Most builders we speak to pick their marketing budget the same way they pick a new van — a number that feels right, plus a bit of VAT, minus whatever the last quote for a kitchen came in over. That works for a van. It doesn't work for marketing, because marketing is a function of two numbers the builder already knows: average job value and how many jobs they want to win a month.

This post walks through the math properly for a firm doing £30k extensions. The numbers scale either way if your average ticket is £15k or £60k — the structure of the calculation doesn't change.

Before a single pound goes into Google Ads, you need these four on a Post-it note:

1. Average won-job value. Not your biggest ever, not your dream ticket. The average across the last 20 jobs you closed. Call this J.
2. Gross margin on that job. After materials, labour, subcontractors — before your own salary. For a UK extension builder doing quality work this is usually 18–28%. Call this M%.
3. Quote-to-won rate. Of the quotes you put out, how many you win. Typically 25–45% for premium work. Call this Q%.
4. Enquiry-to-quote rate. Of the enquiries you get, how many you actually quote (rather than politely pass on because it's tyre-kicking or out of scope). Typically 40–70%. Call this E%.

For our worked example: J = £30,000, M% = 22%, Q% = 35%, E% = 55%.

This is the number nobody calculates and everyone should.

Gross profit per won job = £30,000 × 22% = £6,600.

Probability an enquiry becomes a won job = E% × Q% = 55% × 35% = 19.25%.

So the average expected gross profit for each raw enquiry through the door is £6,600 × 19.25% = £1,271.

Read that slowly. Every single enquiry — including the ones who ghost you, quote-shop you, or turn out to be a neighbour of a neighbour — is worth an average of £1,271 in gross profit.

That doesn't mean you should pay £1,271 per enquiry. It means the ceiling is higher than most builders think, and the reason for-per-lead marketplaces exist is that £30–£60 per enquiry is a rounding error on that number.

Two rules of thumb we use with clients. They're conservative on purpose.

Rule one — the 10% rule. Spend up to 10% of gross profit per enquiry to acquire that enquiry. On our numbers, that's up to £127 per enquiry. At that cost, you're keeping 90% of expected gross profit per enquiry for the business. Anything under this is a healthy unit.

Rule two — the 5% of turnover rule. Once the pipeline is steady, aim for total marketing spend (agency fees + ad spend + tools + content) to sit around 4–6% of turnover. Under 3% and you're under-investing; over 8% and either the work is broken or you're in a launch phase.

For a £1.5m turnover firm that's £60,000–£90,000 a year, all-in. That covers website, SEO, ads, content, any agency fee — not just "ad spend."

Here's what the math looks like for our worked builder, if he wants to add four extra won jobs per month:

• Target won jobs per month: 4
• Required quotes per month: 4 / 35% = ~12
• Required enquiries per month: 12 / 55% = ~22
• Target cost-per-enquiry (at 10% rule): £127
• Maximum sensible monthly marketing budget to hit that target: 22 × £127 = £2,794

That's the budget ceiling to win four extra £30k jobs a month — which is £120,000 in gross revenue at £26,400 in gross profit.

You're spending under £2,800 to generate £26,400. That's a 9.4× return on marketing. Not an agency promise — just what the math says if the rates above are honest.

When we audit a firm, the gap between what they could afford and what they currently spend is almost always the same: they're spending 30–40% of the sensible number, entirely on per-lead marketplaces, getting enquiries that convert at a third of the rate of their own website's.

In practice that looks like:

• £800–£1,000/month on Checkatrade or MyBuilder
• £0–£200/month on Google (sometimes an ad set up in 2019 and never paused)
• £0 on their own website beyond the hosting fee
• £0 on content
• Total: ~£1,100/month, all paying rent to someone else's platform

And then wondering why the phone goes quiet from October to February every year.

For firms doing £500k–£1.5m, a healthy mix looks like:

• Website, SEO, content (owned assets): £1,400–£2,200/month
• Google Ads (brand + service + town): £500–£1,200/month
• Platforms (Checkatrade capped, Google Business Profile): £200–£400/month
• Tooling (email, booking, CRM): £80–£150/month

Roughly £2,200–£4,000/month all-in. That sounds like a lot until you notice the 9.4× math — one £30k job a quarter pays for the entire year, and the whole point of the system is that it compounds.

Nothing pays back in month one. That's the thing nobody tells builders.

• Month 1–2: Setup, website, GBP, local pages going live. Some ads running. Enquiries start arriving, but at the old rate.
• Month 3: First measurable lift. Roughly 20–40% more enquiries than baseline, with better quality (more £20k+, fewer tyre-kickers).
• Month 4–6: SEO pages start ranking for quieter "best builder in [town]"-type searches. Google Business Profile starts outperforming Checkatrade on direct calls.
• Month 6–9: Content and case studies compound. Cost-per-won-job drops every month because fixed costs stay flat while won jobs grow.
• Month 9+: The system prints. Won jobs plateau at whatever capacity the business can physically deliver.

If a builder can't sit through months 1–3 without a hand on the plug, the math doesn't help them. This is why we insist on a 6-month minimum — not because we want the revenue locked in, but because the numbers don't mean anything below that window.

Write your four numbers on a Post-it. Do the sum. Compare to what you're currently spending. If the gap is wide — as it usually is — the fix isn't to panic-spend the difference on Monday. It's to put the ceiling on Checkatrade, get your own assets in order (website, GBP, local pages, content), and let the ad budget grow after the owned stuff is working.

Last thought. This isn't a finger-in-the-air budget. It's what the arithmetic says. Every quarter, run the sum again with fresh numbers. The inputs move. The logic doesn't.