Nothing in the Shop Got Replaced. It All Got Connected.
The Coan Engineering audit mapped five workflows worth $18,200 to $37,700 per year in manual labor. When we walked through the recommendations, the first question was: "What would we have to replace?"
Nothing. Every automation we scoped reads from what already existed and adds processing that wasn't there before. No CRM migration. No new platform the team has to learn. No gap where the old process breaks before the new one works.
The Spec Form Was Already the Input
The 40-field spec form existed before we arrived. The customer fills it out, hits submit, and the shop was manually routing it to a technician who read it and hand-built a quote recommendation.
The AI triage build doesn't change the form. The webhook fires on submit, same as before. The AI reads the fields and the dyno file, drafts a recommendation, and a tech reviews it in two minutes instead of analyzing from scratch in 25. The form didn't change. The tech still approves every quote. What changed is the 20 to 35 minutes per submission that was coming off skilled-tech time.
The Inbox Still Exists
The routing fix doesn't rebuild the shared inbox. Three form types were landing there: spec forms, repair intakes, general contact. The fix reads the form type from the submission data and assigns it to the right person before the email lands. The inbox still exists for visibility. It stops being the place where someone has to figure out who should have this.
Three nodes in an automation flow. The inbox doesn't go away. The manual triage step does.
The Quote Trigger Was Already There
The quote follow-up doesn't change how quotes are sent. The tech sends a quote, same as always. The automation puts a timer on that event and a stop condition on any reply. If the customer doesn't respond in the window, a follow-up goes out. If they reply, the sequence stops. The tech doesn't change their process. The quotes that were dying in silence stop dying.
Status Updates Read From an Existing Record
The repair status system doesn't create a new tracking layer from nothing. The shop already had a way to know where a converter was in the rebuild process. The automation reads from that state. When status changes, a text goes to the customer. One additional field update, which staff was probably making anyway. The difference is the customer finds out without calling.
The RMA Form Replaces a PDF, Not a Process
The digital return form is the one build that changes something visible to the customer: the downloadable PDF becomes a web form. The underlying workflow is the same. Customer submits return info, staff processes the return. The difference is a tracking number issues automatically, the customer knows the form was received, and staff sees it in a dashboard instead of an inbox thread.
What This Means Before Any Build Starts
Every operation we audit has existing stuff. The intake forms are usually already built, even if the processing behind them is manual. The data is already somewhere: in a form submission, in a CRM record, in a status field someone updates by hand.
Automation that reads from what exists and removes the manual steps in the middle is faster to build, faster to prove, and doesn't require the team to change how they work before it's shown to work.
The question before any engagement is: where does the data live? If the answer is "in a form submission," that's the starting point. Everything else can be wired from there.