Every contractor asks this question and every agency dodges it. The dodge is technically correct — it depends — but it is also useless, so here is a way to arrive at your own number in about ten minutes.
Work backwards from a booked job
You need four numbers, and you already have three of them.
| Number | Where to find it | Typical range |
|---|---|---|
| Average job value | Your invoices | $4,000 – $18,000 |
| Close rate on quoted jobs | Your CRM or your memory | 20% – 40% |
| Leads needed per quote | Count last month | 1.5 – 3 |
| Cost per lead | The one you are guessing at | — |
Multiply the first three and you know what a lead is worth to you. That is your ceiling. Anything below it is profitable, anything above it is tuition.
An example that is not yours
A roofer with an $8,000 average job, a 30% close rate and two leads per quote earns roughly $1,200 in gross profit per lead at a 30% margin. If a lead costs $180, the maths works comfortably. If it costs $900, it still works — just barely, and only if nothing goes wrong.
Most contractors we talk to have never done this multiplication. They evaluate ad spend by whether the invoice felt large, which is a feeling, not a number.
Why the minimum matters more than the maximum
Google Ads needs volume to learn. Below roughly thirty clicks a day, the algorithm is guessing, and you are paying for its education without getting the benefit.
- Under $1,500/month in a market like the GTA, you are usually too thin to compete on the terms that convert
- Between $1,500 and $4,000, you can own a narrow set of high-intent searches
- Above $4,000, you can start expanding into adjacent services
Spreading $1,200 across roofing, siding, eavestrough and skylights is the most common way to waste a budget. Pick one, win it, then expand.
A budget that is too small is not a cheap experiment. It is an expensive way to learn nothing.
Seasonality changes the answer
Roofing in the GTA is not a flat business. Demand in May does not look like demand in January, and neither does cost per click.
The mistake is setting one monthly number and holding it all year. The better approach is an annual budget with a seasonal curve — heavier in spring and early autumn, lighter in the deep winter, with enough left over to stay visible when your competitors go dark.
Going completely dark in the off-season costs more than it saves. You lose the account history, the algorithm restarts cold, and your first month back is spent buying back the ground you gave up.
