Google Ads can be the fastest lead source a service business has, or the fastest way to burn a budget. Here's the math that decides which, and the five mistakes that waste most of the spend.
Picture two businesses. A plumbing company spends $4,000 a month on Google Ads for a year and never once checks what a lead costs. A two-person law practice turns $1,500 a month into its most reliable source of new clients. Same platform. The difference isn't skill with the software. It's whether anyone did the math before turning it on.
This is that math, plus the mistakes that quietly eat most small business ad budgets.
The Numbers to Know Before You Spend a Dollar
WordStream's 2025 benchmarks, drawn from more than 16,000 US campaigns, put the averages for Google search ads at a 6.66% click-through rate, $5.26 per click, a 7.52% conversion rate, and $70.11 per lead. Service categories run higher: home and home improvement averages $90.92 per lead, business services $103.54, and attorneys $131.63.
Those are averages across the whole country. Smaller markets run cheaper than mid-sized cities, which run cheaper than major metros like Atlanta or Chicago. But use them as a sanity check. If you're paying $200 a lead for gutter cleaning, something's wrong. If you're paying $150 a lead for estate planning, you may be doing fine.
Now the only equation that matters:
A lead is worth (average customer value × close rate). If that number is comfortably above your cost per lead, ads make sense. If it isn't, no amount of optimizing will save you.
Work it through. A residential HVAC company closes one in four leads into a $6,000 install with decent margin. A lead is worth roughly $1,500 in revenue. At $90 a lead, ads are a bargain. A dog groomer closing one in three leads into a $60 appointment gets $20 of value per lead. At $40 a lead, ads lose money on the first visit and only pay off if the customer comes back for years.
Both businesses can run ads. Only one should run them for cold leads.
Free toolGoogle Ads calculatorCustomer value, close rate and budget in. A yes, a test, or a no, with the break-even math.
Run your numbersWhen Google Ads Make Sense
Search ads work best when three things are true:
- The customer has urgent or specific intent. A burst pipe, a DUI charge, a broken transmission, a tax deadline. They're searching because they need it now, and they'll call the first credible option.
- The transaction is large enough to absorb the lead cost. Or the relationship is long enough that the first job is only the beginning.
- You can actually answer. Ads produce calls and forms. If nobody picks up between noon and two, or the form goes unanswered until tomorrow, you're paying for leads to go to your competitor.
If your business is more considered (people research for weeks before choosing an architect or a financial planner), search ads still work, but the job of the ad is to start a relationship, not close a sale. Your landing page needs to reflect that.
The Five Mistakes That Waste Most Budgets
1. Broad match with no negatives
Google's default settings will happily show your "kitchen remodel" ad to someone searching "kitchen remodel ideas Pinterest" or "how to remodel a kitchen yourself." Every click costs you the same. Tighten match types, and build a negative keyword list from your search terms report every single week. This one habit routinely cuts wasted spend by a third.
2. Sending clicks to the homepage
Someone searches "emergency water heater replacement," clicks your ad, and lands on a homepage about your company's values. They leave. Every ad group needs a landing page that matches the search: the specific service, the specific promise, one clear way to get in touch, and proof you've done it before.
3. Not tracking calls
For most service businesses, the best leads call rather than fill out a form. If you're not tracking calls from ads (call extensions, a tracking number, or call reporting), you're optimizing on half the data and probably cutting the campaigns that actually work.
4. Ignoring Local Services Ads
If you're eligible (home services, legal, financial, real estate, and a growing list of others), Google's Local Services Ads sit above regular search ads and charge per lead rather than per click. The Google Guaranteed badge does a lot of the trust work for you. For most local service businesses this is the first paid thing to turn on, not the last.
5. Setting it and forgetting it
An account that isn't reviewed weekly drifts. Costs climb, search terms wander, a competitor changes the auction. Thirty minutes a week reviewing search terms, pausing what doesn't convert and shifting budget to what does is the difference between an asset and a leak.
"Google Ads isn't a faucet you turn on. It's an auction you show up to every day, and the businesses that win it are the ones paying attention."
A Sane Starting Setup
If you're beginning from scratch, here's the structure I'd build for a local service business:
- Local Services Ads first, if you qualify. Set a weekly budget you can live with, get every review you can onto the profile, and answer the phone.
- One search campaign with two or three tightly themed ad groups for your highest-value services. Phrase and exact match only. A starter negative list (free, DIY, jobs, salary, cheap, near me if you're not truly local to the searcher).
- A landing page per ad group, not per campaign.
- Conversion tracking on calls and form submissions before the first ad runs. Not after.
- A budget you'll hold for 90 days. Ads need data to learn from. Turning them on for two weeks and off again tells you nothing.
How to Know Whether to Keep Going
After 90 days, answer four questions:
- What did a lead cost?
- What did a customer cost (cost per lead ÷ close rate)?
- What is that customer worth in the first year?
- Would you make that trade every month?
If the answer to the fourth question is yes, scale carefully. If it's no, fix the close rate and the landing page before spending more, because those are usually the leak.
Google Ads is a tactic. Whether it belongs in your marketing is a strategy question, and it's answered with a calculator, not a hunch.