Calloway Mechanical Services, Inc.

Nine of eleven recurring decision types require Dana's sign-off, with an average 6.4 hour wait. This is the root constraint. It caps throughput, delays crews, and makes time off functionally impossible.

Identical scopes vary 19% depending on who prepares the estimate. Margin is being decided by habit rather than standard, and no one can see it happening until the job closes.

There is no role with standing authority to act when the owner is unavailable. The business has depth in the field and none between the field and the owner.

Nine of eleven recurring decisions still route through the owner.

Pricing exceptions, scheduling conflicts, warranty calls, and any purchase over $500 all stop at Dana's desk. Two supervisors described their approval path as "text Dana and wait." The cost is not the decision itself, it is the queue in front of it. Average wait time from request to answer measured 6.4 working hours, which pushes crew dispatch into the next morning roughly twice a week. Until decision authority is written down and delegated by dollar threshold and by category, every other improvement in this report will bottleneck at the same desk.

Core workflows exist in people's heads, not in writing.

Three of the four revenue-producing workflows have no documented standard. New technicians are trained by shadowing, which produces a different version of the job with every trainer. The estimating process in particular varies by a measured 19% on identical scopes depending on who prepared it. This is the single largest source of margin leakage found in the assessment and the fastest thing to fix, because the knowledge already exists. It simply has never been captured.

Strong crew, no second layer of leadership.

Field capability is genuinely good and retention is above the trade average at 3.8 years. What is missing is the layer between the owner and the crews. There is no one whose job description includes making the call when Dana is unreachable, so the organization behaves as if it has thirty-one employees and one decision maker. A single operations lead with real authority would move this domain, and Decision Rights with it, faster than any hiring push.

Accurate books, but 30 days behind the decisions they inform.

The accounting is clean and the CPA relationship is solid. The problem is latency and granularity. Job-level profitability is reconstructed after the fact rather than tracked during the job, so unprofitable work is identified in the month after it is completed rather than in the week it goes sideways. Two of the last six large jobs were confirmed as loss-making only at month close. Weekly job costing and a simple cash-forward view would raise this domain quickly.

Referral strength is real, but the pipeline depends on the owner's relationships.

Roughly 68% of new work traces back to relationships Dana personally holds. That is a strength today and a liability in any scenario where Dana steps back. Follow-up on quoted work is inconsistent, with 22 open quotes over 30 days old and no owner assigned to any of them. The engine runs, it just runs on one person's calendar and one person's memory.

Good tools, poorly connected.

The field service platform, the accounting system, and the scheduling sheet each work well in isolation and do not speak to each other. Staff re-enter the same job data an estimated three times per job, which the assessment priced at 41 administrative hours per month. This is an integration problem, not a software purchase problem. Replacing systems here would cost more and deliver less than connecting the ones already in place.