funnel-map
Skill6 starsChanged 4 months agoMITView on GitHub
Maps the funnel as it actually is, stage by stage with real numbers, names the conversion question each step must answer, finds the leak that costs the most money, and designs one fix sized to test in two weeks. Use when traffic exists but revenue disappoints, or nobody can say where buyers actually drop.
In this repository29 other files it gives its agents.
- ad-scriptsSkill
- battlecardSkill
- brand-voiceSkill
- campaign-debriefSkill
- case-study-writerSkill
- channel-pickerSkill
- cmoSkill
- competitor-teardownSkill
- content-calendarSkill
- creative-briefSkill
- customer-deep-diveSkill
- email-sequenceSkill
- experiment-backlogSkill
- hook-factorySkill
- kpi-treeSkill
- landing-pageSkill
- launch-planSkill
- lead-magnet-builderSkill
- marketing-auditSkill
- market-research-sprintSkill
- message-mapSkill
- objection-mapSkill
- offer-builderSkill
- onboardSkill
- positioning-workshopSkill
- pricing-testSkill
- repurposeSkill
- research-to-briefSkill
- weekly-reviewSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 24
If key numbers came up UNKNOWN, the first fix is measurement — instrument the missing stages before spending anything on optimisation.
- Very specific: line 25
Update Channels & funnel in marketing-brief.md with the mapped stages and the numbers as of today.
- Distinctive: line 33
Check message match before mechanics.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Funnel Map — find the leak before touching anything
Funnels rarely fail everywhere at once; they fail at one stage while everyone argues about a different one. The map is the argument-ender: every step from first touch to money, the real number at each, the question the buyer is silently asking there — and the discipline to fix the leak that costs the most, not the one that is most visible.
Inputs
- The funnel steps as they exist today, and any numbers per stage: $ARGUMENTS
- From
marketing-brief.mdif present: Channels & funnel (current stages and numbers), the offer (the thing being converted to), and the objections (they predict where buyers stall).
Do this
- Draw the funnel as it IS, not as designed — every real step from first touch to money, plus renewal or repeat if it applies. Attach the actual number at each step. Where a number does not exist, write UNKNOWN — never an estimate dressed as data.
- For each stage, write the conversion question the buyer is silently asking there — "is this for me?", "do I believe this?", "what does it cost me to find out?", "is it safe to say yes now?" — and name the asset currently answering it. A stage whose question has no asset is a leak by design.
- Find the leak that matters. For each drop-off, weigh drop size × volume reaching it × value per person — a 5-point leak at the proposal stage can cost more than a 30-point leak at the top. Benchmark against your own history first, category norms second.
- Diagnose before designing. The usual suspects, in order:
- The promise upstream does not survive to this stage (message mismatch).
- The stage's question has no answer (missing proof, missing clarity).
- The ask outruns the trust built so far (a step too steep).
- Design ONE fix, sized to test inside two weeks, with the single number that will prove or kill it. Inside Claude Code, the funnel-strategist agent can pressure-test the diagnosis and the fix.
- If key numbers came up UNKNOWN, the first fix is measurement — instrument the missing stages before spending anything on optimisation.
- Update Channels & funnel in
marketing-brief.mdwith the mapped stages and the numbers as of today.
Output
The funnel map as a table — stage, number (or UNKNOWN), the buyer's question, the asset answering it, status — followed by the leak with the costing behind it, the one fix with its test and success number, and the measurement gaps.
Rules
- UNKNOWN beats invented. A map with honest gaps drives better decisions than one padded with guesses.
- One leak at a time. Funnels compound — fixing the biggest leak changes every number downstream, so re-map before fixing the next.
- Check message match before mechanics. Most "page problems" are broken promises from the ad or post before it.
- The buyer's question at each stage decides the asset, not the other way round.
- Fixes are sized to learn in two weeks. Big-bang rebuilds hide which change did the work.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

