Group campaigns by how the job gets sold, not by the pipe involved
A dispatch now call and a job that needs a site visit run on different clocks, and mixing them makes both sets of numbers meaningless.
Emergency work closes on the phone in minutes. Sewer replacement, repipe, well equipment and remodel work close after someone stands in the yard, sometimes weeks later. Put them in one campaign and the fast, cheap conversions eat the budget while the profitable ones look like failures.
Build at least two structures. One buys immediate dispatch and gets judged on same week bookings. The other buys consideration traffic and gets judged on bids issued and bids won, with a reporting window long enough to see them close.
Landing pages should match the split. A dispatch page needs a phone number, a service area and an availability statement. A bid page needs photos of comparable work, an explanation of what the site visit covers, and a form that captures enough to route the visit properly.
Budgets stay separate too. The moment a slow week tempts you to move project money into emergency clicks, the project pipeline empties out and you find out about it two months later.
Three tiers of service area, and only the first one gets top of page money
Draw the map by what a truck can reach and still bill a second job, then pay accordingly instead of bidding one flat rate across the metro.
Tier one is the core you service every day without thinking about it. Tier two is the ring you will happily take with a scheduled window: Round Rock, Pflugerville, Cedar Park, Buda. Tier three is the edge you take only when the job is big enough to justify the drive, out toward Georgetown, San Marcos or Dripping Springs.
Bid them differently. Tier three should almost never win a top of page auction for a drain call, and it can bid aggressively for sewer, well or repipe terms where the ticket carries the windshield time.
A radius drawn from a central yard does none of this. It buys the same click at the same price whether the address is fifteen minutes away or an hour up the corridor, and MoPac at four in the afternoon does not care what your radius said.
Review the tiers every quarter against real bookings. Suburbs move. A yard relocation, a second truck or a new tech living in Williamson County can promote a whole tier overnight.
Do not put a rebate promise in a water heater ad when the utility changes by address
Program and incentive eligibility here depends on which side of a service boundary a house sits, so ad copy that promises one is writing a check the address may not cash.
Austin Energy is city owned, while most of Texas runs on a retail market where a customer picks a provider. Statewide advice about shopping for electricity is simply wrong inside the city, and eligibility for utility programs varies by service territory.
For a plumbing account, the effect is practical. Water heater and electrification searches are valuable, and the temptation is to write a headline about rebates. Do not name plans, rates or amounts in an ad or on a landing page. Say that incentives may be available depending on the utility serving the address and that a homeowner should confirm current terms with them directly.
The honest version still converts. A page that explains the difference between a tank swap and a change in fuel type, and what generally has to happen either way, holds attention better than a number that turns out not to apply.
It also protects the call. Nothing burns a tech's morning faster than a homeowner who booked expecting a discount your ad implied.
A full schedule is a reason to lower bids, not to celebrate
Buying calls you cannot answer or serve this week costs you reviews, and the damage outlasts the month the budget was spent in.
Pacing should follow capacity. When the board fills, pull back on the highest cost emergency terms first and let the consideration campaigns keep running, because those jobs get scheduled anyway.
When the board is thin, the lever is not more budget on the same terms. It is wider geography, longer hours, and the project campaigns that fill next month rather than this afternoon.
Somebody has to own the connection between the dispatch board and the account. In small shops that is a five minute conversation on Monday, not a report nobody reads on the fifteenth.
Hold something back for weather and for the weeks in Central Texas when everything breaks at once. An account with no reserve either misses the surge entirely or blows a month of budget in three days.
Send the invoice back to the campaign, or you will keep buying the cheapest job
Lead counts and cost per lead will point an account straight at the least valuable work in your book.
Every inquiry needs an outcome and a dollar figure attached to it, matched back to the campaign, the keyword group and the geography that produced it. Call tracking, a booking field in the CRM and a monthly reconciliation against invoices is usually enough.
Once the invoices flow back, the account starts making obvious decisions. Terms that produce steady drain calls hold their budget. Terms that produce three site visits a month and one large job get more, even though their cost per lead looks terrible.
Report in the language of the business. Booked jobs, revenue by campaign, and the share of inquiries you declined. A dispatcher understands trucks and days. A cost per lead chart tells them nothing.
Be patient with the project campaigns. A bid issued in March that closes in May belongs to March's spend, and an account read on a calendar month alone will kill the campaigns that pay best.
Say the fee and the service area in the ad, and half the wrong calls stop
Ad copy is a filter as much as a pitch, and the money you save on unusable calls beats most bid optimizations.
State the towns you cover and whether a diagnostic or trip fee applies. People who will not pay it self select out before they cost you a click and a phone call, and the ones who call are already past the first objection.
Back that with a negative list that reflects this market: DIY and how to searches, apprenticeship and training queries, supply and parts shopping, salary research, rental and tenant repair questions, and anything about pool equipment or irrigation if you do not touch it.
Watch the search terms report weekly for the first two months, then monthly. Growth here keeps generating new phrasings, especially around new construction and builder warranty issues in the outer suburbs.
One caution on refusals: log them. If you are declining the same job type in Kyle every week, that is either a service line worth adding or a tier worth cutting, and you only see it if somebody wrote it down.
Questions we actually get
- What should we budget to start?
- Work backward from capacity rather than picking a round number. Decide how many additional jobs a week you can actually service, estimate what those jobs are worth to you, and set spend so that a bad month is survivable. Starting narrow, in your tier one geography on your highest value services, tells you more than a broad launch that spends fast and teaches nothing.
- Are Local Services Ads enough on their own?
- They are a strong source of dispatch work and worth running, but they mostly serve fast, proximity driven jobs and give you limited control over what you buy. Sewer replacement, repipe and well work generally need search campaigns with real landing pages behind them. Most plumbing companies here run both and hold them to different measures.
- Our cost per lead looks good but revenue is flat. What is wrong?
- Usually the account is optimizing toward the cheapest possible call. Small drain jobs convert fast and look excellent in a leads report while the profitable work goes unbought. Attach invoice totals to campaigns for ninety days and the picture typically reverses. Also check answer rate, since unanswered calls still count as spend.
- Should we advertise in Georgetown and San Marcos at all?
- Only for jobs that carry the drive. Bidding for emergency drain work an hour out generally produces calls you decline or trips you regret. Bidding for sewer replacement, well systems or repipes in those areas can pay well, especially if you can schedule two jobs in the same corridor on the same day.
- Do we need a separate landing page, or can ads point at our homepage?
- A homepage tries to serve everyone and usually converts worse for paid traffic. Match the page to the promise in the ad: dispatch pages for emergency terms, bid pages for the site visit work. It does not need to be elaborate, but it does need the same service area, the same fee statement and a form built for the job type.
