AI-Powered LTV Engine

Know Your Customers' Future Value Before They Spend It

Upload your transaction data. Get AI-powered lifetime value predictions and customer segments in minutes. No data team required.

0 weeks prediction horizon
<5% model error rate (MAPE)
5 segments value-based tiers
prediction_summary.csv
SegmentCustomersPredicted Value% Total
⭐ Top 20%4,701€2,847,30262.3%
High4,700€891,44519.5%
Medium4,700€502,19011.0%
Low4,700€248,6705.4%
At Risk4,701€82,3931.8%

Three Steps to Customer Intelligence

From raw transactions to actionable LTV segments in under 10 minutes.

01

Upload Your Data

Drag & drop a CSV with your transactions. Just 3 columns: customer ID, date, and amount. We auto-detect everything.

customer_id, date, sales
02

We Crunch the Numbers

Our engine fits BG/NBD & Gamma-Gamma statistical models to predict each customer's future purchases, spend, and churn probability.

~3 min processing
03

Get Your Segments

Download per-customer predictions and segment summaries. See who's your top 20%, who's about to churn, and where your revenue will come from.

CSV + PDF Report

Everything You Get

Powered by the same models Google built for enterprise advertisers — now accessible to everyone.

Per-Customer LTV Predictions

Every customer gets a predicted future value, expected purchases, future AOV, and probability of being "alive" (not churned).

  • Probability Alive (p_alive)
  • Predicted Purchases
  • Future Average Order Value
  • Total Expected Value

Value-Based Segments

Customers automatically sorted into segments (e.g., Top, High, Medium, Low, At Risk) by predicted future value.

Model Validation

Built-in MAPE (Mean Absolute Percentage Error) validation so you know exactly how accurate the model is on your data.

Privacy-Safe

Your data is processed and deleted. No data is shared with third parties. We never see individual customer identities.

Export Everything

Download per-customer predictions, segment summaries, and a visual PDF report ready to share with your team.

Looker Studio Template

Pro plan includes a pre-built Looker Studio dashboard template. Connect your output data and get instant visual analytics.

What You'll Get Back

Real predictions from the CDNow sample dataset — 23,502 customers analyzed.

Customer ID P(Alive) Pred. Purchases Future AOV Expected Value Segment
C-00142 0.94 12.4 €47.20 €549.85 ⭐ Top
C-08891 0.87 8.1 €62.30 €438.92 ⭐ Top
C-03205 0.71 5.3 €38.10 €143.41 High
C-15440 0.52 2.1 €29.80 €32.54 Medium
C-21003 0.12 0.4 €15.20 €0.73 At Risk
Segment Customers Avg Retention Avg Pred. Value Total Value % of Total
⭐ Top 20% 4,701 89.2% €605.62 €2,847,302 62.3%
High 4,700 72.1% €189.67 €891,445 19.5%
Medium 4,700 51.4% €106.85 €502,190 11.0%
Low 4,700 33.7% €52.91 €248,670 5.4%
At Risk 4,701 8.9% €17.53 €82,393 1.8%
LTV_Analysis_Report.pdf

Your PDF report includes:

  • ✓ Executive summary with key LTV metrics
  • ✓ Segment distribution breakdown
  • ✓ Model validation score (MAPE)
  • ✓ Repeat transaction charts
  • ✓ Top customer highlights
  • ✓ Actionable recommendations

Simple, Transparent Pricing

Start with a free analysis. Upgrade when you're ready.

Monthly Yearly Save 20%

Starter

For teams getting started with LTV analysis

€49 /month
Get Started
  • CSV upload (up to 500K rows)
  • Per-customer LTV predictions
  • 5-segment customer classification
  • Model validation (MAPE)
  • CSV + PDF export
  • 5 saved analyses
  • Results page with key metrics
  • Extra dimension analysis
  • Looker Studio template

Frequently Asked Questions

You need a CSV file with at least 3 columns: Customer ID, Transaction Date, and Transaction Amount. Optionally, you can include an extra dimension (like product category or channel). We recommend at least 12 months of data (18+ months is ideal).

The model uses BG/NBD and Gamma-Gamma statistical models (the industry standard for CLV prediction). Typical MAPE (Mean Absolute Percentage Error) is under 5%. Every analysis includes a validation report so you can see the exact accuracy on your data.

Absolutely. Your data is encrypted in transit and at rest. We process it, generate predictions, and then delete the raw data within 24 hours. We never share your data with third parties. The pipeline is privacy-safe by design — built by Google Cloud's team with privacy as a core principle.

BG/NBD (Beta-Geometric/Negative Binomial Distribution) predicts how many future purchases a customer will make and whether they've churned. Gamma-Gamma predicts how much they'll spend per order. Together, they produce a full CLV prediction. These are the same models used in academic research and by major enterprises.

Yes! LTV Analyzer is platform-agnostic. As long as you can export your transaction data as a CSV (which every ecommerce platform supports), you can use it. It works with Shopify, WooCommerce, Magento, BigCommerce, custom platforms, and more.

Typically 2–5 minutes for datasets under 500K rows. Larger datasets (Pro plan) may take up to 15 minutes. You'll receive an email notification when your results are ready.

Ready to Discover Your Most Valuable Customers?

Join marketers who focus their budget where it matters most.

Start Your First Analysis