Emergency vs scheduled work: when one budget should not cover both

A burst pipe and a planned bathroom remodel may share the word “plumber.” They do not share the same buying situation.

One customer wants a response now. The other may compare providers, ask about scope, request a quote, and decide later. If the same campaign, landing page, budget, conversion goal, and response process serve both, the account may optimise toward whichever action is easiest to generate, not necessarily the work that is most useful to the business.

The answer is not always “split everything.” It is to separate emergency and scheduled campaigns when their customer intent, service promise, economics, capacity, or measurement materially differs.

What “emergency” should mean

Do not define emergency work only by words such as “emergency,” “24 hour,” or “same day.” Define it by the operational promise.

An emergency service generally has a response expectation that is materially shorter than the business’s normal scheduled workflow. The business may need to answer immediately, confirm service area and availability, dispatch a crew, and explain what can happen before a full repair or replacement is quoted.

A scheduled service may still be urgent. A landlord may need a repair before a tenant moves in. A business may need commercial HVAC work before opening. A homeowner may need a roof repaired before a storm. The query can be scheduled in the campaign structure but operationally time-sensitive.

Write the actual promise in plain language: answer and dispatch within the hours the crew can reliably cover, and never advertise a response time the crew cannot deliver. That sentence is more useful than a keyword label.

Why the buying situations diverge

These are operating differences, not universal customer behaviour. A word such as “repair” does not settle the classification: a water-heater repair may be urgent or a planned appointment, and the call or form record should resolve the ambiguity. Confirm the split with the business’s own call records, booking data, margins, and service process.

When one budget CAN cover both

Keep a shared or combined structure when all of these conditions are substantially true:

1. The same team answers both types of enquiry.
2. The business can accept both types during the same hours.
3. The service area and landing-page promise are the same.
4. The acceptable cost per booked job is similar.
5. The business is comfortable with the platform shifting delivery between them.
6. The same conversion action represents value for both.
7. There is not enough volume to support useful reporting in separate campaigns.

A small account with very few conversions can become harder to evaluate after every service is split into its own campaign. Google says Smart Bidding uses conversion data to optimise bids at auction time, and some strategies rely on historical conversion volume. A fragmented account may have better labels but weaker evidence. In that situation, begin with distinct ad groups, service-specific landing-page sections, and a clean tracking field for emergency versus scheduled intent. Split into campaigns when the data or operational need justifies dedicated control.

When to separate

Separate emergency and scheduled campaigns when one or more of these conditions is important enough to protect.

The more of these conditions apply, the stronger the case for separation. If only the ad language differs, separate ads or ad groups may be sufficient.

A practical campaign structure

For a plumbing, HVAC, electrical, garage-door, or similar account, work through the decision rather than creating four campaigns because four labels appear in a planning document.

1. Separate the service labels and records first

   Tag emergency and scheduled terms in the search-term log and CRM before changing budgets. Reporting labels give visibility while the lines still share settings.

2. Split into ad groups when intent and pages differ

   When service, ad, and landing-page intent differ but budget and hours are shared, for example HVAC repair and HVAC installation within one campaign.

3. Split into separate campaigns when control must be protected

   When budget, schedule, geography, bidding, or the service promise differs, for example emergency plumbing and planned plumbing replacement. Each campaign must have enough demand, a clear offer, and a reporting question.

4. Add a maintenance or replacement campaign only on evidence

   Separate replacement or installation only when project value, process, and qualification differ materially; separate maintenance only when recurring service and customer economics justify it.

5. Reserve separate accounts for governance

   Separate accounts or systems only when ownership, billing, legal, or operating teams differ materially. This is rare.

Emergency keywords do not need to live in a separate campaign merely because they contain the word “emergency.” A scheduled customer can also use “urgent,” and an emergency customer may search only for “plumber near me.” Use the complete query, ad promise, service type, hours, and business process. Do not split campaigns only to make the account look organised: splitting creates more settings, more data requirements, and more opportunities for duplication.

Budget design: protected, pooled, or tested

Google Ads allows an individual budget for each campaign or a shared budget across multiple campaigns.

Do not start with a percentage split borrowed from another account. Start with capacity and economics. For each service line, estimate the maximum acceptable cost per booked job:

maximum acceptable cost per booked job
= contribution from a booked job × acceptable acquisition share

Then compare that with the observed cost per qualified enquiry, booking rate, completion rate, and contribution. The acceptable acquisition share is a business decision, not a Google default. If emergency demand is valuable but the team cannot answer at night, an additional emergency budget does not solve the problem. Fix coverage first, or restrict the campaign to hours that can be served.

Ad scheduling and real operating hours

Google says ad scheduling can specify the hours and days when ads show and can set bid adjustments for specific times, and that campaigns show all day by default. Google also says that if no one searches during the scheduled period the ad will not show, and that campaigns using ad schedules still pace toward a monthly spend of 30.4 times the average daily budget regardless of how many days the campaign is active in a month.

Do not use a 24/7 schedule as a substitute for 24/7 operations. Do not promise “24/7” if the business only answers a voicemail after hours, and do not promise immediate dispatch if availability is not checked.

Local Services Ads are a different budget model

Local Services Ads use an average weekly budget and a monthly maximum. Google says lead prices can vary by location, job type, lead type, and bidding mode, and that reporting includes channels such as calls, messages, and bookings. That is not the same as treating emergency and scheduled Search clicks as identical. In LSA, use the service types, job categories, service areas, profile information, business hours, lead feedback, and booking records carefully. If the business cannot accept new work, Google says the ad can be paused or the budget adjusted through the Local Services workflow. Do not assume that an LSA lead is a booked job. It still needs to be answered, qualified, scheduled, completed, and reconciled with the business record.

Match the ad and landing page to intent

The emergency path should make the next step obvious: service area, the hours actually covered, the types of emergency work accepted, what happens after the call, any minimum charge or inspection policy that should be stated, and a prominent phone path with a backup contact path. It should not bury the phone number beneath a long project explanation.

The scheduled path answers different questions: which planned services are offered, what property or customer types are accepted, what is included in the inspection, quote, tune-up, or installation, what evidence supports the provider’s quality, what information is needed to prepare an estimate, and how long the normal booking process takes. A scheduled customer often needs more context. That is not necessarily low intent.

A shared page can work when both paths are obvious and the business can fulfil both promises. If the page creates confusion, separate pages are easier to measure.

Track the path to a booked job

A phone call is not the same as a qualified opportunity. A form submission is not the same as a booking. A booking is not the same as completed work. Track the funnel separately for each service type.

Google’s phone-conversion documentation explains that calls from ads, website calls, mobile-number clicks, call assets, and imported call conversions can be measured in different ways, and documents importing phone-call conversions as a way to connect advertising to important calls and other valuable customer actions. A call-duration threshold can filter calls, but it does not prove qualification or booking. Import deeper outcomes when the CRM or call system can reliably identify them.

At minimum, reconcile the campaign and search source, service requested, emergency-versus-scheduled classification, answered or missed status, qualified or unqualified, booked or not booked, completed or cancelled, revenue and contribution where available, and the reason for loss. If emergency calls are optimised to “calls” while scheduled installations are optimised to “forms,” compare them on downstream booked-job and contribution outcomes. Otherwise the campaign with more cheap front-end actions can appear to win.

The operating check

Use a short review cycle for operational problems and a longer period for economic conclusions. The daily checks confirm operations; the weekly checks review the mix; the verdict-period checks decide the economics.

A single week can identify a broken call path; it usually cannot prove the long-term economics of a seasonal service line.

Economics by service line

Assume an HVAC business separates repair and replacement reporting for one month.

Emergency repair:       $1,200.00 ÷ 9 = $133.33 per booked job
Scheduled replacement:  $1,500.00 ÷ 4 = $375.00 per booked job

Emergency repair:       $6,400.00 × 30%  = $1,920.00
Scheduled replacement:  $15,000.00 × 30% = $4,500.00

These values do not prove one service line is better. The replacement jobs may take more sales time, have delayed completion, and carry different cancellation or installation costs. The repair line may carry more dispatch pressure and after-hours labour. Include emergency-specific response cost or scheduled-project cost where the owner’s model requires it. Do not compare an emergency call and a scheduled project using revenue alone.

What can go wrong after splitting

Campaign separation creates its own risks:

• Each campaign receives too little conversion data for useful bidding.
• The same query can be eligible for multiple campaigns.
• Negative keywords or location settings block valuable demand.
• Emergency and scheduled definitions are inconsistent in the CRM.
• Staff classify every call as “emergency” because the word sounds valuable.
• A service is profitable only after completion, but the account optimises to a first call.
• One campaign gets the budget while the business’s actual capacity is elsewhere.
• A separate campaign creates clean reporting but no operational improvement.

Use exclusions, priorities, landing-page routing, and naming conventions to reduce overlap. Document which campaign owns each service and what happens when a customer asks for more than one service. If budget, bidding, schedule, landing page, conversion goal, and geography all change at once, the next result will be difficult to explain: change one major variable at a time where practical.

I count conservatively. Every untracked job is your upside, not my credit.

The Payback Window

Before separating anything, name the boundary of the first test. Set the acceptable cost per booked job from the contribution a booked job carries, run the proving period long enough that the emergency and scheduled lines each accumulate a defensible booking record, and write the verdict date before spend begins. That cap, that proving period, and that verdict make up the Payback Window: the first test has a boundary, and the boundary is agreed before the money moves.

What to do next

Separate budgets deserve separate verdicts. The proving period guide shows how to set a spend cap and a review date for each side of the business, so neither side subsidizes the other in the report.