for them. Visitors arriving from a “rota software” ad have to work out the benefit themselves.
The headline and subheadline contain no outcome, number or time frame.
“Build next week’s rota in 20 minutes, and stop covering no-shows by phone.”
- Hero value proposition
- Copy specificity
Why we say so. The element, the measurement, the number.
What to change. Often the literal rewritten line, set apart so it can be copied straight into a brief.
Impact and effort. A quick read on whether it’s worth doing this sprint.
The principle behind it. Each finding cites the CRO principle and research it rests on.
- Ranked. Ordered by severity and expected impact, so the top of the list is where you start.
- Severity. Critical, High, Medium or Low, always shown with a label as well as colour.
- What’s wrong. One to three sentences, about this page, not pages in general.
- Why we say so. The element, the measurement, the number.
- What to change. Often the literal rewritten line, set apart so it can be copied straight into a brief.
- Impact and effort. A quick read on whether it’s worth doing this sprint.
- The principle behind it. Each finding cites the CRO principle and research it rests on.
- Confidence. How sure the analysis is. Low-confidence findings say so.
Measured or analysed, always labelled.
Some findings are measured. Crobi renders your page in a real browser and checks positions, sizes, contrast and load times. Others are analysis: reasoning about your message, offer and trust signals against Crobi’s CRO knowledge base. Solid markers are measured. Hollow markers are analysis. You always know which is which.
How it works
-
Step 1: Enter a URL.
Any public landing page or home page, whether it’s yours or a client’s.
-
Step 2: Crobi captures and checks it.
The page is loaded in a real browser at desktop and mobile size and the cookie banner is handled. Every element is measured, and the page is analysed against Crobi’s knowledge base.
-
Step 3: You get a ranked backlog.
Typically 8–20 findings, ranked by severity and impact, in about three minutes.
25 places where landing pages lose conversions.
Each check is grounded in established CRO practice and published research on attention, persuasion and usability. Every finding cites the principle it rests on.
Message and offer6
- Hero value proposition
- Copy specificity
- Benefits over features
- Offer and pricing clarity
- Motivation and urgency
- Language and locale fit
Trust5
- Specific social proof
- Trust signals and risk reversal
- Visual credibility
- Human contact options
- Payment and delivery reassurance
Friction and the ask6
- Primary CTA clarity
- Lead form fields
- Commitment level of the ask
- Clickability and affordances
- Landing page focus
- Homepage routing
Structure, mobile and speed8
- Visual hierarchy and scannability
- Content pyramid
- Mobile first view
- How-it-works steps
- Showing the product
- Objection handling and FAQ
- Carousels and auto-motion
- Page speed
Plus measured checks on every page: CTA visibility in the first view, form length and labelling, text contrast and load time.
Send the report, not a slide deck.
Share any report as a read-only link. It opens on a phone and needs no login. It expires when you choose, from 7 to 90 days, and you can revoke it at any time. You’ll see when it’s been opened.
- Brief your developer.
- Align with your designer.
- Make the case to your boss.
- Hand your agency a backlog.
- 2 “Book a demo” is not visible in the first mobile view
- 3 The demo form asks for nine fields before a first call
Straight answers
What Crobi will and won’t tell you.
-
No promised uplift.
Crobi tells you what to test first, not what will win. Your offer, traffic and market decide that.
-
No black box.
Every finding shows its evidence, its reasoning and its source. Measured facts are kept apart from analysis.
-
No generic advice.
Findings point to an element on your page. If it would apply to any page, it doesn’t make the report.
FAQ
Who is Crobi for?
Growth and performance marketing teams at companies and startups that send meaningful paid traffic to their landing pages, where a better page means more pipeline from the same budget.
How is this different from pasting my page into ChatGPT?
A chat model sees the text you paste. Crobi loads the actual page in a browser on desktop and mobile, measures every element, and checks it against a curated CRO knowledge base. Findings point to the exact element, show their evidence and are ranked, and measured facts are kept separate from analysis.
How is it different from Lighthouse or PageSpeed Insights?
Those measure technical performance. Crobi checks load time too, but it focuses on what makes visitors convert: the message, the offer, trust, friction and the ask.
Which pages can Crobi audit?
Public landing pages and home pages. Pages behind a login, and pages that block automated browsers, can’t be audited.
Does it work on German pages?
Yes. Crobi audits pages in German and English. Reports are written in English for now.
What happens to my data?
Crobi stores the screenshots and findings of your audits in your account. A report is only visible to your team and to the people you send a link to. Processing is EU-first; see our subprocessors for details.
Why is a Crobi bot visiting my website?
Someone asked Crobi to audit one of your pages. Read what our crawler does and how to reach us: crobi.ai/bot.
What does it cost?
Crobi is a monthly subscription, priced by your traffic volume. Book a demo and we’ll walk you through it.
