Deferred-Work & Found-Work Upsell Tracker: Turn What Techs Spot on PM Visits Into Quoted Revenue

The problem this kills

Your technicians are walking gold mines. On every preventive-maintenance visit they spot the worn belt, the corroded contactor, the panel that won't pass code, the compressor on its last legs. Then the visit ends, the finding lives in a scribbled note or a line buried in a work-order comment, and it dies there. Nobody quotes it. Nobody follows up. The customer's unit fails three months later, they're annoyed, and the repair revenue you already found goes to whoever picks up the emergency call.

Found work is the highest-margin revenue in field service and the easiest to lose. The problem isn't that techs don't find it — they find tons of it. It's that there's no clean path from "tech spotted something" to "customer has a quote in their inbox" to "someone is chasing the yes." You don't need a developer to build that path. You need a tool that captures the finding at the moment it happens, drafts the quote, puts a manager in the loop before it goes out, and never lets a follow-up fall through the cracks.

What you'll build

A simple internal web tool for your service team. A tech logs a finding — tied to the asset and the work order, with a photo and a severity (especially safety-critical ones) — right from their phone. The tool drafts a quote for each finding by pulling parts and prices from your price book (a CSV or Google Sheet). A manager reviews every found-work item and approves the quote before anything reaches the customer — the human gate. On approval, the tool emails the quote to the customer via Resend and starts tracking it through quoted → won / lost / deferred. Deferred items get linked back to the asset so they resurface on the next visit. The whole upsell pipeline exports to CSV any time.

What's inside the Implementation Plan

The downloadable plan is a step-by-step file you paste into an AI coding agent. It opens by interviewing you about your business — how your techs report findings today, what your work orders and assets look like and how they're identified, the exact shape of your price book and how you mark up parts and labor, your severity and approval rules, who's allowed to approve a quote, your typical and peak volumes, and the messy edge cases (a finding with no matching price-book part, the same worn part flagged on two visits, a customer who keeps deferring). It reads a short spec back to you for a thumbs-up, then builds the tool around your workflow and price book instead of a generic template. From there it walks the agent through the data model, the mobile-friendly finding-capture flow with photos, the quote drafting engine, the manager-approval gate, the customer email, the follow-up pipeline board, and the CSV export. Every step ends with a ready-to-copy prompt.

The governance it includes (this is the point)

This isn't a toy. The plan builds in the controls a real service operation needs: login so only your team can use it, row-level security so people only see their own company's findings and quotes, a complete audit trail of every finding, quote, approval, send, and status change (who, what, when), a hard human-approval gate so no quote reaches a customer until a manager signs off — the AI drafts, a person approves, and only then does it send — and duplicate guards so the same worn part on the same asset can't be logged and quoted twice. Safety-critical findings are flagged and surfaced first. The tool exists to make a careful human decision fast: the agent prices the work, a manager decides what goes out.

Who it's for

Service managers and account managers who know found work is leaking and want it captured and chased, and technicians who are tired of spotting problems that go nowhere. If you can describe how your techs report what they find and how you price a repair, you can build this.

You've got this — open the plan, paste the first prompt, and you'll be turning your techs' findings into quoted revenue by the end of the afternoon.