Stage Loss Lanes
Show where people drop out — and which leak to fix first.
Shows where people actually drop out of a funnel: each stage keeps its survivors moving forward while a side lane carries the losses off, labeled with why they left.
Install it in 3 steps
No sign-up. No CLI tools. Pick your AI, copy one line, paste it over.
Prefer doing it yourself? Buy the skill ($4.99), download the ZIP from your checkout confirmation, unzip it into the directory shown above, restart your AI session — same result. Already bought it? Your AI assistant can download it for you.
What’s inside
One SKILL.md (the playbook) + one starter template. A preview:
Stage Loss Lanes
When to use
- You're walking through a funnel — signups, onboarding, checkout, applications — and the headline conversion rate hides where the real bleeding happens
- Your audience keeps asking "where do we lose people?" and the answer is a bar chart that shows totals but not the exits
- You're reviewing a process (hiring pipeline, sales pipeline, course completion) where each stage loses people for a different reason and the reasons matter more than the counts
- You're presenting an audit or post-mortem and need the audience to see, stage by stage, which leak to fix first
- You want a funnel visual that works in a video, not just a slide — one where the flow is visible over time
Workflow
- Define the stages (input: process + data → output: 3–6 stages with counts)
Name 3–6 stages in the audience's language, not the database's: "Visited pricing," "Started trial," "Added payment," "Still active after 30 days." Each stage gets two numbers: how many entered, how many moved on. The gap between them is the loss for that stage. If you have more than 6 stages, merge the ones with tiny losses — a funnel with 11 stages is a spreadsheet with arrows. Every number needs a date range and a source ("product analytics, March 1–31") — a funnel without a date range is a rumor with boxes. - Name the loss reason per stage (input: stage gaps → output: one loss reason per stage)
For each stage, write the single most common reason people left, in plain words: "Price shock at the paywall," "Form asked for a phone number," "Never got the confirmation email." One reason per stage, the biggest one — not a list of five. If you genuinely don't know why people left a stage, label it "Unknown — needs investigation" rather than inventing a reason. An honest unknown is more useful than a confident guess, because it tells the team what to measure next. - Build the main flow (input: stages + counts → output: forward flow)
Draw the survivors' path as a horizontal flow, left to right, with one block per stage. Block width is proportional to the number of survivors — a stage that keeps 80% of people is visibly wider than one that keeps 30%. Animate the flow as moving dots or a stream: a steady trickle of dots entering stage one, thinning as they pass each stage. The thinning must be visible — if the stream barely changes width, the funnel has no story and you should say so rather than exaggerating the drop.
Template preview
Stage Loss Lanes — Funnel Template
One funnel per process per period. Fill in the numbers before drawing.
Process: {name} | Period: {date range} | Data source: {e.g. product analytics}
Stages (3–6, audience-language names)
Merge rule: {more than 6 stages → merge the ones with tiny losses}
Diagram build
Main flow: {horizontal, left → right, block widths proportional to survivors, moving dots/stream} Loss lanes: {one per stage, branch over 0.6s, thickness on same scale as main flow} Lane direction: {downward = recoverable · muted/gray = final} Lane label format: "{—214 · Price shock}"
