Structure the account around which truck answers, not around your service menu
Campaign structure should match dispatch reality, because dispatch reality decides whether a click can become revenue.
Most plumbing accounts are built from the service list on the website. Drain cleaning, water heaters, sewer, repipe, fixtures. The list is fine for a menu and wrong for a budget, because those jobs are answered by different people with different equipment and different availability.
Sort by the truck instead. Work that needs someone tonight sits in one set of campaigns. Scheduled work that can wait a week sits in another. Sewer work needing a camera, a jetter and possibly an excavator sits in a third, because only so many of those run in a day.
The split gives you controls that mean something. Pause the emergency campaigns at nine at night when the on-call tech is already committed, and leave the scheduled campaigns running, because a water heater quote can wait until morning.
It also stops the cheapest clicks from eating everything. Drain cleaning searches are plentiful and inexpensive next to sewer replacement searches. Left in one campaign, they will spend the budget before the expensive keyword gets an impression.
Every thaw sends a second wave, and that is where the budget usually runs dry
Chicago does not get one freeze event, it gets a season of them, and the thaw between cycles produces emergencies of its own.
Pipes split during a freeze and often stay quiet until things warm and water starts moving again. Ground shifts. Sump pumps that idled all month are suddenly asked to work. Demand arrives as a series of spikes across the cold months, not as one week you can plan around.
Budget in seasons and hold a reserve. A flat monthly number will either sit underspent through a calm January or run out three days into a cold snap, and either way work goes to a competitor with headroom.
Set the mechanics up before you need them. Caps that can be raised in an afternoon, alerts when impression share drops because you have gone budget limited, and a written rule covering who approves an increase and how quickly.
The summer peak behaves differently. Remodels, sewer replacements and water heater upgrades pace steadily, so the same account should feel calm from June through September and twitchy from December through March.
The North Side drive is short, and the parking is what costs you the hour
Geographic bidding should reflect time on site, not distance on a map.
A twelve mile radius drawn from your yard treats a call in Lincoln Park and a call in Skokie as equivalent. They are not. Dense blocks with permit parking, narrow gangways and rear alley access can eat more of a technician's day than a longer run to a suburban house with a driveway.
Build target areas from your own job history. Which neighborhoods and towns produced work that ran on schedule, and which produced jobs where the crew spent longer looking for parking than working?
Then bid them apart. Cook County and the collar counties can sit in separate campaigns with separate budgets when their economics differ, and inside the city you can adjust by neighborhood cluster rather than treating the whole North Side as one bucket.
Exclusions matter as much as targets. If you do not cross into Will County, do not let a broad radius spend there simply because one suburb sits near the line.
Bid a rodding call like a coupon and an overhead sewer conversion like a project
Different plumbing jobs justify wildly different click costs, and averaging them hides both.
A blocked kitchen line is a small ticket and a short conversation. Sewer replacement, overhead sewer conversion and flood control work involve a site visit, a camera, permits and often a week of thinking. The second group supports a far higher click cost, because one closed job is worth many drain calls.
Keep those keyword sets in separate campaigns so their bids and budgets never compete. Give the expensive terms room to look unprofitable for a while, since they close slowly and a single month will make them look worse than they are.
Watch the phrasing. Someone typing sewer rodding wants a truck today. Someone typing overhead sewer cost is comparing contractors, and belongs on a page that explains the work rather than a form demanding a phone number.
Emergency terms follow their own rule. Frozen pipe and burst pipe searches convert on availability, so the ad and the landing page should both answer whether anyone can come tonight before either mentions the company.
Build the negative list from last winter's call recordings, not from a template
The exclusions that save real money are specific to your shop, and you already have the evidence.
Every downloadable negative keyword list holds the obvious items. Free, DIY, jobs, salary, how to. Adding them takes ten minutes and prevents the cheapest sort of waste.
The expensive waste is local and boring. Renters searching because a landlord will not answer. People looking up a license or a permit record. Homeowners pricing a sump pump at a big box store. Callers who saw a power outage and are unsure whether they need a plumber, an electrician or ComEd.
Pull a month of winter call recordings and search terms and read them side by side. Every call your dispatcher had to end politely is a term to exclude, and the pattern is usually visible within an hour.
Repeat the review quarterly rather than once. Query patterns move with the season, and the terms wasting money in February are not the ones wasting money in July.
The landlord with three two-flats is not one booked job
Cost per booked job is the right measure, and it still understates what some customers are worth.
Report on booked jobs rather than leads, because a form fill that never became work is not a result. Getting there takes call tracking, a way to mark outcomes in the dispatch software, and the discipline to reconcile the two every month.
Then look at who the buyer is. Chicago holds a deep stock of two-flats and three-flats owned by people living in one unit or nearby, plus small managers running a handful of courtyard buildings. One acquired customer can mean several buildings and years of calls.
Cost per booked job on a first visit makes those clicks look expensive. Cost per customer across a year tells a different story, and it should change what you are willing to pay for a search mentioning apartments, units or a building.
Keep the reporting honest in the other direction too. Include jobs booked and later canceled, and calls turned down for being out of area, because both were paid for and both point at something in the targeting.
Questions we actually get
- What should a Chicago plumbing company expect to spend on Google Ads?
- There is no honest number without your average job value, your close rate and the geography you want covered. We will not quote a cost per click or a budget from a template. The starting point is arithmetic on your own figures: what a booked drain call is worth against what a sewer replacement is worth, then a budget that can survive a week when a cold snap doubles the searching.
- Are Local Services Ads worth running alongside search?
- In most cases they are worth testing, since they sit above search results and are priced per lead rather than per click. They also require verification and reviews, and they will not replace a search account for planned work like flood control or a repipe. Run both, tag them separately, and compare them at the booked job.
- How should we handle after-hours calls in the account?
- Decide first whether you truly answer at three in the morning. If a live person picks up, run emergency campaigns overnight and say so in the ad. If calls go to voicemail, either schedule those campaigns off or route to a service that can ask real questions. Buying emergency clicks nobody answers is the most expensive mistake in the account.
- Do we need separate campaigns for the suburbs?
- Usually yes, once the suburbs are a meaningful share of your work. A search in Naperville and a search in Pilsen behave differently, and they carry different drive times, different competitors and often different permitting steps. Separate campaigns let you fund them apart and see which one is paying.
- How quickly can paid search produce calls?
- Ads can serve as soon as they are approved, so calls can arrive the same week. We will not promise a volume or a cost, because both depend on your market, your ads and how fast the phone gets answered. What we will say is that the first month is largely spent removing waste, and the account gets more efficient as the negative list and the geographic data fill in.
