CLVTools

CLVTools is a free, GPL-3-licensed R package for customer lifetime value modeling and individual-level forecasts. It is best suited to analysts working with non-contractual customer histories who want established probabilistic models and control over model extensions. Its focused scope and requirement for R make it a specialist tool rather than a general-purpose customer platform.

Overview

CLVTools models purchase and attrition behavior in settings where customers can stop buying without a formal cancellation. It predicts expected future transactions and each customer’s probability of still being active, which can support individual forecasts as well as cohort analysis and segmentation.

The models are typically applied to cohorts whose customers joined at the same time. That makes the package a more natural fit for aligned customer groups than for analysis that mixes very different acquisition periods without accounting for them. Input consists of transaction history with customer IDs and purchase dates; the example also uses prices.

Key features

Probabilistic model selection

The package implements Pareto/NBD, BG/NBD, Gamma/Gompertz/NBD, and Gamma/Gamma. This gives analysts several established model choices within a single R package, including purchase-frequency and monetary-value modeling. The individual-level forecasts are useful when a cohort average would hide meaningful differences between customers.

Covariates and extensions

Time-invariant covariates are supported across the models, while time-varying covariates are available for some. Extensions include correlation between purchase and attrition processes, L2 regularization for contextual-factor parameters, and equality constraints between those parameters. These options make CLVTools more adaptable to structured modeling questions, though they also place it firmly in an analyst’s toolkit rather than a turnkey workflow.

Installation and documentation

The stable release is installed from CRAN; a development version is available from GitHub through devtools. The project provides a walkthrough, reference manual, and vignettes on modeling, advanced techniques, and package classes. Building from source requires the external GNU Scientific Library (GSL), an extra dependency to account for when setting up an environment.

Pricing

CLVTools R package: free. It is open source under GPL-3 and requires R. The price makes it accessible for individual analysts and teams able to work in R, but it does not remove the cost of the surrounding work: users need to prepare transaction data, choose and fit models, and interpret forecasts themselves. Compiling from source requires GSL.

There are no paid tiers described for the package, so there is no higher plan to add managed infrastructure or a different workflow. For readers seeking a packaged customer-marketing service instead of a modeling library, a free R package may not meet that operational need.

Platforms

CLVTools supports Linux, macOS, and Windows. Installation guidance covers Linux, Intel and Apple Silicon Macs, and Windows. In every case, use depends on having R; building from source also requires GSL.

Who it's for

CLVTools is a strong candidate for data analysts and researchers who need probabilistic customer forecasts from non-contractual transaction histories, particularly when customer cohorts are meaningfully aligned. Its model range and covariate options suit users who want to select and extend statistical models rather than rely on a ready-made business interface.

It is less suited to teams seeking a hosted analytics product, a general customer engagement system, or modeling outside non-contractual purchase behavior. The package’s R and data-preparation requirements are also a poor fit for users who need forecasts without statistical tooling.

Pros and cons

• Pros: Four implemented model families give analysts choices for modeling repeat purchases and monetary value.
• Pros: Individual forecasts include expected transactions and probability of being alive, enabling customer-level analysis rather than cohort totals alone.
• Pros: Covariates, regularization, correlation, and parameter constraints support more tailored model specifications.
• Pros: Free, GPL-3 licensing and documented CRAN installation lower the cost of adopting the package.
• Cons: The non-contractual focus limits its relevance for contractual businesses or other prediction tasks.
• Cons: Its typical cohort assumption means users need to consider whether their customers’ joining times are comparable.
• Cons: R is required, and source builds add a GSL dependency, creating setup work for users outside an R environment.

Alternatives

For a broader comparison of tools in this category, browse Customer Lifetime Value Software.

• PyMC-Marketing is another free, open-source modeling library; choose it if you prefer an Apache 2.0 Python library to an R package.
• Pushwoosh has a free plan with web, desktop, and mobile platforms; choose it when the need is a multi-channel product rather than an R modeling package.
• RevenueCat offers a free tier up to $2,500 in monthly tracked revenue, then charges 1% of tracked revenue; choose it if that revenue-based pricing and its listed platform range better match your use case.
• VOZIQ AI Platform is available free of charge with BYOL, though AWS infrastructure costs may apply; consider it if that deployment arrangement better suits your needs.
• WorthSignal is a free alternative.
• Klaviyo has a free plan capped at 250 active profiles, 500 email sends per month, and $5 of monthly mobile messages; choose it if those messaging limits fit better than a modeling library.
• SaaSFlow offers a free plan up to €120k ARR (€10k MRR), with all features, unlimited users, and unlimited data sources; consider it if those SaaS-focused limits suit your needs.
• Siftmo has a free Starter plan capped at 50 monthly orders and paid plans for higher order volumes; consider it if order-based pricing better fits your use case.

Verdict

Choose CLVTools if you work in R and need a free, flexible set of probabilistic models for individual forecasts from non-contractual purchase histories. Its range of models and extensions is a substantive advantage for analysts, but the package’s narrow modeling scope and technical setup make another tool a better fit for teams seeking a broader platform or a non-R workflow.