Lead Scoring Engine

The problem this kills

Right now your "lead score" probably lives in someone's head, or in a spreadsheet formula nobody fully trusts, or it's locked inside an expensive marketing automation platform that bills you per contact and still hands sales a pile of half-warm names. When a rep asks "why is this lead a 92 and that one a 41?" you can't answer in plain English. So sales quietly ignores the score, works leads in whatever order they feel like, and the good ones go cold while someone chases a tire-kicker.

The real problem isn't math - it's trust and control. You need a score sales believes, that you can defend in a meeting, that you can change when the market shifts, and that a human approves before anyone gets stamped "MQL" and dumped into a rep's queue.

What you'll build

A small, private web tool that does one job well: take your leads, give each one an honest score, and let you hand sales the genuinely hot ones - on purpose, not by accident.

• Import a leads CSV (or pull from your CRM later) and merge duplicates by email so nobody gets double-counted.
• Score every lead two ways: fit (industry, company size, role - is this even our kind of buyer?) and behavior (page views, downloads, email opens and clicks - are they actually interested right now?), then combine them into one number.
• Rank all leads and flag the ones over your MQL threshold (the line where a lead is "ready for sales").
• Show a clear breakdown per lead - exactly which signals added or subtracted points - so the score is never a black box.
• Let marketing ops tune the weights, see the rankings re-shuffle live, and log a reason for each change.
• Put a human approval gate in front of the handoff: ops reviews the proposed MQL list and approves the batch before any lead is promoted and exported for the CRM.

What's inside the Implementation Plan

The plan is a complete, paste-and-go runbook. Its best feature: it starts by interviewing you about your business - your funnel, your CRM, the actual columns in your leads file, your buyer's industries and roles, your typical and peak lead volumes, your scoring instincts, and your messy edge cases (free-email signups, competitors, students, decayed old leads). It reflects a short tailored spec back to you, gets your thumbs-up, and only then builds. You get a tool shaped around how you sell, not a generic template.

Inside you'll find: a tailored data model, the import + dedupe step, the transparent scoring engine, the weight-tuning screen, the human approval gate, the audit trail, and the CRM-ready CSV export - each step ending in a ready-to-copy prompt you paste into your AI coding agent.

The governance it includes (this is the point)

This isn't a toy spreadsheet replacement. The plan bakes in real controls from the first build step:

• Login so only your team can open the tool.
• Row-level security so you only ever see your own organization's leads.
• A complete audit trail - every weight change, every approval, who did it and when, with the reason logged.
• A hard human-in-the-loop approval gate - the engine drafts the MQL list, a person reviews and approves, and only then are leads promoted and exported.
• Duplicate guards - lowercased email is the merge key, so the same lead can't be scored or promoted twice.

Who it's for

Marketing ops and demand-gen folks who need a scoring model they can defend and adjust - without buying (or fighting) a pricey automation platform. If you can use a spreadsheet and write a clear sentence about why a lead matters, you can build and run this. No coding background required.

You've got this - paste the first prompt and let the interview tailor the rest.