A representative build: a single-rooftop service and parts operation moved from a monthly departmental summary to a daily view of absorption, effective labor rate, and technician and advisor productivity — with the constraint on throughput identified rather than assumed.
Loaded nightly, so a bad week is visible in the week it happens
Hours per RO and ELR down to the individual, against a written definition
Capacity utilization by hour and day, against the brand-tier range
Assessment to full deployment on a single rooftop
A store with a healthy-looking service department on paper: absorption in an acceptable range, gross trending up modestly, no obvious problem. The service director's actual complaint was that the shop felt busy all the time and the numbers never moved much, and nobody could say where the ceiling was.
The reporting available was a monthly departmental summary out of the DMS. It could say what happened. It could not say whether the constraint was bays, technicians, parts availability, or dispatch — and every proposed fix, including hiring two more technicians, rested on a guess about which.
A dashboard over the DMS, the scheduling tool, and the parts system, loaded nightly and reconciled to the financial statement. The design goal was narrow: make the constraint visible, and make the person-level numbers defensible enough to act on.
Fixed gross against the fixed expense it is meant to absorb, monthly and rolling, with the labor and parts contributions separated. Tied to the statement so the number survives contact with the controller.
Customer pay, warranty, and internal shown separately against door rate. The internal transfer price turned out to be doing more damage to the blended rate than the discounting everyone was worried about.
Hours sold against available and worked, by technician and job type, with the recurring jobs where flagged and actual time had separated flagged for a look at the labor guide.
Hours per repair order, ELR, and declined-service capture per advisor — published to each advisor first, and to the group only after the definitions had been agreed.
Utilization by hour and day against appointment volume and parts availability. This is the report that answered the original question.
Fill rate by source, days supply, obsolescence aging, and the emergency-purchase pattern — which is where the throughput constraint turned out to be hiding.
Submission lag, rejection reasons, and resubmission outcomes, with warranty receivable aging treated as the working-capital line it is.
Defection timing by selling store and maintenance interval, because next year's fixed-ops volume is decided by this year's retention and nothing in the monthly summary showed it.
Not bays, and not technicians. Utilization by hour showed capacity sitting idle mid-morning and again mid-afternoon, in a pattern that lined up with parts fill rate rather than with appointment volume. Technicians were waiting on parts, absorbing the wait as unproductive time, and the monthly average had been smoothing it into invisibility.
The fix was a stocking and pre-pull change, not a hire. We are describing this because it is the ordinary outcome of measuring properly rather than an unusual one: the expensive answer is frequently the wrong one, and the only way to know is to instrument the question.
The reporting did not fix anything. A service director changed a stocking policy and a dispatch practice. What the dashboard did was make the argument for that change specific enough to win, and then show whether it worked. Reporting that is not attached to a decision and a review is decoration.
| Phase | Work |
|---|---|
| Weeks 1–2 | Assessment and a definitions workshop with the service director, parts manager, and controller. Every metric written down before anything was built. |
| Weeks 2–5 | Extracts from DMS, scheduling, and parts. Nightly loads, reconciliation to the statement, and the departmental view live. |
| Weeks 5–8 | Technician and advisor level reporting, throughput and capacity analysis, parts performance. |
| Weeks 8–10 | Warranty and retention modules, alerting, and the weekly review cadence the reporting feeds. |
About this page. This describes a real build and the decisions behind it. Your data sources, your constraint, and your timeline will differ — we are describing an approach, not quoting a project.
Yes. Each location loads and reconciles separately, then rolls up, which is the same pattern we use for multi-rooftop groups. The mapping work is the part that grows, not the reporting.
That is a fair concern and the sequence matters more than the software. Definitions get published first, each technician sees their own numbers before anyone else does, and the scorecard is paired with the pay plan it relates to. Where a store is not ready for that, we ship the departmental layer and leave person-level reporting for later — it is a real choice, not a delay.
Reconciliation to the statement happens before the first release, so the departmental numbers are defensible from day one. Person-level numbers typically take another few weeks of running in parallel with existing reports, because that is how long it takes for people to stop finding discrepancies — and the discrepancies they find in that period are usually worth finding.
No. We read from it. Scheduling tools are generally fine at scheduling; what they do not do is join appointment and capacity data to DMS labor and parts data, which is where the throughput picture comes from.
If the honest answer is that nobody knows, that is the report worth building first. Let's talk.
Schedule a Consultation
K Auto Tech