Clark and Floyd County clicks are only worth buying if you can cross the river to serve them
Geography is the first decision in a Louisville HVAC account and the one most likely to be quietly wrong.
A mileage radius from a shop in the Highlands reaches New Albany and Jeffersonville without noticing it crossed a state line. Registration, licensing and tax treatment differ between Kentucky and Indiana, so confirm your standing with the relevant state agency and your own counsel before you spend a dollar over there.
If you are licensed only on the Kentucky side, exclude the Indiana side properly. Not a bid adjustment, an exclusion. Buying clicks you cannot invoice is the most expensive habit in local paid search, and it hides inside a radius setting where nobody looks.
If you do serve both, run them as separate campaigns with separate budgets. Cost per booked job almost always differs across the river, and a merged campaign averages the two into a number that tells you nothing.
Inside Kentucky, keep going. Louisville Metro and Jefferson County are one merged government, while Oldham and Bullitt are separate counties with their own buyer mix. Separate them in the account so the reporting can separate them too.
Local Services Ads are worth pursuing alongside search, and the verification paperwork is state specific. Sort out which entity is verified where before you assume the ads will show on both sides.
Bid on the furnace that quit tonight and the furnace that quits in January in separate campaigns
Emergency intent and replacement intent behave so differently that averaging them ruins both.
A no-heat search at nine at night is a dispatch decision. Short copy, phone number, arrival honesty, high bids for a few hours, and the whole value is in whether somebody answers.
A furnace replacement search is a considered purchase with a longer path. The click is worth more, the close is slower, and the landing page has a different job. Same account, different economics, different measurement window.
Maintenance plan and tune up searches are a third thing again. Cheap clicks, low immediate revenue, real value if your intake actually converts a tune up into an agreement and a future changeout. Track them as their own line or they get judged unfairly.
Keep heating and cooling structurally separate even where the keywords overlap. In a market with real winters and hot summers both, one blended campaign will always overspend on the season that happens to have more search volume and underspend on the one making you money.
Half the clicks in a heating campaign come from people who will never buy equipment
An HVAC account lives or dies on what you refuse to pay for.
Parts lookups, model number searches, DIY troubleshooting, thermostat wiring diagrams and space heater shopping all sit adjacent to your keywords and all cost the same as a real inquiry.
Job seekers are the quiet drain. Apprentice, technician, hiring and trade school searches turn up constantly against HVAC terms and convert on nothing. So do home warranty callers, whose work is assigned rather than chosen, and new construction bids if you do not do that work.
Build the exclusion list from your own search terms report, weekly at first, then monthly. Nobody can hand you a finished list, because the wasteful terms in your account depend on what you sell and where.
Landlord and property manager terms are a judgment call rather than an automatic exclusion. Rental stock is a real part of the market in Germantown and Old Louisville, and portfolio work is worth something if you are set up to bill a manager rather than a homeowner.
Cold snaps do not respect the calendar month, so neither should the pacing
Even monthly budgets are the default setting and the wrong one for a business with real winters.
Demand in this market arrives in bursts. A hard freeze produces more searching in three days than the previous three weeks, and an account pacing evenly runs out of budget on exactly the day it should be spending.
Watch the forecast the way your dispatcher does. Raise budgets ahead of a cold stretch, not after it, because the searching starts when the house stops holding temperature and not when the weather makes the news.
Pace against capacity, not against ambition. If your trucks are booked three days out, more spend buys inquiries you will lose to whoever answers faster. Turning spend down during a genuine backlog protects your reputation and your cost per job at the same time.
Summer has its own peak and the two peaks are separated by shoulder seasons where replacement selling is easiest and competition is lightest. Budget the year in four moves rather than twelve equal ones.
Air freight shift work means your ad schedule should not go dark at ten
A large logistics and air freight employment base makes unusual hours normal here, and paid search should follow the phone.
Plenty of people in this metro work nights and handle household business at hours that most contractors have switched off. If you genuinely take those calls, bid into those hours, because competition thins out there.
If you do not take them, say so and schedule accordingly. Paying for a click at two in the morning that lands on an unanswered phone is worse than not showing at all.
Test the after-hours path yourself before you raise the bid. Call your own number at eleven at night from an unfamiliar phone and see what the caller actually hears. Whatever the answering service says is what your ad copy is really promising.
Relocating workers are a steady presence in this market and they arrive without a contractor. Competitor terms and brand searches for shops they have never used are generally cheaper than head terms, and worth a small, honest campaign.
Feed Google the jobs your dispatcher ran, not the leads your form counted
The bidding system optimizes toward whatever you tell it a conversion is, so tell it the truth.
Form fills and phone calls are inputs, not results. If every ring counts as a conversion, the system will happily find you more rings from people who wanted a part number.
Send booked jobs back into the account. Offline conversion import from your dispatch software is the single highest value change most HVAC accounts can make, and it changes what the bidding chases within a few weeks.
Report on cost per booked job, split by county and by intent. A cheap no-heat call in Jefferson County and an expensive replacement inquiry in Oldham County can be equally good or equally bad, and only the booked number tells you which.
Give any structural change enough time to be judged fairly. A week of weather can swing paid search results in either direction, and no honest report draws a conclusion from a cold Tuesday.
Questions we actually get
- What should we budget for HVAC Google Ads in Louisville?
- There is no honest per month figure to quote you without seeing your service area, your close rate and your capacity. What we can say is that the budget should be built around booked job economics and dispatch capacity, and paced against cold snaps rather than split evenly across the calendar. Any agency that quotes a cost per click for this market before looking at your account is making it up.
- Should we advertise into Southern Indiana?
- Only if you are set up to serve it. Clark and Floyd County work sits under Indiana rules, and requirements differ by state, so confirm your registration and licensing standing before spending there. If you are Kentucky only, exclude the Indiana side outright rather than reducing bids.
- Are Local Services Ads worth running alongside search?
- For most residential HVAC shops, yes, and they generally reward paperwork you already keep. The wrinkle in this metro is that verification is state specific, so sort out which entity is verified where before assuming your ads appear on both sides of the river.
- How do we stop paying for people looking for parts and jobs?
- Read the search terms report every week at the start, and build the exclusion list from your own data rather than a generic list. Parts lookups, DIY wiring questions, apprentice and hiring searches and home warranty traffic are the usual offenders, but the exact mix depends on what you sell.
- What number should the monthly report end on?
- Cost per booked job, split by county and by intent, alongside how much capacity you actually had. Lead counts flatter everyone. If your dispatch software can send booked work back into the ad account, the bidding starts chasing the right thing instead of the loudest thing.
