Intake Completeness & SLA Priority Scorer: Triage Work Orders by Impact, Not by Who Shouted Loudest

The problem this kills

A work order lands. It's missing the equipment ID, the contact is just a first name, and the description says "broken." Three more come in behind it. One is from your biggest contract customer, one is a genuine safety hazard, and one is a routine filter swap from someone who happens to call twice a day. Right now, the order they get worked in depends on who shouted loudest, who the dispatcher happens to like, and whether anyone noticed the contract tier buried in a CRM field nobody opens.

So the urgent-but-quiet jobs slip. SLA clocks blow past their deadline before anyone realizes there was a clock. A VIP account churns because their "high priority" got the same treatment as a walk-in. And half the orders that reach a tech are missing the one field that would have let them fix it on the first visit — so they roll a truck, find they're missing a part, and roll back out again. None of this is a people problem. It's a triage problem, and triage is exactly the kind of repetitive, rules-driven judgment you can hand to a tool. You don't need to be a developer to build it.

What you'll build

A simple internal web tool that grades and ranks your incoming work orders. A work order comes in — typed into a short intake form or imported from your existing sheet — and the tool does two things at once. First, it runs a completeness check: does this WO have everything a tech needs to actually do the job (site, equipment/asset, contact, a usable description, access notes)? It flags what's missing before the order ever reaches a queue. Second, it runs your priority rules — the customer's contract/SLA tier, whether there's a safety risk, whether equipment is down, whether this is a VIP account — and proposes a priority tier plus a concrete SLA due timestamp computed from that customer's contract terms. Critically, it shows the "why" behind every score: "Tier 1 contract (4h response) + equipment down = P1, due today 3:40 p.m."

Then a human steps in. A dispatcher reviews the suggested priority and SLA, and can override it — bumping a job up or down with a reason — before the score is locked onto the work order. Nothing is final until a person says so. The whole prioritized queue exports as a clean CSV in the exact columns your dispatch board or ERP expects, and every score, override, and reason is kept as an audit trail.

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 — your real priority tiers and what each one means, how your contract/SLA levels map to response and resolution times, what counts as a safety risk or "equipment down" in your trade, which fields make a work order genuinely complete, how your WO IDs and customer codes are named, your typical and peak intake volumes, and the messy edge cases (a customer with two contract tiers, an after-hours arrival, a WO that's a duplicate of one logged an hour ago). It then tailors the scoring rules, the SLA math, the completeness checklist, and every later build 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 the data model, the intake form and sheet import, the plain-language editable rules (so you can change a tier or an SLA window without touching code), the scoring engine that explains itself, the dispatcher review-and-override screen, the duplicate guard, and the CSV export — each step with a ready-to-copy prompt. There's also a fallback so you can build and use the whole thing today even with no integration to your existing dispatch system.

The governance it includes (this is the point)

This is a tool you can trust to touch real work orders, because the guardrails are built in from the start:

• Login so only your team can use it.
• Row-level security so each organization only ever sees its own work orders, customers, and rules.
• A human-in-the-loop gate: the tool proposes a priority and SLA, but a dispatcher must confirm or override before anything is locked onto the WO. The AI drafts; a person decides.
• A complete audit trail: who scored it, what the rules suggested, who overrode it and why, and when — for every work order.
• Duplicate guards keyed on your WO ID so the same order can't be scored and queued twice.
• Editable rules in plain language, versioned, so a change to your SLA terms is deliberate, visible, and traceable.

Who it's for

Dispatchers, service coordinators, and service managers in any field-service operation — HVAC, plumbing, electrical, facilities, equipment maintenance, IT field support — who are drowning in incoming work orders and want to triage by real impact instead of by volume of complaints. If you can write down your priority rules on a whiteboard, you can build this.

You've got this — paste the first prompt and let the agent interview you.