A radius drawn from your shop spends Virginia money on District addresses
Geographic targeting is the first decision in a DMV HVAC account and the one most often made by dragging a circle on a map.
The District, suburban Maryland and Northern Virginia are separate legal jurisdictions that license and permit on their own terms. A twenty mile circle around a shop in Falls Church covers all three. If your standing does not, you are paying for clicks that cannot become work.
Build the targeting from the jurisdictions you actually serve, then trim by drive time. Beltway congestion means two addresses the same distance away can be a forty minute difference at four in the afternoon.
Exclude aggressively at the boundary. It feels wrong to turn off spend, and it is the single change that most reliably improves cost per booked job in accounts like this.
One more targeting detail that gets missed: set location settings to people in your targeted area rather than people interested in it. A transient population searching from elsewhere about a future move is a different buyer and should not be paid for at emergency rates.
Run the District, suburban Maryland and Northern Virginia as three budgets
One campaign covering the metro hides three different auctions, three different competitor sets and three different job mixes inside one average.
Split at the campaign level so each jurisdiction has its own budget, its own bids and its own reporting line. The moment you do, the differences become visible.
Typically the job mix diverges. Suburban Maryland and Northern Virginia produce more straightforward changeouts on accessible equipment. District addresses produce more complicated work on tight lots, which is often better margin and always a longer sale.
Landing pages should follow the split. An ad for Bethesda that lands on a page naming Montgomery County, and an ad for a Capitol Hill address that lands on a page about rowhouse equipment placement, both beat a single generic page. Message match is not a nicety in a market this fragmented.
Budget can then move deliberately. If District work is worth more to you and takes longer to close, fund it as a separate line rather than letting a cheaper suburb quietly absorb it.
Bid on heating failure the week before the first hard freeze, not after it
Demand here swings with the weather in a way that a flat monthly budget cannot follow.
Heating season, freeze risk and snow response are ordinary parts of the calendar. When the first sustained cold hits, no-heat searches spike and so do competitor bids. Being ready a week early is worth more than reacting a day late.
Structure for the swing. Keep emergency heating campaigns separate from planned replacement campaigns so you can raise one without dragging the other along.
Pace budget against dispatch capacity rather than evenly across the month. Buying calls you cannot answer for two days during a cold snap converts a marketing cost into a reputation cost.
Watch the shoulder seasons too. Spring and fall carry planned replacement and maintenance intent at lower competition, which is often where the best cost per sold system in the account hides.
The exclusion list is where an HVAC account is actually built
Most wasted spend in this trade is not bad bidding, it is buying searches that were never going to become a job.
Filters, thermostats, window units, parts lookups, DIY diagnostics, manuals and error codes all pull clicks from people who will never book. So do employment searches, which spike whenever a large employer in the region is hiring.
Renter language is worth its own attention. A transient rental population produces searches from people who cannot authorize a repair. Terms tied to landlords, building maintenance and lease issues generally belong on the exclusion list unless you have a property management program built to receive them.
Exclusions are not a one time task. Review search terms weekly at first, then monthly. In a metro with this much varied vocabulary, new waste appears constantly.
Add jurisdiction awareness to the exclusions as well. If a place name repeatedly produces inquiries you have to decline, the money spent on it is not a targeting inconvenience, it is a direct loss.
Local Services Ads reward the paperwork you already keep in a filing cabinet
The verification steps that make this format tedious are the same steps that keep casual competitors out of it.
Local Services Ads require credential and background verification before you can run. For an established contractor with current documentation, that is an afternoon of admin. For a low effort competitor it is a barrier.
In a three jurisdiction market, keep your documentation organized by jurisdiction from the start. Requirements differ and are worth confirming with the relevant licensing authority rather than assuming one set covers everything.
Treat the format as a separate channel with separate economics, not as an extension of search. Lead quality and dispute handling behave differently and should be reported on their own line.
It is also not a substitute for a search campaign. The two reach overlapping demand at different moments, and killing one to fund the other usually shows up as a quiet drop in booked work.
The number that matters is what a booked Bethesda changeout actually cost you
Cost per lead is the easiest metric to improve and the least connected to whether the account is working.
A campaign can halve cost per lead by buying cheaper, vaguer clicks and produce fewer jobs. Any account judged on that number will eventually drift toward it.
Push the measurement downstream. Cost per booked job, then cost per sold replacement, reported by jurisdiction. That requires connecting the ad platform to whatever your team books in, and tagging every call to a source. It is tedious once and useful forever.
Call tracking matters more here than in most trades because the majority of high intent HVAC inquiries arrive by phone. Untracked phone calls make the whole report fiction.
Expect the picture to change once the connection exists. Campaigns that looked expensive per lead frequently turn out to be the ones producing sold systems, and the cheap campaign turns out to be answering parts questions.
Questions we actually get
- What should an HVAC company in the DC area expect to spend on Google Ads?
- There is no honest single figure, and any number quoted without seeing your service area, job mix and dispatch capacity is invented. Budget should be set against how many jobs you can actually run in a week and what a sold system is worth to you. We would rather start smaller in one jurisdiction and expand on evidence.
- Should we run one campaign for the whole DMV?
- Generally no. Splitting the District, suburban Maryland and Northern Virginia into separate campaigns lets you fund them differently, land ads on pages that match the jurisdiction, and see which one actually produces sold work. A single campaign averages all of that into one misleading number.
- Do Local Services Ads replace regular search ads?
- In most cases they work alongside them rather than instead. The verification requirements are worth completing because they keep casual competitors out, but the two formats reach demand at different moments. Report them separately so you can see what each is really producing.
- How quickly can paid search deliver calls during a cold snap?
- Ads can be live quickly, which is the main advantage of the channel, but nobody should promise a call volume. The more useful preparation is structural: have the heating campaigns built, the exclusions in place and the budget headroom agreed before the first hard freeze, so raising spend is a decision rather than a scramble.
- Why do you want call tracking before launching anything?
- Because most high intent HVAC inquiries in this market arrive by phone, and an account measured only on form fills is measuring the minority. Without call tracking tied to booked jobs, there is no way to tell a campaign that produces sold systems from one that produces parts questions.
