Win-Rate & Conversion Funnel Analytics

The problem this kills

You know deals are leaking somewhere in your funnel. You just can't prove where. The CRM dashboard shows a single win rate, but it can't tell you that mid-market deals stall between Qualified and Opportunity, or that one source produces leads that never convert, or that your sales cycle quietly stretched by three weeks last quarter.

So revenue ops ends up rebuilding the funnel by hand in a spreadsheet every quarter - re-pasting deal exports, arguing about which deals are "test" deals, fighting over what counts as a real opportunity, and shipping a number nobody fully trusts. The next quarter, it all happens again.

The real problem isn't the math. It's that the definitions and exclusions live in someone's head, the numbers aren't reproducible, and nobody approved them before they went out as official. This tool fixes that.

What you'll build

A small internal web app - yours, private, and login-protected - that:

• Imports your historical deals and their stage history (when each deal entered each stage, its source, segment, owning rep, and final outcome).
• Computes the funnel: stage-to-stage conversion rates, overall win rate, average deal size, and sales-cycle length.
• Slices everything by source, segment, and rep so you can find the single weakest stage.
• Handles the messy reality: deals that skip stages, test/junk deals, and duplicates - all excluded consistently, with a clear paper trail of what was excluded and why.
• Puts a human gate before "official": ops reviews and approves the stage definitions and any exclusions, and only then is the funnel report published - as a numbered, versioned snapshot.
• Exports a clean CSV in the exact column shape your downstream tools and decks expect.

What's inside the Implementation Plan

The plan is a single markdown file you paste into Claude Code. It walks the AI agent through building the whole tool, step by step, each step ending with a ready-to-copy prompt.

Crucially, the plan opens by interviewing you about your business - your stages and what they really mean, the systems and spreadsheets you pull deals from, how your data is actually named, your typical and peak deal volumes, your exclusion rules, and your messiest edge cases. It reflects a short tailored spec back to you and waits for your thumbs-up before it builds a thing. You get a tool shaped around your funnel, not a generic template that assumes everyone sells the same way.

Inside you'll find: a definition of done, the discovery interview, the data model, the conversion/cycle-time math, the slice-and-dice dashboard, the ops approval gate, report versioning on publish, and the CSV export - plus the "No API yet?" path so you can build the entire thing this weekend straight from a Google Sheet, with zero CRM integration.

The governance it includes (this is the point)

This isn't a throwaway script. Every runbookify plan bakes in the controls that make a number trustworthy enough to act on:

• Login so only your team can open the tool.
• Row-level security so people only ever see their own organization's data.
• A complete audit trail - who changed a definition or exclusion, and when.
• A human-in-the-loop approval gate - the tool drafts the funnel and the exclusion list; a person reviews and approves before anything is published as official.
• Duplicate guards keyed on deal ID, so the same deal never gets counted twice in the funnel, and a report version stamped on every publish so last quarter's official number is always reproducible.

Who it's for

Revenue operations leads and sales managers who are done eyeballing a single win-rate number and want to find - and fix - the weakest stage in their funnel. If you can keep a deal export in a spreadsheet, you can build this. You don't need to be a developer.

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