Should I measure marketing against revenue or against margin?
Against margin, once you can. Return on ad spend measured on revenue ranks channels by ticket size, not by profitability, so it consistently favors whichever channel sends big low-margin jobs. Contribution margin per marketing dollar ranks them by what actually reaches the bottom line. If job costing is weak, apply average margin by job type as an approximation and label it as such.
Why revenue-based ROAS misranks channels
Two channels can produce identical revenue per marketing dollar and contribute very different amounts of profit. Equipment replacement carries a different margin structure than diagnostic and repair work, and heavily subcontracted or heavily materials-loaded jobs carry less than either.
A channel that fills the schedule with high-revenue, thin-margin installs will look like your best performer on a revenue-based report and your worst on a margin-based one. Which one you look at determines where the next budget increment goes.
Getting to a usable margin number without perfect job costing
Most operators do not have reliable per-job costing, and waiting for it is not a plan. Two approximations work well enough to change decisions.
Two workable approximations
- Margin by job type. Compute an average gross margin for each major job type from a period where costing was reasonably captured, then apply it to every job of that type. Crude, but it captures the biggest driver of variance.
- Margin by business unit. If your field service system has business units with their own labor and materials rollups, use the unit-level margin. Coarser than job type, but usually already accurate.
- Label the estimate. Whichever you use, show it as modeled margin, not actual margin, on the report. The number is for ranking channels, not for the tax return.
What changes when you switch denominators
Expect the channel ranking to reorder. Expect at least one campaign that everyone considered a winner to move down. Expect maintenance and service agreement channels to move up, because their revenue understates their value and their repeat behavior is not in a revenue-per-click number at all.
Also expect the conversation to get better. A revenue-based argument between marketing and operations has no resolution, because each side is holding a real number. A margin-based one has a shared denominator. That shared denominator is the reason we join field service costing data into marketing intelligence rather than reporting the two separately.
The line you should not cross
Do not push fixed overhead into the calculation. Contribution margin means revenue minus the costs that vary with doing the job. Allocating rent and salaried office payroll across jobs produces a number that moves when volume moves for reasons unrelated to marketing, and it makes incremental decisions actively wrong.
Keep two views: contribution margin for channel decisions, and full profit and loss for the business. Mixing them in one table is how organizations end up cutting the channel that was funding their overhead. Keeping them separate but reconcilable is a reporting design problem, and it is worth solving in your dashboards once.
Topics: contribution margin · ROAS · job costing · channel comparison
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.