funnel-strategy
Skill3 starsChanged 32 days agoMITView on GitHub
Works out a company's go-to-market motion (product-led, sales-led or hybrid) from its website and pricing, then maps the funnel stage by stage: a definition, strategy, leading inputs the team controls, lagging outputs with targets, cadence and owner for each stage, MQL, SQL and PQL criteria, post-sale stages, closed-lost re-entry paths and backward funnel maths from a revenue target. Triggers: "funnel strategy", "funnel stages", "pipeline stages", "PLG vs SLG", "lead qualification criteria", "map our sales process", "how do leads flow". Feeds /goals. Not for setting the target itself (use /goals).
3 other projects use a version of this.
Adopted by
Most recently taken up by
- matteotitta/repo-structure-agencyword for word, first seen Nov 8, 2026
- fnlfx/aitheir own version, first seen Oct 15, 2026
- matteotitta/genesys-skillstheir own version, first seen Jul 30, 2026
Other versions3 projects use a version of this.
Word for word · 1
- matteotitta/repo-structure-agency★ 1100% same
Edited · 0
None found.
Wrote their own version · 2
- matteotitta/genesys-skills★ 46same name
- fnlfx/ai★ 26same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository10 other files it gives its agents.
- brand-kitSkill
- company-contextSkill
- competitor-researchSkill
- expert-povSkill
- goalsSkill
- icp-researchSkill
- messagingSkill
- positioningSkill
- tov-guidelinesSkill
- win-lossSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
When marketing and sales disagree on what a qualified lead is.
- Very specific: line 24
Don't use it to set the revenue target (/goals takes a target someone has already committed to) or to design individual campaigns.
- Very specific: line 32
Paths: Write to context/goals/MMYY-funnel.md, next to the scope file /goals will write.
- Distinctive: line 43
Unclear: ask the user.
- Distinctive: line 54
Never put a metric in the inputs.
- Distinctive: line 58
Run the backward maths: if there's a revenue target.
- Distinctive: line 77
Every input has a frequency, every output a target, every stage a cadence and an owner.
≈ “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 strategy
Map how a buyer moves from first touch to renewal, in the shape this company actually sells. The motion decides the stages: a product-led company and a sales-led company should not share a funnel. Every stage separates what the team does (inputs) from what it measures (outputs), and the maths runs backwards from the revenue target.
When to use it
- Building the context folder, after positioning and messaging, before
/goals. - Setting up or cleaning up CRM stages.
- When marketing and sales disagree on what a qualified lead is.
Don't use it to set the revenue target (/goals takes a target someone has already committed to) or to design individual campaigns.
Inputs
Required: the company website URL.
Recommended: context/company/ (deal size, team size, funding sharpen the motion call), context/icp/ (buying triggers shape qualification), the real CRM stages if they exist (they override anything inferred), pricing or average deal size.
Paths. Write to context/goals/MMYY-funnel.md, next to the scope file /goals will write. Agency or consultant? Swap context/ for clients/<client>/context/ throughout.
Tools. Page fetching and web search. Exa helps with company research if you have it.
Steps
- Read the signals. Homepage CTA ("Start free" vs "Book a demo"), pricing (self-serve checkout vs "Contact sales"), free trial or freemium, docs and API in the navigation, average deal size, sales team size.
Call the motion.
- Product-led: signals point to self-serve.
- Sales-led: they point to demos and sales.
- Hybrid: both are strong.
- Unclear: ask the user.
State it with your two or three strongest signals and a confidence level.
- Pick the stage set for that motion only. Don't mix product-led stages into a sales-led funnel. Hybrid runs two tracks with a rule for when a product user gets routed to sales.
Define each pre-sale stage:
- a definition for this company
- the strategy
- inputs: leading activities the team controls, with a frequency ("3 LinkedIn posts a week")
- outputs: lagging results with a target ("500 visitors a month")
- cadence and owner
Never put a metric in the inputs.
- Set qualification criteria. 3 to 5 measurable signals each for MQL (marketing to sales), SQL (confirmed by sales) and, for product-led or hybrid, PQL (product usage). "Good fit" isn't a criterion; "50+ employees and visited pricing twice" is.
- Map closed-lost re-entry. Loss reason, which stage they re-enter, the trigger and the timing.
- Map post-sale stages in the same format: onboarding (the path to first value), retention, expansion, churn prevention. Match the motion: product-led is in-app, sales-led is account-manager driven.
- Run the backward maths if there's a revenue target. Revenue to deals, deals to opportunities, then SQLs, MQLs, leads and traffic, using real conversion rates where you have them and marked estimates where you don't. Check whether current volumes can get there.
- Draw the journey as a simple text diagram from first touch to expansion.
- List gaps and recommendations, then hand off to
/goals.
Output shape
- Motion assessment table: signal, what you saw, what it points to.
- Pre-sale stages, each in the same format.
- Qualification criteria.
- Post-sale stages.
- Closed-lost re-entry table.
- Journey diagram.
- Backward maths (if a target exists).
- Gaps and recommendations.
- Sources with URLs and confidence.
What good looks like
- The motion is backed by at least three signals.
- Every input has a frequency, every output a target, every stage a cadence and an owner.
- Inputs and outputs never swap places.
- Qualification criteria are thresholds someone could put into a CRM.
- Post-sale matches pre-sale. A product-led company doesn't suddenly get a customer-success-led renewal motion.
Mistakes to avoid
- Inventing stages the company doesn't run, instead of asking for the real ones.
- Assuming the motion from the category ("it's dev tools, so it's product-led").
- Making up conversion rates or KPI targets without marking them as estimates.
- One funnel template forced onto every company.
Chain
Step 10 of the suggested chain: /company-context → /icp-research → /competitor-research → /win-loss → /expert-pov → /tov-guidelines → /brand-kit → /positioning → /messaging → funnel-strategy → /goals.
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.

