agentleFS
MadalaVijayMadalaVijay/marketing-skills-deep/skills/unit-economics/SKILL.md

unit-economics

Skill0 starsChanged 2 months agoMITView on GitHub

Work out whether a channel or cohort actually makes money, separating marginal from blended economics and accounting for payback. Use before scaling spend, when comparing channels, or when a profitable-looking channel is not producing cash.

23 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use market-sizing, pricing-strategy and 1 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • OpenAISkill

    market-sizing

    Estimate market, segment, or opportunity size with transparent assumptions and uncertainty. Use for TAM/SAM/SOM, sizing scenarios, or comparing the scale of possible opportunities.

    Also in 3 projects that use this

  • silvainfmSkill

    pricing-strategy

    When the user wants help with pricing decisions, packaging, or monetization strategy. Also use when the user mentions 'pricing,' 'pricing tiers,' 'freemium,' 'free trial,' 'packaging,' 'price increase,' 'value metric,' 'Van Westendorp,' 'willingness to pay,' or 'monetization.' This skill covers pricing research, tier structure, and packaging strategy.

    Also in 3 projects that use this

  • birne-skSkill

    market-research

    Professional market research and competitive analysis using McKinsey/BCG/Bain consulting methodologies. Use when analyzing markets, competitive landscapes, customer insights, or strategic positioning. Generates hypothesis-driven research with PPTX presentation deliverables.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 17 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    4. Retention-adjusted LTV: Only after the above. LTV built on an optimistic retention curve is the most common way a bad channel looks good.

    line 31

“Common” is measured: kept by at least 6 in 10 of the 17 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Unit economics

The question is never "what is our ROAS". It is "does the next unit of spend make money, and can we survive the wait".

Four numbers, and the order they matter

1. Contribution margin per customer, not revenue. Revenue minus COGS, payment fees, shipping, support and any per-unit variable cost. Marketing against revenue overstates every channel, and by more in low-margin businesses.

2. Marginal CAC, not blended. Blended CAC averages your cheapest historical customers with your most recent ones. Approximate the margin:

marginal CAC ≈ (spend_recent − spend_prior) / (customers_recent − customers_prior)

using two adjacent equal-length windows. It is crude and seasonality-sensitive. Say so. It still answers the question blended CAC does not.

3. Payback period. How many months of contribution margin to recover CAC. A business with three months of runway cannot fund a nine-month payback at any return multiple. This kills more companies than bad ROAS does.

4. Retention-adjusted LTV. Only after the above. LTV built on an optimistic retention curve is the most common way a bad channel looks good.

The trap

A channel at 3x blended return can be losing money on its most recent spend while the average still looks healthy, because early cheap conversions drag it down. You scale, the average worsens, and the scale-up gets blamed. It was already unprofitable at the margin before you touched it.

Always compute both, and lead with the marginal number.

LTV honesty

  • Use realised retention where you have it. A cohort with three months of history gives you three months of LTV, not a projection to 24
  • State the retention assumption explicitly whenever you project
  • Cap the horizon at something the business can finance. LTV over 36 months is irrelevant to a company that needs cash in six
  • Segment it. LTV averaged across a whale and a churner describes neither

Ratios, and their limits

LTV:CAC of 3:1 is a convention, not a law, and it is silent on timing. Two businesses with identical 3:1 ratios and payback of 2 months versus 14 months are in completely different situations. Report the ratio and the payback, or the ratio misleads.

When to refuse

  • Contribution margin unknown. Ask; do not substitute revenue
  • Fewer than ~30 conversions in the window
  • Cohorts too young to observe retention, and a projection would carry the answer
  • Attribution not verified — see the data-trust skill first
  • Delivery cost is per slot rather than per customer. Where a class, van, appointment or route costs roughly the same however many customers it serves, per-customer margin is a function of occupancy and this skill will understate the real lever. Use capacity-economics instead

Output

Contribution margin per customer, blended CAC, marginal CAC, payback in months, and the retention assumption stated in words. Then a verdict: scale, hold, or stop, with the single number that would change it.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.