Bid the 215 loop first, then decide about Boulder City on its own budget
Geography in a paid account should follow where a truck can finish a call and still make the next one, which is almost never the circle a radius tool draws around your shop.
A radius buys open desert and towns you would decline anyway. Target by city and ZIP instead. Build the core around the valley inside and around the 215, then treat Henderson and North Las Vegas as their own line items, because they are separate cities with their own competitors and their own drive times.
Outlying work deserves a separate decision rather than an accidental one. Boulder City, and anything further out, either gets its own small campaign with its own budget and its own trip charge in the ad copy, or it gets excluded. What it should not do is silently absorb spend inside a metro campaign.
Set location targeting to people present in your area rather than the looser option that includes people merely showing interest in it. A city that hosts constant visitors and constant relocation research will otherwise send you clicks from people nowhere near a house you can service.
Report by geography from day one. Cost per booked job in Summerlin and cost per booked job in North Las Vegas will diverge, and you cannot act on that difference if the account only reports one blended number.
Run the graveyard shift as its own campaign, not as a bid adjustment
A round the clock hospitality economy puts real demand into hours when most competitors' ads are still running and their phones are not.
Overnight and early morning clicks are worth buying only if a human answers. If calls roll to voicemail at 1am, turn the ads off then and stop paying for the privilege of losing the call. Honesty about your own coverage saves more money than any bidding tactic.
If you do cover those hours, split them into a separate campaign rather than burying them under a bid adjustment. A separate campaign shows you cost per booked job by daypart, gives the overnight ads their own copy, and lets you fund it without touching the daytime budget.
Write the copy to match. Ads that say when a person actually picks up, and what happens if you call at four, get fewer clicks and better ones. Call only ads earn their keep in those hours because the visitor has no interest in reading a page.
Watch the pattern for a month before drawing conclusions. Overnight volume in this valley is real but uneven, and one busy week during a heat run is not a trend.
Swamp cooler parts and window units are where a valley budget leaks
The negative keyword list is the part of an HVAC account that actually decides your cost per booked job, and in the desert it has entries no national template includes.
Start with the local ones. Evaporative cooler questions, pads, pumps and parts. Portable and window units. Do it yourself repairs, wiring diagrams, capacitor part numbers, refrigerant prices. Every one of those clicks costs the same as a customer and none of them invoice.
Then the ones that look like customers and are not. Hiring and training searches, apprenticeships and pay questions, home warranty claims, builder warranty callers, tenants who cannot authorize a repair, and visitors describing a hot room in a hotel on the Strip who need a front desk rather than a contractor.
Read the search terms report weekly for the first month, then monthly. An HVAC account never finishes generating garbage, because Google keeps widening matches on your behalf. Twenty minutes a month is the highest paid twenty minutes in the account.
Keep the list in one shared negative set applied across campaigns, so a new campaign inherits everything you already learned instead of repeating a year of tuition.
The clicks you buy in April pay for the changeouts you cannot install in July
Spending in even monthly twelfths guarantees you overpay in exactly the weeks your board is already full.
During a hard heat run, demand arrives on its own. The marginal click you buy at that point often becomes a booking you cancel, a customer who waits three days, and a review that follows you for years. Turning spend down when capacity is gone is a legitimate move, not a failure of nerve.
The shoulder months are the opposite. Nobody is desperate, which is precisely when a planned replacement conversation is possible, when your install crews have room, and when competition for the click is softer. Fund the replacement campaigns there.
Pace against the dispatch board rather than the calendar. If installs are booked out two weeks and service is triaging, that is data the account should respond to within days, not at the end of the month.
Keep a small always on core for emergency service so the account never goes cold, and let the replacement side flex. Accounts that get switched off entirely lose their footing and take weeks to steady.
If your crews are set up for rooftop package units, say so in the ad
Ad copy that names the equipment you actually work on filters clicks better than any audience setting Google will sell you.
Split ad groups by equipment and job type, not by clever keyword themes. Service and repair, system replacement, ductless and mini split work, light commercial. The buyer changes with the category and so does the page they should land on.
Say what you are equipped for. If your trucks and crews are set up for rooftop equipment, put that in the copy. If you do not touch a particular class of system, or you subcontract it, keep it out of the account entirely rather than paying for calls you hand away.
Light commercial deserves separate treatment. A restaurant manager on Spring Mountain and a homeowner in Centennial Hills are running different searches, judging different things and needing different proof, and one campaign cannot serve both without wasting money on each.
Match the landing page to the ad group. Sending a mini split search to a general home page is the most common and most expensive mistake in HVAC accounts, and it is free to fix.
A changeout decided in March starts with a click in July, and last click will not show it
Price the account on booked jobs and released installs measured over a window long enough to include the summer somebody spent thinking about it.
Form fills flatter everyone. Import outcomes from your dispatch or field software back into the ad platform so the bidding optimizes toward jobs that actually got run, not toward whoever fills in a form fastest.
Use call tracking with dynamic numbers on the site, and record the calls if your counsel confirms your notice practices are in order. Half of what you need to know about a paid campaign is audible in the first thirty seconds of the call it produced.
Set conversion windows that respect how people buy a system. A replacement is a considered purchase, often with a spouse, sometimes with financing, and a thirty day window quietly credits none of it. Look at assisted paths before you cut a campaign that looks weak on last click.
End every report on one line: what did a booked job cost, by campaign and by area. If that number is not in the report, the report is decoration.
Questions we actually get
- What should we budget to start?
- There is no honest figure anyone can quote you sight unseen, because it depends on your service area, your capacity and what your competitors are already paying. The useful way to set it is backward from the field. Decide how many additional jobs a week your crews can absorb, work out what a job is worth to you after cost of delivery, and set spend so a plausible cost per booked job still clears that. Then check the assumption against real data after the first several weeks.
- Should we run Local Services Ads as well as regular search ads?
- Often yes, and they behave differently. Local Services placements sit above search ads, are billed toward leads rather than clicks, and generally require license and insurance verification before they run. Requirements and availability change, so it is worth confirming the current criteria with the platform and making sure your Nevada licensing and insurance paperwork is in order first. Treat it as a separate channel with its own reporting rather than folding it into the search numbers.
- Is it worth bidding on competitor names?
- Sometimes, and it should be a small, closely watched experiment rather than a habit. Competitor traffic converts worse and costs more, and it invites the same treatment in return. It is more defensible to bid on your own brand, which is cheap, protects the searches you already earned and stops other contractors from collecting people who were looking for you.
- Do we pause the account in winter?
- Turning it off entirely is usually a mistake. Heating work exists here, the valley does get cold nights, and the shoulder season is when replacement conversations are actually possible because nobody is in crisis. Shrink the emergency service campaigns, keep a small always on presence, and shift the money toward replacement. An account switched off and back on loses momentum you then pay to rebuild.
- Our cost per lead looks fine but the schedule is not full. What is happening?
- Almost always the leak is between the lead and the booking, not in the auction. Check how many calls are answered live, how long they take, what the after hours path actually does, and how many booked jobs the ads produced compared with how many forms. Until outcomes from your dispatch system feed back into the account, the bidding is optimizing toward whatever is easiest to submit, which is rarely the customer you want.
