VIP & At-Risk Account Flagger

The problem this kills

Your biggest accounts are sitting in the same queue as everyone else. A VIP customer who pays you six figures a year opens a ticket, and it lands behind forty routine "how do I reset my password" requests — handled in order, by whoever's free, with no idea who they are. By the time someone notices it was a key account, the renewal conversation has already gotten awkward.

The accounts most likely to churn are exactly the ones least likely to get noticed in time. Today the only defense is that a sharp agent happens to recognize the company name, or an account manager happens to be CC'd. That's luck, not a process. There's no record of which key-account tickets came in, how fast they were caught, or whether anyone was told.

This tool watches every incoming ticket, matches it against your list of key accounts, and flags the VIPs and at-risk accounts the moment they arrive — then makes a human decide the special-handling path before anything is reassigned or anyone is pinged.

What you'll build

A small internal web app where your team:

1. Keeps an editable, audited list of key accounts (tier, revenue, renewal date, risk flag) right inside the tool.
2. Feeds in incoming tickets — from a CSV export, a paste, or a simple inbound form to start.
3. Watches each ticket get matched to an account by email domain or account ID, with unknown senders handled gracefully instead of silently dropped.
4. Sees a VIP / at-risk flag applied with a plain-language reason ("Enterprise tier, renews in 21 days, marked at-risk").
5. Lets the support lead review the flagged tickets and confirm the handling path — escalate, assign a senior agent, notify the account manager — before any of it happens.
6. On confirmation, the ticket is tagged/escalated and the account manager is emailed, and the whole thing is logged.

What's inside the Implementation Plan

The plan is a complete, paste-and-go runbook for an AI coding agent (Claude Code). You don't write code — you paste, answer questions, and approve.

It opens by interviewing you about your business — how tickets reach you today, what your account list actually looks like (the real field names, how tiers and risk are recorded), how you match a ticket to a company, your volumes, and your exact rules for what counts as VIP or at-risk — so the tool fits how you run support, not a generic template. The agent reads a short spec back to you and waits for your thumbs-up before it builds anything.

From there it walks step by step through standing up the database, importing and editing the account list, ingesting tickets, the matching-and-flagging logic, the flagged-ticket review queue, the lead's confirmation gate, the tag/escalate + account-manager email, the audit log, and the CSV export — each step ending in a ready-to-paste prompt.

The governance it includes (this is the point)

• Login so only your team can open the tool.
• Row-level security so each organization only ever sees its own accounts and tickets.
• A complete audit trail — which ticket was flagged and why, who confirmed the handling, what was sent, and when.
• A hard human-in-the-loop approval gate — the tool flags and recommends; the support lead reviews and confirms the special-handling path; only then is anything reassigned or any account manager emailed.
• Duplicate guards — the dedupe key is the ticket ID, so the same ticket can't be flagged, escalated, or emailed about twice.

Who it's for

Account managers, support leads, and customer-success teams who are protecting key accounts and can't afford to let a VIP or a wobbling renewal sit unnoticed in the standard queue.

You've got this. Open the plan, paste the first prompt, and let the agent interview you.