New-Agent Ramp Tracker

The problem this kills

You hire a new support agent, drop them into the queue, and then... you find out six weeks later that they were drowning the whole time. By then it's a retention problem, a CSAT problem, and a "why didn't we catch this" problem.

The signals were all there - their volume was low, their QA scores lagged, their handle time crept up - but they were buried in four different reports nobody compares side by side. Nobody had a single view of "is this person on track for this week of their ramp?"

This tool gives you that view. It takes the metrics you already collect, lines them up against a ramp curve you define, and tells you - this week, not next month - who needs help.

What you'll build

A small, private web app for your training team that:

• Loads each new hire's weekly metrics (volume, QA score, CSAT, average handle time) from a spreadsheet or CSV.
• Compares every agent to a ramp-target curve - the numbers a new hire should be hitting in week 1, week 2, week 3, and so on (these differ by role and complexity, so you set your own).
• Flags behind-target hires early and highlights ahead-of-curve stars worth recognizing.
• Lets a trainer review each flagged agent and approve an intervention plan before it's logged - the AI drafts the suggestion, a human decides.
• Emails a clean weekly ramp digest to the training team via Resend.

What's inside the Implementation Plan

The plan is a single file you paste into an AI coding agent (Claude Code). It walks the agent through building the whole tool, step by step, with a ready-to-copy prompt at the end of each step.

The best part: the plan starts by interviewing you about your business. Before it writes a single line, the AI asks how your team onboards agents - how long your ramp is, what metrics you track and what you call them, what "on target" looks like in week 1 vs week 6, which roles have different curves, and where the messy exceptions live (part-time hires, mid-ramp role changes, holiday weeks). Then it reflects a short spec back to you for a thumbs-up. You get a tool tailored to how you actually work, not a generic template you have to bend to fit.

Inside you'll find: the discovery interview, the data model shaped around your answers, the ramp-curve comparison logic, the at-risk flagging rules, the trainer approval gate, the dashboard, and the Resend email digest - plus a no-integration fallback so you can build it today even if your help-desk has no API.

The governance it includes (this is the point)

This isn't a throwaway spreadsheet macro. The plan builds in the controls that make a tool safe to use on real people's careers:

• Login so only your training team can get in.
• Row-level security so each team or org only ever sees its own agents' data.
• A complete audit trail - who flagged whom, who approved which intervention, and when.
• A human-in-the-loop approval gate - the AI drafts an intervention suggestion, but nothing is logged or acted on until a trainer reviews and approves it.
• Duplicate guards - the same agent-week can't be loaded or flagged twice (dedupe key = agent + week).

Who it's for

Team leads and trainers onboarding new support agents who want to catch a struggling new hire in week 2, not week 8 - and who'd rather coach proactively than do exit interviews.

You've got this. Make a folder, start your agent, and paste the first prompt.