Do you need a new website if you’re already running ads?

The short answer: not automatically.

Running ads is not, by itself, a reason to commission a new website. Start by checking whether enquiries are being measured correctly, whether the page matches the offer in the ad, and whether a customer can complete the next step on a phone. A focused repair may be enough. A rebuild is a larger decision about the site’s limitations, not a default response to disappointing campaign results.

For a UK service business, the website is usually one part of a longer buying journey: a visit, an enquiry, a response, a quote and a booking. A weak result at the end can originate at any step. This guide explains how to locate the problem before choosing a solution.

1. Check that a lead is actually being counted.

Submit a test enquiry through the page people reach from your ads. Check that the confirmation appears, the business receives it, and the conversion is recorded in the intended system. A button click and a successful form submission are different events; counting the former as a lead can make a campaign look better than it is.

Google Ads supports measurement for website actions, calls and offline conversions.^[1] That makes it possible to look beyond a simple click count. Ask your campaign manager exactly what the reported conversion means. Is it an enquiry received, an answered call, a qualified lead or a booked job?

Also check the less glamorous details: messages in spam folders, an outdated destination address, a form that fails on mobile, or a business that cannot respond quickly. A redesign cannot fix an enquiry that never reaches the right person.

2. Match the landing page to the reason for the click.

If an ad offers boiler servicing in a particular area, the landing page should explain that service and area. A generic homepage makes the visitor find the relevant information again. The useful question is whether the page continues the conversation the ad started.

Google describes landing-page experience as one of the components of Quality Score, alongside expected click-through rate and ad relevance. Quality Score is a diagnostic tool, not a business outcome or a guarantee of lower costs.^[2]

Review the first screen and the next few sections. Can someone understand the service, see whether you cover their location, recognise credible evidence, and find a sensible next step? If the answers are unclear, the first experiment may be a revised landing page rather than a whole new site.

3. Test the journey on a real phone.

Try the page on mobile data as well as Wi-Fi. Check the menu, text size, call link, form fields, validation and confirmation. Watch for keyboard obstruction, horizontal scrolling, a sticky banner over the submit button, or images that move the form while it loads.

Google’s Core Web Vitals assess loading, interaction responsiveness and visual stability. The published “good” thresholds are LCP at or below 2.5 seconds, INP at or below 200 milliseconds, and CLS at or below 0.1, assessed at the 75th percentile of visits.^[3] These are experience indicators, not proof that a page will convert.

Use a performance report to locate the cause. A large hero video, third-party scripts or oversized images can slow a custom-coded site too. WordPress, a builder or static HTML is not a diagnosis on its own.

4. Decide between a repair and a rebuild.

Compare the options against a written brief. Include content migration, URLs and redirects, contact flows, account access, analytics, accessibility and ongoing support. A cheaper headline quote may exclude the work needed to protect the existing customer journey.

5. Keep the financial test honest.

Estimate whether an improvement is commercially useful with your own data. For example, 200 relevant visits producing 4 enquiries is a 2% visit-to-enquiry rate. If a change produced 6 enquiries from comparable traffic, the observed rate would be 3%. That is a worked example, not a forecast.

Then ask what those enquiries become. If they are outside your service area or cannot afford the service, a higher enquiry rate may not help. Google distinguishes qualified leads and converted leads, which can support a more useful view of the sales process.^[4]

What to ask an agency before agreeing to a rebuild.

• Which problem have you identified, and what evidence supports it?
• Could a smaller repair solve it?
• What is included in the fee, and what is separate?
• How will you test forms and preserve important URLs?
• Who owns the files and accounts after payment?
• What will you measure after launch, and when will you review it?

At AllVia-AI, standalone websites start from £799 and website-plus-marketing packages start from £1,000/month, with media spend separate. The useful first step is reviewing the current journey, then agreeing the scope. A rebuild should have a clear reason.

Questions people ask