A radius drawn from Maplewood spends money in three counties and two states
Circle targeting ignores every boundary that actually governs whether a click turns into a truck you can bill.
Draw a twenty mile circle from a yard near Maplewood and watch where it lands. Part of the City of St. Louis. A long list of St. Louis County municipalities. A reach toward St. Charles County. Depending on the radius, the far edge starts picking up the Illinois side.
Those are not variations on one market. They are different governments, different permitting practice, and in the case of the Metro East a different state. The clicks cost roughly the same and they are not worth the same.
Replace the circle with a list. Target by municipality and by county, group the ones that behave alike, and set bid adjustments where drive time and administrative variation genuinely change what a job is worth to you.
The immediate benefit is not efficiency, it is visibility. Once geography is listed rather than drawn, your reporting finally shows which places send work worth having, and that answer usually surprises the owner.
The Illinois half of the account needs its own money and its own copy
Metro East demand runs under different state rules and deserves a separate budget rather than a leftover share of a Missouri campaign.
Belleville, Edwardsville, Madison County and St. Clair County searchers are not accidentally in your service area. They live in a distinct part of this metro, and they notice when an ad speaks only to the Missouri side.
Licensing differs by state. Ad copy should reflect that you work there without stating requirements you would have to keep current. Say where you are licensed to work, and let the specifics sit with the reader and the relevant authority.
Give it a separate budget for a practical reason: shared budgets are decided by whichever half wins the auction that morning, not by which half you want to grow. Separate budgets make that a decision rather than an accident.
If you do not cross the river, exclude it outright. Paying for clicks you have to refuse is worse than paying for nothing, because it also costs you the minute a dispatcher spends saying no.
Sewer lateral bids can carry a click price a drain cleaning campaign cannot
Job value should set the bid, which means big ticket searches belong in their own campaigns with their own leash.
A cleared kitchen drain and a replaced sewer lateral are the same trade and completely different businesses. Put them in one campaign and the cheaper term will consume the budget every time, because it converts faster and looks better on any surface metric.
Separate them. Emergency and small repair terms in one structure, with tight geography and aggressive hours. Replacement and repipe terms in another, with a higher tolerable cost per click and a much longer window before you judge them.
The second group closes slowly. A homeowner researching a water service replacement in an older brick home may take weeks and several quotes. Any pacing decision made on a two week lookback will kill that campaign right before it pays.
Water heaters sit between the two, and they deserve their own treatment. Demand spikes with cold weather, the ticket is predictable, and the decision window is short, which makes it the one place where fast bidding actually matches buyer behavior.
Your negative list starts with tuckpointing, because brick sends the wrong searches
In a brick metro with freeze-thaw winters, a large share of the traffic your plumbing keywords attract is not plumbing at all.
Water in a basement is a search that lands on plumbers constantly here, and a meaningful part of it is masonry, grading or foundation work. Tuckpointing, spalling, waterproofing and settlement queries will find your ads if you let broad matching do its thing.
Build the negative list before you build the ad copy. Masonry and foundation terms. DIY and how to phrasing. Parts, fittings and supply house searches. Employment and apprenticeship queries. Utility account and bill lookups that have nothing to do with hiring anyone.
Add the local ones you only learn by reading search terms weekly for a month. Landlord and tenant phrasing, home warranty language, and municipal lookups all show up in this metro and all waste money quietly.
Negative keyword work is not a setup task that gets ticked off. In a plumbing account it is most of the ongoing work, and skipping a month is visible in the numbers the following month.
Freeze-thaw does not run on a monthly budget calendar
Winter demand here arrives in bursts, so pacing has to be a weekly decision with reserve held back for the week nobody scheduled.
Real winters are ordinary here. So are frozen lines, split hose bibs and the run of calls that follows a hard cold stretch. Demand does not distribute itself evenly across a month to suit your budget settings.
A flat monthly cap handles that badly in both directions. During the cold week you cap out by midday and hand the overflow to whoever is still bidding. During the mild weeks you spend the same money on softer demand at the same price.
Hold reserve deliberately. Set a working daily budget below what you can afford, keep the difference available, and release it when the forecast turns. The decision has to belong to someone who is allowed to make it in a morning, not to a monthly review.
There is a dispatch constraint too. Buying more calls than you have trucks for during a freeze does not produce revenue, it produces bad reviews. Pacing up should follow crew availability, not just the auction.
Judge the account on the jobs the dispatch board actually filled
Cost per lead flatters a plumbing account, and cost per booked job is the only figure that survives contact with the schedule.
Leads are cheap to generate and easy to misread. Wrong numbers, price shoppers, renters who cannot authorize work, and callers outside your area all count as leads and none of them fill a slot on the board.
Track calls and forms through to an outcome, then divide spend by booked jobs. Do it separately for the City of St. Louis, for the county municipalities you serve, for St. Charles County and for the Metro East. Blending them produces an average that describes no part of your business.
Expect the split to be uneven, sometimes badly so. One area may produce cheap calls that rarely book while another produces expensive calls that almost always do. Both look identical on a cost per lead report.
Once the outcome data exists, most optimization becomes obvious. Move money toward the geography and the service lines that fill trucks, and stop defending the ones that only fill a spreadsheet.
Questions we actually get
- What should a St. Louis plumbing company expect to spend on Google Ads?
- There is no honest benchmark we can hand you, because click costs in this trade move with weather, competitor behavior and which municipalities you target. What we can do is size a budget against your capacity: how many jobs per week you can absorb, what a booked job is worth to you, and how much of the account has to be held in reserve for cold snaps.
- Do Local Services Ads replace search ads for plumbers?
- They do a different job. Local Services placements tend to suit fast emergency demand, while search campaigns are where you compete for replacement and repipe work with copy and landing pages that can actually make an argument. Most accounts we would run include both with separate budgets and separate expectations.
- Should city and county campaigns really be separate?
- Yes, for the same reason your dispatcher already treats them separately. The City of St. Louis and St. Louis County are different governments, and permitting practice varies again between county municipalities. Splitting them costs you a little management time and tells you which half of the metro is actually paying you.
- How quickly can you tell whether a campaign is working?
- Emergency campaigns show their character quickly because the decision window is hours. Replacement and repipe campaigns take much longer, since a homeowner may collect quotes over several weeks. Judging those on a short lookback is the most common way a working campaign gets shut off early.
- What happens to my ads during a hard freeze week?
- Ideally three things at once: released reserve budget, adjusted copy that reflects a realistic response time, and a conversation with dispatch about how much demand you can actually take. The failure mode is spending harder while the board is already full, which converts ad money directly into complaints.
