Statistical LTV Analysis

Everyone wants their LTV.
Nobody measures it right.

Most LTV estimates fall apart when someone asks how they were calculated. LTV Analyzer applies survival analysis — using only minimal customer data you already have — to easily produce defensible LTV estimates with the same statistical methods used in clinical research.

Survival
Analysis
Statistically rigorous
LTV∞
Expected lifetime value
CAC
Ceiling
Know your limit
LTV Analyzer Pro
The problem

Why most LTV numbers
don't hold up

Whether you're defending a valuation, justifying ad spend, or reporting to the board — LTV figures are central to the conversation. Yet the methodology behind them rarely survives a skeptical audience.

I.

Back-of-envelope calculations

Simple average revenue × average tenure ignores churn patterns, censored data, and the statistical uncertainty that any rigorous buyer or investor will immediately probe.

II.

Methodology that can't be explained

When someone in the room asks how the number was calculated, a back-of-envelope method has no answer. The credibility of the estimate collapses with the first follow-up question.

III.

No time-horizon flexibility

Most LTV estimates produce a single number with no way to specify when. LTV Analyzer lets you define the horizon — 1 year, 3 years, 5 years — so you can compare, plan, and present on your own terms.

Methodology

Clinical-grade,
zero configuration

When someone asks how your LTV was calculated, you need an answer that doesn't collapse. The engine behind this tool uses the same statistical framework applied in clinical trials and actuarial science — no statistical expertise required to get expert-level results.

Kaplan-Meier Estimator

Accounts for what simple averages miss

The Kaplan-Meier estimator handles the fundamental challenge of customer data: censoring. Customers still active at analysis time haven't churned yet — but their data still carries information. This is why survival analysis exists — and why simple averages fall short.

Survival Curve
Weibull Distribution

Projects LTV beyond your observation window

Fitting a Weibull distribution to the hazard function allows projection beyond the observed time window. The shape parameter κ reveals whether churn risk increases, decreases, or stays constant over time — critical context for any forward-looking valuation.

Weibull Linearization Plot

"The combination of non-parametric estimation with parametric projection is the gold standard for survival analysis in contexts with censored, time-to-event data."

LTV by Observation Period LTV Horizon Table
How it works

From raw data to
defensible numbers

Upload a minimal dataset and get statistically rigorous LTV estimates in minutes — no data science team, no complex setup required.

Step 01

Drop your CSV and go

Prepare a simple CSV — just customer IDs, dates, and revenue. No personal data required. Set your business parameters, hit Analyze — results appear in seconds.

CSV data
Step 02

Rigorous analysis — without the statistician

Outlier detection, statistical modeling, LTV projections — methods that normally require a data science team. They run automatically. You get the rigor without needing the expertise.

Outlier Detection
Removes noise
Survival Analysis
Kaplan-Meier estimation
Weibull Distribution
Parametric hazard model
LTV Projection
Revenue × retention
Step 03

Results your audience already understands

Charts, tables, and segment comparisons — all formatted for business conversations. Export to PowerPoint or PDF in one click. Walk into your next meeting ready.

Slide 1Slide 2Slide 3Slide 4Slide 5Slide 6Slide 7Slide 8Slide 9

See it in action →

Pricing

Own it. No subscription.

Perpetual access. No monthly fees. Pay once, use it on every engagement.

Standard
Full LTV∞ analysis on your entire customer base — revenue and gross profit basis, with CAC ceiling built in.
$499
one-time purchase
  • Kaplan-Meier survival curve
  • Weibull LTV projection (k, λ)
  • LTV∞ — revenue & gross profit basis
  • LTV & CAC ceiling by time horizon (1yr, 3yr, 5yr)
  • CAC ceiling (set your LTV:CAC ratio)
  • R² model fit score
  • Revenue outlier filtering (percentile-based)
  • Export: Excel · PowerPoint · PDF
  • AI interpretation prompts included
  • Segment-level LTV analysis (Advanced)
Get Standard →
Requirements

Recommended minimum of 1,000+ customer observations for the overall analysis. Segment analysis (Advanced) requires 200+ observations per segment for reliable Weibull fitting. Smaller segments will run but R² fit score may be lower.

Built by

A consultant who needed
this tool

TH
Tomotake Hirata
Management Consultant  ·  Angel Investor

I built LTV Analyzer because the problem keeps appearing across consulting engagements: businesses need LTV estimates that can withstand scrutiny — in campaign planning, in CAC optimization, in fundraising decks, in PE due diligence — but the available tools are either too simple or require a data science team.

My work sits at the intersection of statistics and strategy. I don't just read the numbers — I build them, pressure-test them, and translate them into decisions. LTV Analyzer is the tool I wished I'd had on the projects I was running.

Past engagements span major financial institutions, global e-commerce, worldwide food delivery chain, apparel, and online healthcare — bringing data-driven analysis and strategic insight to each.

FAQ

Common questions

How is this different from a simple LTV calculator?
Simple calculators use avg. revenue × avg. tenure — ignoring churn patterns and censored data. LTV Analyzer uses survival analysis (Kaplan-Meier + Weibull) to produce statistically grounded LTV estimates that hold up under scrutiny.
What does LTV∞ mean in practice?
LTV∞ is the theoretical lifetime value derived from integrating the Weibull survival function — the expected total revenue (or gross profit) if a customer remained indefinitely. The tool outputs both a revenue-based and gross profit-based figure. In practice, you can apply any time horizon (1 year, 3 years, 5 years) to compare scenarios on your own terms.
Can I compare LTV across customer segments?
With the Standard plan, you can run separate analyses per segment — for example, one CSV per acquisition channel, age group, or region. The Advanced plan handles this automatically: upload a single file with up to five segment columns, and the tool runs segment-level comparisons and ranks your highest-value segments in one report.
What exactly does the tool output?
The Kaplan-Meier survival curve, Weibull fitting parameters (κ and λ), and the expected customer lifetime value — the theoretical LTV derived from integrating the Weibull model. Results export directly to Excel, PowerPoint, and PDF.
What data format do I need?
Less than you'd expect. A minimal CSV with anonymized customer IDs, start date, churn date (or active status) / last purchase date, and total revenue per customer. No personal data required. The tool handles censored observations — customers still active at analysis time — automatically.Advanced — add up to five segment columns to the same CSV. One file, full breakdown.
Do I need a statistics background to use this tool?
No. The tool handles all statistical modeling internally. You don't need a statistics background — just the ability to make business decisions based on the results. Each report includes plain-language explanations, so you can interpret the output and act on it without technical expertise.
How does the CAC ceiling work?
The tool calculates LTV∞ on a gross profit basis. You set your target LTV:CAC ratio using a slider (e.g. 3.0:1), and the tool automatically derives your CAC ceiling. You decide the ratio — the tool does the math.
Why do I need to enter GPM?
GPM (gross profit margin) lets the tool output LTV on both a revenue basis and a gross profit basis. The gross profit LTV is what matters for CAC decisions — spending against revenue LTV overstates how much you can actually afford to acquire a customer. For businesses where GPM varies across products or services, use your blended average GPM as a starting point — or split your dataset by category and run separate analyses for more precise results.
What are the AI prompts included?
Each report includes pre-written prompts you can paste into Claude, ChatGPT, or Gemini to interpret your results — covering model fit, marketing strategy, and analytical validation. No statistics background required to get actionable insights.
Why 1,000+ observations?
Kaplan-Meier estimation and Weibull fitting both require sufficient sample size for a reliable R² fit score. Below ~1,000 observations the model will run but R² may be lower and results should be treated as directional only. For segment analysis (Advanced), 200+ observations per segment is recommended — below that, Weibull parameter estimation becomes unstable.
Can I use this for client deliverables?
Yes. The single-user license covers use across your client projects. For team or firm-wide licensing, contact us directly.
Get started

LTV numbers you can
defend in any room.

For digital agencies, in-house marketing teams, growth consultants, M&A advisors, and PE analysts who need statistically rigorous LTV — without hiring a data science team.