Landing page checklist

This is the list I go through before a landing page receives a single paid click from Google or Meta. It is written for UK businesses that send ad traffic to a dedicated page, whether you built the page yourself, a developer built it, or an agency handed it over. Every item is something you can check in a browser, on your phone and in your ad account, without special software.

How to work through this list

Sections 1 to 8 are pre-launch checks, roughly in the order a visitor experiences the page: the ad, the first screen, the offer, the proof, the form, the load, the legal small print, then the tracking that tells you whether any of it worked. Section 9 covers what to change once real traffic arrives.

Open the ad and the page side by side, and do the whole check twice: once on a laptop, once on a mid-range phone on mobile data rather than office wi-fi. Mark each item yes, no or not relevant. Any “no” in sections 7 or 8 should stop the launch, because a page that breaks the rules or cannot report a conversion costs money you will never be able to account for.

This checklist covers the page. For the ad account behind it, use my Google Ads account checklist or the Facebook and Instagram ads checklist.

1. The ad and the page say the same thing

Someone clicked because the ad promised something specific. If the page opens with something else, they assume they have landed in the wrong place and leave. Getting this message match right is the cheapest improvement on this list.

• The headline repeats the promise of the ad. If the ad says “Boiler service in Croydon from £85”, the page headline mentions a boiler service and Croydon, not “Welcome to our heating company”.
• The price, discount or offer in the ad appears on the page With the same figure and the same conditions. A price that appears in the ad but not on the page is a complaint waiting to happen.
• One page per distinct promise. If one ad group sells emergency call-outs and another sells annual servicing, they need separate pages or at least separate headlines.
• The visual carries over from Meta ads. On Facebook and Instagram the image does most of the selling, so the same product, person or colour scheme should appear on the first screen.
• The search term fits the page. Read the search terms report after the first few days. Clicks from searches the page does not answer are a keyword problem, not a page problem.

Google also judges this. Landing page experience is one of the three parts of Quality Score, and a page that does not answer the search costs more per click as well as converting worse.

2. The first screen

Look at the page on a phone before scrolling. A visitor should be able to answer four questions from that screen alone.

• What is this? A plain headline naming the product or service. No slogans that need decoding.
• Is it for me? The location, type of customer or problem, stated in the headline or the line under it. “For landlords in Leeds” does more than a stock photo of a smiling family.
• Why this one? One concrete reason: a fixed price, a same-week appointment, free delivery over a stated amount, a guarantee you actually honour.
• What do I do now? A button that says what happens when it is pressed (“Get a quote in 2 minutes”, “Book a free survey”), visible without scrolling.

Things that should not take up the first screen: a cookie banner covering half the page, a full-width image with no text, an auto-playing slider, or a navigation menu with twenty links to the rest of the site. A landing page has one job, so remove or shrink anything that offers a way out before the visitor has read the offer.

3. The offer and the price

A landing page that looks tidy can still fail because of the offer. If the page gives no reason to act now, the visitor goes back and compares three other businesses first.

• The offer is specific. “Free 20-minute consultation by video” is an offer. “Get in touch to discuss your needs” is not.
• Prices are shown where you can show them. If the work genuinely varies, give a starting price or a typical range and say what moves it. Hiding all pricing sends price-sensitive visitors back to the search results.
• Consumer prices include VAT. If you sell to the public, the price you show must be the price they pay. Business-to-business pages can show prices excluding VAT, provided they say so clearly.
• Conditions sit next to the offer Not in a footnote three screens down: minimum spend, areas covered, end dates, what “free” includes.
• The next step is low-commitment for a cold visitor. Someone who has never heard of you is more likely to request a quote than to pay a deposit on a first visit.

4. Proof that holds up

A visitor from an ad has no reason to trust you yet. Proof closes that gap, but only if it is specific and real.

• Reviews link to where they live. A star rating with a link to your Google Business Profile, Trustpilot or Checkatrade page is worth more than anonymous quotes.
• Testimonials are genuine and attributable. Use a first name, a town and the job done, with the customer’s permission. Fake or invented reviews are banned outright under the Digital Markets, Competition and Consumers Act 2024, and that includes hiding the fact that a review was paid for or incentivised.
• Memberships and accreditations are current and link to the register where one exists, such as Gas Safe, the FCA register or the SRA.
• Photos are of your work, your premises or your team. Stock photography is easy to spot and does nothing for trust.
• Every claim can be backed up. “Rated number one in Bristol” needs a source. If you cannot show where it comes from, remove it.

5. The form, the phone number and other ways in

The form is where most paid traffic is lost, and it is also the easiest part to test yourself. Fill it in on a phone as if you were a customer.

• Ask only for what you need to reply. Name, one contact method and a short description is enough for most enquiries. Every extra required field is another reason to give up.
• Labels stay visible above each field, rather than placeholder text that disappears when someone starts typing.
• Fields use the right mobile keyboard A number pad for phone numbers and an email keyboard for email addresses. Postcode fields accept UK formats with or without a space.
• Error messages say what is wrong and keep everything the visitor already typed.
• The button says what happens next (“Send my quote request”), not “Submit”.
• A phone number is tappable on mobile, in UK format, and is answered during the hours the ads run. If nobody picks up after 5pm, schedule the ads or show the number only in office hours.
• There is an alternative for people who will not fill in forms A phone number, WhatsApp or a booking calendar, depending on how you actually prefer to take enquiries.
• The confirmation is useful. After submitting, the visitor sees what happens next and when, and gets an email copy if you collected an email address.
• Someone receives the lead. Send a test enquiry and check it arrives in the inbox or CRM it should, not a junk folder or an old employee’s mailbox.

6. Speed and mobile

Check your own ad reports for the device split: for many campaigns, especially on Facebook and Instagram, most clicks come from phones, often on patchy mobile data. A page that takes several seconds to show anything loses visitors you have already paid for.

• Run the page through PageSpeed Insights and look at the mobile result. Google’s published “good” thresholds for Core Web Vitals are a Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint within 200 milliseconds and Cumulative Layout Shift under 0.1.
• The hero image is compressed and sized for the screen In a modern format such as WebP. An uncompressed 3MB photograph at the top of the page is one of the commonest causes of a slow first load.
• Nothing jumps as the page loads. Buttons that move just as someone taps them cause mis-taps and frustration.
• Chat widgets, pop-ups and third-party scripts earn their place. Each one slows the page. Remove anything nobody uses.
• Text is readable without zooming Buttons are big enough for a thumb, and nothing overflows the screen sideways.
• The page works when opened inside the Facebook or Instagram app. The in-app browser handles some scripts and logins differently, so test it from a real ad preview, not only in Chrome or Safari.

7. UK rules: claims, privacy and consent

This is the section most often skipped, and the one that can turn a weak page into a liability. None of it is legal advice, but these are the checks I would expect any UK landing page to pass.

8. Tracking before the first click is paid for

If the page cannot report a conversion, the ad platforms cannot optimise towards one and you cannot tell which ads pay for themselves. Set this up and test it before launch, not after the first invoice.

• Every way in is tracked: form submissions, phone calls (clicks on the number, or call tracking if calls matter), WhatsApp clicks and bookings.
• Each conversion fires once, at the right moment. Count the confirmed submission, not the button click, and make sure refreshing the thank-you page does not count it again.
• Google Ads click IDs survive. Redirects, cookie banners and form plugins sometimes strip the parameters that connect a click to a conversion.
• Meta and other non-Google links carry UTM tags so GA4 credits the right campaign. My UTM link builder keeps the naming consistent.
• Enhanced conversions or the Conversions API are considered, if consent and your privacy notice allow it, to recover conversions that browser tracking misses.
• You have sent a real test. Click a test link, submit the form, and watch the conversion appear in Google Ads, Meta Events Manager and GA4. My guide on checking that conversion tracking works walks through each platform.

9. After launch: what to test, and in what order

Once traffic arrives, change one thing at a time, starting where the biggest losses usually are.

1. Traffic quality. Before touching the page, read the search terms and placements. A page cannot convert people who were never looking for what you sell.
2. The offer and headline. These move results far more than button colours or layouts. Test a different offer, or a headline aimed at a different problem.
3. The form. Fewer fields, a two-step form, or a phone-first option for urgent services.
4. Proof and its position. Moving reviews higher or swapping generic praise for specific outcomes customers describe.
5. Layout and design details Last.

Be honest about traffic. Many small UK accounts do not get enough conversions for a split test to reach a reliable answer within a few weeks. Before you declare a winner, put the numbers into my A/B test significance calculator. If the result is not significant, make bolder changes and compare whole periods instead, noting the date of each change.

The faults that cost the most

These faults waste the most money because the ads keep spending while they go unnoticed. If you only have an hour, check these first.

• Ads sent to the homepage. The visitor has to find the relevant service on their own, and many will not bother.
• A form that silently fails Or delivers to an inbox nobody reads, so enquiries vanish without anyone noticing.
• Conversions counted twice, or not at all So the bidding algorithms learn from the wrong signal.
• A mobile page that loads slowly or hides the button below a large image.
• An offer that matches nothing in the ad Usually because the ad was updated and the page was not.
• Tags firing before consent Which is a compliance problem first and a data quality problem second.

When to ask for help

If the checklist turns up more “no” answers than you can fix, or the page needs rebuilding rather than adjusting, I design and write landing pages for Google and Meta ad campaigns, with the tracking and UK compliance checks above built in. If you would rather start with a second opinion, send me the page and the ad and I will tell you what I would change first.

Updated 6 October 2026