A slab leak click and a water heater click do not belong in one budget
Emergency intent and planned replacement intent behave differently at every stage, so they should not share a campaign, a bid or a landing page.
An emergency search converts in minutes or it never converts. A repipe search converts over days, sometimes weeks, after an estimate, a follow-up call and a conversation at a kitchen table.
Separate them into distinct campaigns with distinct budgets. Emergency work justifies aggressive bidding during the hours you can actually dispatch. Planned work justifies a longer attribution window and a landing page built to explain rather than to hurry.
Mixed together, the fast half eats the budget and the slow half looks like a failure. Both conclusions are wrong, and both are expensive.
Water heaters usually sit between the two. Give them their own ad group and their own page, because urgency varies with whether the tank is leaking or simply old.
An hour of freeway sits between your Fort Worth ad and your Dallas truck
Two anchor cities inside one enormous metro means a wide radius around a single pin will quietly spend in both.
Radius targeting is the default setting and it is almost always wrong here. A twenty-five mile ring drawn from Addison reaches neighborhoods your dispatcher would decline on the phone.
Build geography from where crews actually go. Target by city or ZIP code in tiers, bid the core up, bid the edges down, and drop the areas that only look good on a map.
If you want Tarrant County, run it as its own campaign with its own budget and its own landing page, then judge it on its own numbers for a full quarter. Folding it into a Dallas County campaign guarantees you never learn whether it works.
Set location targeting to presence rather than presence or interest. Otherwise you pay for people in other cities researching Dallas plumbers.
LBJ at five o'clock belongs in your ad schedule, not just your route
Drive time across the Metroplex changes what a job costs you, and your bidding should know that before your dispatcher has to explain it.
A four in the afternoon call from the far side of an interchange can consume the rest of a technician's day. The click cost the same as any other. The job did not.
Use ad scheduling deliberately. Bid up the hours you can serve well and bid down the hours where the drive eats the margin, particularly on the outer edges of your map.
Late evening and overnight deserve a decision rather than a default setting. If you genuinely run overnight and answer the phone, bid it. If you do not, stop paying for those clicks entirely.
Weekends often price and convert differently from weekdays. Pull them apart in reporting before you assume they behave the same.
Foundation repair, home warranty and DIY are three exclusions that pay
In a plumbing account, most of the improvement in the first ninety days comes from what you stop buying.
Foundation repair queries will find you through shared vocabulary. Slab, crack, pier, movement, leveling. Unless you do that work, exclude the whole cluster and keep excluding it as new phrasings appear.
Home warranty searches are a specific trap in this market. Somebody hunting for a warranty-dispatched plumber is not choosing you and generally cannot, no matter how good the ad is.
Add the standard DIY and price-research terms, then the ones owners forget: jobs, hiring, apprenticeship, supply house, parts, school. Review the search terms report weekly for the first two months and monthly after that.
Negative keyword lists are not a one-time task. Query patterns shift with the weather, and a freeze or a heat wave brings vocabulary your list has never seen.
The freeze week that spends a quarter's budget in seventy-two hours
Demand here is not flat, and a fixed daily budget handles a hard freeze badly in both directions.
During a freeze event, search volume surges and auction prices climb with it. A capped daily budget spends out by mid-morning, so you miss the afternoon calls entirely while the phone would have been answered.
Decide in advance which way you want it. Either raise budgets deliberately for the event, accepting a higher cost per click, or accept the cap and put the overflow through the phone and a callback list.
The mistake is leaving settings untouched and then reading the month as a failure. It was three days, and the average buried them.
Apply the same discipline coming out. Demand falls quickly afterward and bids should follow within days, not at the next monthly review.
Stop reporting cost per lead to a dispatcher who thinks in trucks
A lead is not a job, and an account optimized to form fills will happily buy you the wrong ones.
Get the booking outcome back into the ad platform. Import offline conversions from your field service software so the bidding learns which clicks turn into work orders rather than which ones turn into rings.
Cost per booked job is the only number an owner can act on. It survives a slow week, a bad month and a change of agency, because it maps to the schedule instead of to a dashboard.
Segment it by campaign and by area. Two inquiries at the same cost can close at different rates and different tickets depending on which county they came from.
Record and score calls as well. Without that, you are optimizing toward a ringing phone and hoping the rest takes care of itself.
Questions we actually get
- Should a Dallas plumbing company run Local Services Ads alongside search ads?
- They do different jobs. The lead-focused format tends to catch urgent, proximity-driven demand, while search campaigns can be built around planned work like repipes and sewer replacement. Many shops run both, but track them separately or the reporting becomes meaningless.
- How much budget does a plumbing account in DFW need to be worth running?
- There is no honest published figure, and any agency quoting one is guessing. The practical test is whether your budget can hold a full day of coverage in your core area during your dispatch hours. If it spends out by noon, tighten the geography before adding money.
- Why exclude home warranty searches when they are clearly plumbing traffic?
- Because the caller is generally routed by their warranty company rather than choosing a contractor. The click costs the same as a real one and the conversation ends the same way most of the time. Exclude it and revisit only if you take that work directly.
- Should Dallas and Fort Worth run in one campaign or two?
- Two, in most cases. They are separate markets with an hour of freeway between them, and combining them hides which one pays back. Separate campaigns also let you set different budgets, schedules and landing pages.
- What should I do with the ad account during a hard freeze?
- Decide in advance. Either lift budgets knowing costs will rise, or hold the cap and route overflow to a callback list you actually work afterward. What does not work is leaving the settings alone and reviewing the damage a month later.
