Most of a contractor's wasted ad spend goes to two places: searches from people who were never going to hire anyone, and clicks that land on a homepage instead of a page about the thing they searched. Close those two leaks and the same budget produces more booked jobs. Nothing else in a paid account matters as much.
High intent versus research
"Emergency AC repair near me" and "why is my AC blowing warm air" are different people. The first is holding a phone, sweating, ready to call whoever answers. The second is sitting on the couch deciding whether this is a problem. Both cost you money per click. Only one becomes a job this week.
Put your budget on the high-intent side: the service plus a location, the word "repair" or "replacement," "near me," "emergency," "24 hour." Let the research searches be answered by your blog and your service pages, where they cost nothing per visit and pay off whenever that person's problem finally becomes urgent.
Local Services Ads versus Search ads
Local Services Ads sit above everything else on the results page, show a verification badge, and charge per lead rather than per click. Getting in requires submitting licensing and insurance and passing a background check, which takes some paperwork but keeps out the companies that will not bother. For most residential trades this is the first place to spend, because you pay for a contact rather than a click and disputed leads can be challenged.
Search ads are the traditional text ads below that. They give you far more control — exact keywords, your own landing pages, your own copy — and they are the right tool for services Local Services Ads does not cover and for commercial work, where the buyer is searching for something more specific than the consumer categories.
Most contractors should run both: Local Services Ads for the bread-and-butter residential emergency work, Search ads for the services and commercial segments the program does not reach.
Negative keywords are the budget
A new account without negative keywords pays for "HVAC technician jobs," "AC repair training," "how to fix AC yourself," "cheapest AC repair," and the name of the competitor across town. Every one of those clicks costs what a real lead costs.
Build a negative list before launch with the obvious categories: jobs and careers, DIY and how-to, training and certification, free, parts and manuals, and the towns you do not serve. Then go into the search terms report every week for the first couple of months and add whatever you actually got charged for. This single habit separates accounts that work from accounts that do not, and it is most of the weekly work in our paid media management.
One landing page per service
An ad for water heater replacement should land on a page about water heater replacement — the same words the person searched, in the headline, with a phone number and a short form. Sending that click to your homepage forces the visitor to navigate, and a meaningful share of them simply leave.
This is where paid media and the website meet. If your site already has a real page for every service, your ads have somewhere to point. If it does not, building those pages does more for your ad performance than any bidding adjustment.
Track the calls or you are guessing
Ads produce phone calls, and a phone call leaves no record in your ad account unless you set one up. Use call tracking so each call is attributed to the campaign and keyword that produced it, and record them if your state allows — the recordings show you what people are actually asking for and whether the office is booking them.
Without call tracking you will optimize toward form fills, which for a trade business are a minority of the leads, and you will make decisions that quietly harm the channel producing your best work.
Judge everything by cost per booked job
Clicks, impressions and click-through rate tell you the machinery is running. They do not tell you whether it is profitable. The only number that decides the budget is what it costs in ad spend to produce one booked job, and that requires connecting the lead to what the office did with it.
That number is different for every service. A drain cleaning that turns into a repipe is worth a different acquisition cost than a filter change. A rooftop unit replacement justifies spending that would be absurd for a service call. Once you know cost per booked job by service, the budget decisions make themselves: spend more where the math works, stop where it does not, and ignore the metrics in between.
Where to start
Build the negative keyword list before you turn anything on. Get call tracking in place on day one so you are not reconstructing attribution later. Start narrow — your two or three most profitable services, in the towns closest to your shop, with a real landing page for each. Run it for a full month, look at cost per booked job by service, and expand only into the ones that paid.