How much should a local service business spend on Google Ads?

The honest answer: it depends on your unit math

Every agency gets this question, and most answer with a number that conveniently matches their minimum retainer. The honest answer is that a Google Ads budget is not a number you pick — it is a number you derive. It falls out of three things you already know about your business: what an average job is worth, how many calls it takes to book one, and how many booked jobs you can actually service this month.

Once you frame it that way, "how much should I spend" becomes "how many jobs do I want, and what am I willing to pay for each one." That question has a real answer.

Work backward from the job, not forward from a budget

Take a simple illustration. Say your average job is worth $400 and you close one out of every three qualified calls. If a click costs $15 and one in ten clicks turns into a call, a call costs you about $150 — which means a booked job costs about $450. That campaign loses money before the truck leaves the lot, and no amount of "optimization" fixes arithmetic.

Now change the two levers that actually matter: send those clicks to a landing page that converts one in five visitors instead of one in ten, and tighten keywords so you pay for buyers instead of browsers. The same math lands closer to $180 per booked job — less than half your average ticket. That is a campaign worth scaling.

This is why we measure every account we run on cost per acquisition, not clicks or impressions. The budget question answers itself once the CPA is known and stable.

The five-line budget math you can copy

Here is the same logic as a procedure. Grab last quarter's numbers and run it for your own business — it takes ten minutes and settles the budget argument permanently:

• Line 1 — average ticket: what a booked job is actually worth. Use the median if a few whales skew your average. Example: $400.
• Line 2 — affordable cost per job: for one-off work, roughly a quarter to a third of the ticket keeps the job profitable after labor and materials — call it $130 on our example. Recurring trades can spend up to first-job breakeven, because the plan value pays you back.
• Line 3 — calls per job: one divided by your close rate. Closing one in three means every booked job costs you three qualified calls.
• Line 4 — affordable cost per call: line 2 divided by line 3 — about $43 here. If clicks in your market make a call cost more than this, the fix is the landing page and keywords, not a bigger budget.
• Line 5 — the budget: jobs you want this month times line 2. Want 25 more jobs? 25 × $130 ≈ $3,300 a month — and now the number has a reason attached to it. Or skip the napkin: the interactive calculator runs all five lines on your numbers.

Realistic starting ranges

With the caveat that your market and trade change everything, here is what we typically see work as a starting structure for a single-market service business. These are ad-spend figures only — what an agency or freelancer charges to run the account is a separate line on top:

• Testing phase ($2,000–$5,000/mo): enough spend to gather real conversion data in 30–60 days without betting the quarter. Below roughly $2,000 in a competitive metro, data trickles in too slowly to learn anything.
• Proving phase ($5,000–$10,000/mo): CPA is known and under target; budget expands into more service lines, neighborhoods, and hours.
• Scaling phase ($10,000+/mo): the constraint stops being the ad account and starts being your capacity to answer calls and staff crews.

Starting budgets by trade, from our own accounts

Generic ranges only get you so far, because the same $30 click is a bargain in one trade and a budget-killer in another. These are the observed ranges from accounts we manage, as of August 2026 — every market, season, and service area moves them, so treat the spread as the message. Each trade links to its full economics breakdown, and the tables live on in our quarterly benchmarks report.

Why the spread is that wide

Click prices track case value, not difficulty. A personal-injury firm can pay $250 a click and win, because a signed case costing $1,500–$6,000 to acquire is measured against a five- or six-figure settlement. A pest-control company paying $12 for an urgent click wins a different way — a $250 one-off treatment converts into a $1,200–$2,400 recurring plan a third of the time, so the lifetime value carries the math. Run line 2 of the budget procedure against what a customer is worth to you over a year, not just the first invoice, and the "expensive" trades stop looking expensive.

Where budgets actually die

In our experience across $30M+ in managed spend, budgets rarely die because Google Ads "doesn't work." They die for three preventable reasons: the account optimizes for traffic instead of conversions, nobody wired up conversion and call tracking so decisions are made blind, and leads sit unanswered. Industry research is blunt on that last one — Lead Connect data shows around 78% of customers buy from the company that responds first, and a well-known Harvard Business Review study found contacting a lead within five minutes makes a qualified conversation dramatically more likely than waiting even an hour.

Fix response speed before you raise the budget. It is the cheapest ROAS improvement available.

When Google Ads is the wrong answer

An honest budget article has to include the cases where the right budget is zero — for now:

• The ticket can't carry the click: sub-$150 one-off work with no recurring plan behind it rarely survives auction prices in a competitive metro. Build the review base and the Business Profile first — the free channels have to carry cheap-ticket work.
• You're already at capacity: if the crews are booked three weeks out, more leads buy you angrier customers, not more revenue. Raise prices or hire before you advertise.
• There's no page and no tracking: sending paid clicks to a homepage with an untracked contact form means the budget produces anecdotes, not data. Wire conversion and call tracking and a real landing page first — it is the difference between a test and a donation.
• LSA volume hasn't capped yet: in most trades, Local Services Ads book cheaper trustworthy calls than search until their market volume runs out. Max that floor first; scale past it with search.

When to raise spend

Seasonal businesses have a different question — what to hold through the quiet months — but for everyone else: raise the budget when your cost per booked job has held under target for at least four consecutive weeks and your close rate has not slipped — that combination means the extra dollars are buying profit, not noise. Raise it in steps of 20–30%, not doublings, so the algorithm re-learns without a performance cliff.

If you want a second set of eyes on your numbers before you commit a budget, that is exactly what our free audit is for: we will show you where the spend is leaking and what your realistic cost per job looks like.

Frequently asked questions