Is my average ticket up because of pricing or because of job mix?
Separate them arithmetically. Take last period's job mix, re-price it with this period's per-job-type averages, and the resulting change is the price effect. Whatever remains is the mix effect. Without that split, a swing toward replacements looks identical to a price increase, and the two call for opposite responses. Mix is mostly driven by demand, season and dispatch. Price is something you control directly.
The decomposition, in plain steps
You need two things per period: the count of completed jobs by job type, and the average ticket by job type. Then run three totals.
First, actual revenue this period. Second, a hypothetical using this period's job counts with last period's per-type tickets, which isolates how much of the change came from doing more work. Third, a hypothetical using last period's counts with this period's per-type tickets, which isolates price and per-job scope. Compare each hypothetical to last period's actual and you have the mix effect and the price effect side by side.
- Price effect. Per-type average ticket changed. Includes real price increases, better add-on attachment, and scope creep on the same nominal job.
- Mix effect. The proportions between job types changed while each type's ticket held.
- Interaction. A small residual when both moved at once. Report it rather than quietly assigning it to one side.
Why the distinction changes the decision
If the increase is price, the follow-up question is whether close rate held. A price increase that holds close rate is pure margin. A price increase that costs you conversions may be net negative even though average ticket looks better.
If the increase is mix, the follow-up question is where the mix came from. Sometimes it is seasonal and will reverse. Sometimes it came from a deliberate change in what your marketing targets. Sometimes it came from nowhere you intended, which usually means your paid media keyword or audience mix drifted.
The classification problem underneath
This analysis is only as good as your job type taxonomy. If half your work is coded to a catch-all bucket, the mix effect will hide inside it. Most companies have a workable taxonomy in the field service system that nobody has audited in years.
Before running the decomposition, check what share of revenue sits in generic categories. If it is large, fixing the categorization is a higher-value project than the analysis. Job type coding is also what makes source comparison meaningful, since it lets you compare like work to like work.
How often to run it
Monthly is enough. Average ticket is a slow metric with meaningful month-to-month noise, and running the split weekly invites reacting to a single large job. Run it monthly, look at the twelve-month trend of each effect separately, and treat any single month's residual as noise until it repeats.
This is the kind of standing calculation worth automating rather than rebuilding in a spreadsheet each month, which is what purpose-built dashboards are for.
Topics: average ticket · price · job mix · decomposition · analysis
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.