runbookify
← All plans
Field Service & Dispatch / Work Order Intake

Email-to-Work-Order Parser: Turn Inbox Chaos into Clean Work Orders

Forward or paste a service-request email and let AI pull out the customer, site, asset, problem, and urgency — match it to an existing account, flag what's missing, and hand a coordinator a ready-to-approve work order before anything is created.

IntermediateA weekendBuilds onNext.jsSupabaseResend
What you'll build

A web tool where a coordinator forwards or pastes an inbound service-request email, the app extracts customer, site, asset, problem, and urgency into structured fields, matches it to an existing customer/site, flags the missing or low-confidence pieces, and — only after the coordinator fills the gaps and approves — creates the work order and adds it to a clean CSV export in the columns your system expects.

Gated download

Enter your email — the plan downloads instantly and a copy lands in your inbox.

By submitting your email you'll also receive the weekly runbookify newsletter. You can unsubscribe at any time.

Before you start

  • A Supabase account (free)
  • A Vercel account (free)
  • A Resend account (free)
  • Your customer/site list as a CSV or sheet
  • Your work-order field and priority rules
  • A handful of real request emails to test with
  • Claude Code or any AI coding agent

The problem this kills

Service requests show up as email, and email is messy. A property manager forwards a thread three replies deep. A tenant types two sentences and a phone number. A facilities contact pastes a screenshot caption and a long signature block. Somewhere in all of that is the stuff you actually need on a work order: who the customer is, which site and which piece of equipment, what's broken, and how urgent it is. Today a coordinator reads each one, mentally maps "the unit by the loading dock" to an asset, retypes it into the work-order system, and hopes nothing got dropped.

That retyping is slow, it's error-prone, and it scales badly the moment volume spikes. Names get misspelled, the wrong site gets attached, urgency gets guessed, and the same request from a customer and their property manager becomes two work orders. The fields you need are right there in the email — they just aren't structured, and a person is doing the structuring by hand, one request at a time. You don't need to keep doing that, and you don't need to be a developer to fix it.

What you'll build

A simple internal web tool that turns an inbound email into a draft work order. A coordinator forwards or pastes the request email. The tool strips the reply chain and signature noise, then uses AI to extract the fields you care about — customer, site, asset/equipment, the problem description, and an urgency/priority — into clean, structured boxes. It matches the extracted customer and site against your existing list so you reuse the right account instead of creating a near-duplicate, and it flags every field it couldn't find or wasn't sure about so nothing silently goes blank. The coordinator sees the draft side-by-side with the original email, fills the gaps, fixes anything the AI got wrong, and approves. Only on approval does the tool create the work order and add it to a clean CSV export in the exact columns your field-service or work-order system expects.

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 — what your request emails actually look like and where they come from, the exact fields and naming your work orders use, how your customers and sites and assets are identified, your priority rules, your typical and peak volumes, and the messy edge cases (forwarded threads, signatures, a request that names a site you don't recognize) — and then it tailors the extraction fields, the matching logic, and every later step to your answers. This is not a generic template: the agent reflects a short spec back to you and waits for your thumbs-up before it builds anything. From there it walks the agent through email ingest and cleanup, the AI field extraction with confidence flags, customer/site matching, the coordinator review-and-approve screen, work-order creation, and the CSV export — each step with a ready-to-copy prompt. There's also a fallback so you can build the whole thing today even with no integration to your work-order software.

The governance it includes (this is the point)

This tool drafts records that become real work orders, so it ships with the controls a service operation needs: login so only your team can use it, row-level security so you only ever see your own company's requests, a complete audit trail of who parsed, matched, edited, and approved each request and when, a hard human-approval gate so the AI only ever drafts — a coordinator reviews and approves before a work order is created or exported — and duplicate guards keyed on the email's message-id and on the same customer + problem within a short window, so a request forwarded twice (or sent by both the tenant and the property manager) can't become two work orders.

Who it's for

Dispatchers, intake coordinators, and office managers at small service shops — HVAC, plumbing, electrical, facilities, property maintenance, refrigeration, IT field service — anyone whose work-order queue starts as an inbox and who is tired of retyping emails into a system. If you can describe what your request emails look like and what a complete work order needs, you can build this.

You've got this — start with the plan, paste the first prompt, answer the interview, and you'll watch your first email turn into an approved work order the same weekend.

Gated download

Enter your email — the plan downloads instantly and a copy lands in your inbox.

By submitting your email you'll also receive the weekly runbookify newsletter. You can unsubscribe at any time.