Case study: Design Build London, one month after launch

Design Build London is an architectural design and build firm working across south and east London, covering Greenwich and Blackheath, Clapham and Battersea, Herne Hill and Dulwich, and Kennington and Stockwell. They handle planning, design and construction on projects like loft conversions in conservation areas.

We rebuilt their website and we report on it monthly. This is the first month of data, taken straight from Google Search Console, and we are publishing the uncomfortable numbers alongside the good ones.

The starting point, honestly

The site we replaced was not a disaster. That is worth saying, because most web design case studies open with a horror story.

It was a competent WordPress build. It had a mobile viewport, a tap-to-call link, structured data, a canonical tag and a meta description. Someone had done a reasonable job.

What it was, was heavy. The homepage delivered 346 KB of HTML and loaded 54 separate scripts. For a business whose customers are homeowners searching on a phone, that weight is the problem, not the design.

What changed

The rebuild targeted weight and structure rather than looks:

• Homepage HTML went from 346 KB to 116 KB, roughly three times lighter.
• Scripts went from 54 to 17.
• Images cut from 29 to 14, with 10 of those lazy-loaded so they only download when scrolled to.
• A real page for each London area served, rather than one page listing them all.
• Structured data, Open Graph tags and canonical tags across the site.
• Five sitemaps covering 43 addresses, all read successfully by Google.

The point of that work is not a score in a testing tool. It is that a homeowner standing in a half-finished kitchen on 4G gets the page before they give up.

Month one: the actual numbers

Reporting period 29 June to 26 July 2026, compared against 1 to 28 June. All figures from Google Search Console.

• Average position: 34.1 to 23.3. An improvement of 10.8 places. In plain terms, roughly page four of Google to roughly page three.
• Clicks from Google: 10 to 17. Up 70 per cent.
• Impressions: 6,662 to 6,510. Down 2.3 per cent.
• Click rate: 0.2 per cent to 0.3 per cent.
• 354 different search terms brought the site up in Google.
• 21 pages appeared in search results.
• No errors, penalties or security issues.

Now the honest reading of that, which is the same reading we gave the client.

Seventeen visits is not a lot. Seventy per cent growth on ten visits is seventeen visits. We are not going to dress that up, and any agency showing you a percentage without the underlying number is hoping you will not ask.

The number that actually matters here is the position change. Rankings move first, then visibility, then clicks. That is the sequence, and moving 10.8 places in a month on a site that has only just gone live on its own domain is the sequence working correctly. The clicks follow the rankings, not the other way round.

Impressions held flat, down 2.3 per cent. On a young site that is unremarkable. We flagged it rather than hiding it because a client who only ever hears good news stops believing any of it.

Where the movement actually happened

The interesting detail is not the headline figure, it is which pages woke up.

• About Us: 1,915 impressions, up 415.
• Our Approach: 245 impressions, up 228, from almost nothing.
• Locations: 215 impressions, up 201, from almost nothing.
• Contact Us: 199 impressions, up 188, from almost nothing.
• Home: 5,989 impressions, down 292.

Four pages went from effectively invisible to hundreds of appearances each. That is the site broadening out from being one page Google recognises into several. For a business that wants to be found in specific London areas, that spread matters more than the homepage number.

On search terms, “design build london” produced 5 clicks from 358 impressions. “design and build london” produced 750 impressions and no clicks yet, which tells you exactly where the next bit of work goes.

What is not fixed yet

Fourteen pages are fully listed in Google. Twenty-eight are found but not yet listed, and seven were read but not listed.

That is normal for a site of this age and it clears as the domain earns trust, but it is not nothing. It means most of the site cannot appear in search results today. The fix is fresh content and stronger internal links to the pages that are waiting, which is what the next month is for.

The agreed actions were: one new article, set up the Google Business Profile, switch on visitor analytics, and strengthen the links from the homepage to each London area page. For a London builder, the Google Business Profile is the highest value item outstanding, because that is what puts a business in the map results and it is where reviews live.

What this costs

This is a pay monthly website. There was no five figure upfront bill, and the monthly reporting is part of the plan rather than an add-on. Plans start at £32 a month with hosting, SSL, updates and support included, and the client owns their domain and their content.

We mention that because the alternative quote for work of this kind is usually several thousand pounds before anything goes live, followed by an invoice every time something needs changing.

The honest summary

One month in: rankings up nearly eleven places, four pages moved from invisible to visible, the site is three times lighter than the one it replaced, and there are seventeen visits to show for it.

Anyone promising a small business page one and a full diary inside a month is either lucky or lying. What good work looks like at week four is the foundations moving in the right order, and that is what this is.

We will publish the next report at the end of August, including the bits that do not go to plan.

If you want the same thing for your business, get in touch or see what is included on our pay monthly plans.