Why does a single blended cost per job target break down across job types?
Because job types differ in gross profit, close rate and demand behavior. An emergency call and a scheduled maintenance visit can differ several times over in what they can afford to pay for acquisition. One blended target quietly overspends on low-margin work and underspends on the high-margin work you actually want, while the blended average sits comfortably in range and hides both errors.
The mechanism
Marketing spend has to be recovered from gross profit. Two jobs with similar revenue can leave very different profit once technician hours, materials and subcontracted work come out. A blended target implicitly assumes every job contributes the same, and every job does not.
The result is systematic misallocation in both directions. Campaigns producing cheap low-margin work look efficient against the blended target and get more budget. Campaigns producing expensive high-margin work look inefficient and get cut, even when each of those jobs pays for itself several times over.
Why the blend hides it so well
Mix shift moves the blended number without any campaign changing. A season heavy in one job type moves cost per job and average ticket simultaneously, and both movements look like marketing performance.
This is the same measurement problem that shows up throughout revenue reporting: a change in composition wearing the costume of a change in performance. The only reliable defense is to report within segments and let the mix be a separate, explicitly stated line.
Building targets per job type
- Group by economics, not by trade jargon. Three to six buckets is usually enough — emergency repair, planned repair, maintenance and agreements, replacement and install.
- Compute gross profit per bucket from actual job costing. If costing is unreliable, fix that before building targets on it.
- Set a ceiling per bucket. Share of gross profit you will pay to acquire, adjusted for repeat value you can observe.
- Publish the ceilings. They are only useful if the person managing spend can see them next to actuals.
Aiming spend at a job type is harder than setting the target
Campaign structure has to correspond to job type — separate campaigns or ad groups by service intent, landing pages that match, and tracking that carries the intent through to the booking record. Otherwise you have targets you cannot act on.
Even then the correspondence is imperfect: a maintenance search can turn into a replacement sale, and an emergency call can turn into a membership. That is fine, as long as the reporting credits the job that actually happened rather than the campaign's intended category. Getting that credit right is standard field service integration work, and it is what makes per-type targets operable rather than theoretical.
Topics: job types · mix · targets · budget
Have a version of this question about your own business?
The useful answer usually depends on which systems you run and how they're connected. That's a conversation, not a blog post.