The short version
- One ad group, one job. Group keywords by the exact service and the exact intent, not by how you organize your business internally.
- Budget follows data, not guesswork. Start with a number you can watch daily for two to three weeks before you touch it.
- Negative keywords are half the job. A search campaign without a negative keyword list is just paying Google to educate you.
- Quality Score is a discount, not a badge. Relevance between keyword, ad, and landing page is what lowers your cost per click.
- Someone has to check it weekly. Automated bidding manages the auction. It doesn't manage your business goals.
Set the campaign up so it can actually be measured
Pick "Search" as the campaign type, not "Search with Display opt-in." That checkbox quietly lets Google spend your budget on the Display Network, which mixes in placements that have nothing to do with someone typing a search query. If you're a plumber or an electrician, you want people who searched, not people who saw a banner while reading a recipe.
Set your goal to leads or calls, not "website traffic." Traffic is a vanity setting. It tells Google's bidding algorithm to optimize for clicks, and clicks without a conversion action attached are the fastest way to burn a budget with nothing to show for it.
Choosing keywords that convert, not just keywords that get clicks
Skip the broad, high-volume terms you'd guess off the top of your head. A term like "plumber" pulls in DIY searchers, job seekers, and people three states away. Go for the phrases someone types when they have a problem right now: "emergency water heater repair," "same day AC repair [your city]." Lower volume, much higher intent.
Use phrase match as your default, not broad match. Broad match hands Google permission to match your ad to loosely related searches, and for a local service business that permission gets expensive fast. Reserve exact match for your highest-value, already-proven terms once you have a few weeks of search term data to confirm they convert.
Build your negative keyword list from day one, not after you've burned through a month of spend. Add job-related terms ("hiring," "salary," "how to become a"), DIY terms ("how to fix," "diy"), and any service you don't offer. Then keep adding to it weekly from your search terms report. This list should grow for the life of the campaign.
Structuring ad groups so your ads stay tight
One ad group per service, not per keyword and not per whole business. A "drain cleaning" ad group with drain cleaning keywords and drain cleaning ad copy will always beat one generic "plumbing services" ad group trying to cover ten services with the same three headlines. Tight groups mean your headlines can name the exact service the searcher typed, which is what actually moves click-through rate.
Writing ad copy and using extensions correctly
Your ad copy needs to answer the question the searcher already asked, in the first headline. If someone searched "24 hour emergency plumber," your first headline should contain "24 Hour Emergency Plumber," not a slogan. Save the slogan for headline three.
Every ad needs one clear action: call now, book online, request a quote. Pick the one your business is actually set up to handle fast. If your phones aren't staffed at night, don't ask night-time searchers to call.
Add sitelink extensions that point to specific service pages, not just your homepage. Add call extensions with a real number your team answers. Add structured snippets listing the actual services you offer. These aren't decoration, they take up real estate on the results page and push competitors further down, and Google rewards ads with more relevant extensions with a lower cost per click.
Location targeting and device bidding for local businesses
Set a radius around your actual service area, and exclude anywhere you don't send trucks. Every dollar spent showing your ad outside your service radius is a dollar you can't get back. If you serve three specific cities and not the whole metro, target those cities by name rather than a blanket radius that bleeds into areas you don't cover.
Check your device breakdown after two to three weeks of data. Most home service searches happen on mobile, someone standing in front of a leaking pipe reaches for their phone, not their laptop. If mobile is converting better, adjust bids toward mobile rather than leaving the split even by default.
What to actually check every week
Conversion tracking has to be set up before you look at anything else. Decide what counts: a phone call over a certain duration, a form fill, a booked appointment. Without that in place, every other number you're looking at is a guess dressed up as data.
Pull the search terms report weekly. This is where you find both new negative keywords and new keyword ideas you didn't think to target. It's the single highest-leverage five minutes you can spend on an active campaign.
Test one variable at a time in your ad copy: a headline, a call to action, a landing page. Changing five things at once tells you nothing about which change actually worked. Let a test run long enough to collect a real sample before you call a winner.
Automated bidding strategies (Target CPA, Target ROAS, Maximize Conversions) are worth using once you have enough conversion volume to feed the algorithm signal. They're not a replacement for someone reviewing performance against your actual business goals. Set the target, then check in on it, don't set it and walk away.
Frequently asked questions
What's the fastest way to waste budget on a new search campaign?
Skipping negative keywords and leaving the Display Network opt-in checked. Both let Google spend your money on traffic that was never going to convert.
How do I write ad copy that converts?
Match the headline to the exact phrase the searcher typed, state the service plainly, and give one clear next step. Test one change at a time against your current best-performing ad.
Are ad extensions worth the setup time?
Yes. Sitelinks, call extensions, and structured snippets take up more space on the results page and typically lower your cost per click because Google treats a more complete ad as more relevant.
How often should I actually touch a live campaign?
Weekly at minimum: search terms report, negative keyword additions, and a look at conversion rate by device and location. If you don't have the time to do that yourself, that's exactly the kind of ongoing management we handle for clients through our Google Ads management service. Get in touch if you want a second set of eyes on what's running right now.