Orphex Customer Cohort & LTV Review
A horizon-aligned cohort-value review with CAC, observed repeat behavior, maturity coverage, and explicit lifetime/profit limits.
When to use
When comparing acquired customer value and repeat behavior at the same observed age.
Bring the right data
Use your own export, or start with the template. The fictional example shows the expected shape.
View the input columns
cohort_idRequired- Acquisition cohort with a stable supplied membership rule
currencyRequired- ISO currency
customersRequired- Unique acquired customers under the stated definition
observed_daysRequired- Minimum fully available follow-up for every cohort member
common_horizon_daysRequired- Same per-customer elapsed observation horizon for comparison
revenue_at_common_horizonRequired- Defined cohort revenue observed within the common elapsed horizon
revenue_to_dateOptional- Revenue over each cohort own available horizon; do not compare unaligned horizons
acquisition_costRequired- Cost attributed to acquiring these unique customers, not all period orders
repeat_customers_at_common_horizonOptional- Unique cohort customers with at least one repeat order in common horizon
Add your business context
Supply your objectives, conversion definitions, currency, constraints, and approved brand facts once, then reuse the profile across reviews. Leave unknown values explicit.
See an example
Fictional data · an illustrative review, not a customer result.
Compare customer value without extrapolating the younger cohort.
View the example result
Fictional example output
| Cohort | Common observed horizon | Net revenue / acquired customer | CAC | Observed revenue-to-CAC ratio | Repeat-customer fraction |
|---|---|---|---|---|---|
| June | 60 days | $8,000/100 = $80 | $3,000/100 = $30 | 2.6667 | 50/100 = 50.00% |
| July | 60 days | $7,000/100 = $70 | $3,000/100 = $30 | 2.3333 | 40/100 = 40.00% |
July's observed 60-day revenue per acquired customer is $10 lower (−12.50%). June's $9,000 at 90 days must not be compared with July's $7,000 at 60 days as equal-horizon lifetime value. This is observed 60-day value, not a lifetime forecast or profit LTV; CAC ratios do not include service/product costs.
Membership and elapsed maturity are supported by the fictional export guarantee. Without per-customer age coverage in an actual export, alignment would remain unverified. Next: compare acquisition mix, offers, returns and repeat ordering at the same horizon. No projected lifetime multiple, payback date, or bid-target change was invented.
Before you start
Review the required context
- Supplied data with the documented task-specific columns, stable scope, and refresh/maturity context
- Business definitions and constraints relevant to the decision; see the reusable business-context reference
The method
Align cohorts before comparing
Define new/acquired customer membership, acquisition date, attribution and cost allocation, unique identity rules, product/business model, currency, revenue/cost basis, snapshot, and per-customer follow-up. Use the same elapsed horizon from acquisition for every eligible customer; a monthly cohort's average age or earliest member age cannot certify maturity for the whole cohort. Exclude right-censored periods from an equal-horizon comparison or disclose a valid censoring-adjusted method rather than extrapolating automatically.
View SKILL.md
Orphex Customer Cohort & LTV Review
Review customer cohort value, repeat behavior and acquisition economics at a common observed horizon. A skill title containing LTV does not make a finite observation a lifetime estimate.
Align cohorts before comparing
Define new/acquired customer membership, acquisition date, attribution and cost allocation, unique identity rules, product/business model, currency, revenue/cost basis, snapshot, and per-customer follow-up. Use the same elapsed horizon from acquisition for every eligible customer; a monthly cohort's average age or earliest member age cannot certify maturity for the whole cohort. Exclude right-censored periods from an equal-horizon comparison or disclose a valid censoring-adjusted method rather than extrapolating automatically.
Net/gross revenue, collected cash, recognized revenue, ARR and contribution are different numerators. For SaaS, a booked annual contract or ARR is not observed recognized cash over 60 days. Define repeat purchasers, active/subscription retention, renewals or churn against the original eligible denominator; do not mix transaction count with unique repeat customers. For repeat metrics, preserve customers with zero subsequent activity when still observable. Consent/deletion or identity gaps are missingness limits, not churn evidence.
Calculate supported value
Observed horizon value per acquired customer = matching cohort value / unique acquired customers. CAC = scoped acquisition cost / those customers. Aggregate from sums, preserve currencies, and avoid zero-denominator ratios. Label revenue-to-CAC separately from contribution-to-CAC; neither is company ROI without the full cost definition. Payback requires cumulative contribution/cash by elapsed time; it cannot be inferred from a revenue multiple alone.
A lifetime forecast needs an explicitly supplied/validated retention or survival model, horizon, cost path, discounting assumptions when applicable, calibration and uncertainty. Do not invent a retention multiplier, claim a younger cohort's ultimate value, or bid to a forecast from these aggregates. Use the optional calculator only for common-horizon arithmetic, not survival modeling.
Deliver observed horizon coverage, weighted value/CAC/repeat metrics, economic definition, uncertainty and a next cohort check. Acquisition attribution is not causal lift; budget or bidding changes remain bounded proposals.
Portable inputs and examples
- Read the input contract when mapping a new export or checking the example's scope and definitions. Copy the header-only CSV template when preparing data; equivalent supplied exports remain acceptable.
- Read the reusable business context only for business facts or constraints this task needs. Reuse user-supplied facts with their source/date; the template contains no default targets.
- Inspect the complete fictional input with its example output when learning the output and calculation boundaries. Never use fictional values for a real account.
State whether the result is complete, partial, or blocked for the requested decision. Link material findings to actual supplied rows/sources and separate observed metrics, hypotheses, and estimates. Lead with a short business conclusion, then evidence, uncertainty, and the next measurable check. A data export or installed skill does not authorize account changes.
For the supported arithmetic only, optionally run the bundled calculator with Python 3: python3 scripts/marketing_math.py cohort < calculation.json. Read its input mapping in the input contract before preparing JSON. It reads JSON, not CSV directly. If Python or the requested method is unavailable, show a reproducible alternative calculation or mark it unsupported; do not report an uncomputed result as verified.
Source:View on GitHub Open SKILL.md