Work out what a lead is worth to you, then budget for enough leads to win the jobs you want. In our example roofing business, a lead is worth about $360, a lead costs about $60, and 4 jobs a month takes about $1,620 a month in ads. To start, about $43–$86 a day for the first fortnight is enough to learn which ads bring work.
Every roofer is different. The figures in this guide show the method; swap in your own job value, margin and close rate to get your budget. Example amounts are shown in your currency.
Step 1: what a lead is worth to you
Three numbers decide it: what a job is worth, the share of that you keep after costs, and how many of your leads become paying work. Multiply them and you get the most you can pay for a lead and still break even.
| Example roofing business | |
|---|---|
| Average job value | $8,000 |
| Margin you keep after costs | 30% |
| Leads that become paying jobs | 15% |
| A lead is worth | $360 |
Pay less than that per lead and your ads make money; pay more and they lose it, however busy the phone is. There’s more on this in what a lead is worth.
Step 2: what leads cost
In the US, the median Facebook cost per lead for home and home improvement businesses is about US$41, against about US$28 across all industries1. Leads are getting dearer too: the average cost per lead rose about 21% in a year2.
Your own cost depends on your area, your offer and your ads. This example uses $60 a lead. Once your ads have run for a few weeks, use your real figure from Ads Manager.
Step 3: budget for the jobs you want
Start from the work you want, not from a number that feels safe. Divide the jobs you want each month by your close rate to get the leads you need, then multiply by what a lead costs.
| Each month | |
|---|---|
| Paying jobs wanted | 4 |
| Leads needed (4 ÷ 15%) | 27 |
| Cost per lead (example) | $60 |
| Monthly ad budget | $1,620 |
| Kept from those jobs before ads | $9,600 |
| Left after the ads are paid for | $7,980 |
If the last line comes out small or negative with your own numbers, more ad spend won’t fix it. Lift the value of each job, your close rate or your margin first, or aim the ads at the jobs worth more.
Step 4: how much to start with
A brand-new campaign needs enough results to tell a good ad from a bad one. Meta has long described about 50 results in 7 days as the point an ad set leaves its learning phase, though some accounts have been shown lower bars3. Most small businesses can’t afford 50 leads a week, so aim for 10–20 leads across the first two weeks instead.
At $60 a lead, that’s about $43–$86 a day. Run two or three ads, judge them on the jobs they bring, then move budget to the winners a step at a time.
The jobs worth advertising
Roofing is a high-value, low-frequency trade. A re-roof or restoration can be worth thousands, so a roofer can afford to pay much more for a lead than a plumber or cleaner and still make money. The catch is that fewer leads book, and quotes take longer, so you need patience before judging an ad.
- Re-roofs and restorations: the biggest jobs, often after an inspection turns up problems.
- Storm and leak repairs: urgent, seasonal and usually found on mobile.
- Gutter replacement and guards: smaller jobs that often lead to bigger ones.
- Roof inspections: a low-cost first step that can feed the restoration pipeline.
Ad ideas for roofers
- Drone footage of a before-and-after restoration on a local street.
- A free or low-cost inspection offer that leads with the warning signs people can see from the ground.
- Storm response ads switched on quickly after bad weather in your area, and off when it clears.
When to spend more, and less
Storm season brings bursts of urgent repair work; restorations are often planned for drier months. Have storm ads ready to switch on, and keep restoration ads steady through the year.
Mistakes that waste budget
Judging ads after a week
Big roofing jobs can take weeks from enquiry to signed quote. Judge ads on a 30-day view and a running close rate, not on last week’s bookings.
Treating every enquiry as equal
An inspection request and a “how much for a whole new roof?” message are very different leads. Track which ads bring the enquiries that become signed restorations.
Changing everything every day
Big edits to budgets, audiences or ads send an ad set back into learning, and results wobble while it relearns. Make fewer, bigger changes and give them a few days.
Spreading a small budget over too many ads
Ten ads on $60 a day means none of them gets enough leads to judge. Fewer ads, each with a fair share of the budget, tell you more, faster.
Agency or do it yourself?
Agencies commonly charge $500–5,000 a month, or 10–20% of ad spend, on top of the ads themselves45. On a $1,620 budget, that fee can be a big share of what you spend reaching customers. A good agency can earn it; if you mainly need someone watching the numbers and telling you what to change, you may not need one. See the signs your agency isn’t worth it.
Questions
Can a roofer afford to pay more for leads than other trades?
Usually, yes. With bigger jobs, a lead is worth more even when fewer leads book. The worked example below shows a roofing lead worth several times a typical home services lead price.
Do drone videos work in roofing ads?
They show the job in a way photos can’t, and they tend to hold attention. Judge them like any other ad: on the signed jobs they bring, not the views.
How long before I know if my roofing ads work?
Give each ad enough budget for about 10–20 leads, then judge it on a 30-day view. Big-ticket trades need longer than a week.
See what your own ads are worth
Tony checks every live ad each morning and tells you which to turn off, which to scale and why, using your numbers. The first audit is free and takes four minutes.
Sources. Dollar figures from studies are shown in your currency, converted at Reserve Bank of Australia rates for 9 October 2026 and rounded; each source has the original. Example amounts are the same number in your currency.
- WordStream (LocaliQ), Facebook Ads Benchmarks 2025: medians of 726 US lead campaigns, April 2024 – June 2025. www.wordstream.com/blog/facebook-ads-benchmarks-2025
- Search Engine Land, “Facebook ad costs jump 21% in 2025, but still beat Google”, September 2025 (WordStream LocaliQ data). searchengineland.com/facebook-ad-costs-jump-beat-google-461690
- Jon Loomer, “Is the learning phase changing?”, May 2024: Meta’s documentation reflects about 50 optimised events in 7 days to exit learning; some accounts were shown 10 in 3 days. www.jonloomer.com/qvt/is-the-learning-phase-changing/
- Margin Media, “How much does it cost to advertise on Facebook in Australia?”, October 2024: management usually A$500–5,000 a month, or 10–20% of ad spend. marginmedia.com.au/our-blog/facebook-ad-costs-australia
- Linear Design, Facebook ads management pricing guide, 2026: US$500–5,000+ a month or 10–25% of spend; many require 3–6 month minimums, some 6–12. lineardesign.com/blog/facebook-ads-management-pricing-guide/