This website uses cookies for analytics purposes. Learn more

Lender Support

Attrition & Lifetime Value Modelling

  • Client: New Unsecured Lender
  • Project Length: Ongoing
  • Personnel: Partner, Consultant + Analyst Team
CS Attrition 1


Understanding and predicting customer attrition was a key need for the client, to ensure resources allocated for retention strategies were well focussed

Additionally, understanding the lifetime value of a typical customer by modelling attrition behaviour in combination with channel-specific costs of acquisition was of high importance for the client

The insights were required to be delivered in a short time frame, with the key findings distilled for senior management and investors

CS Attrition 2

Initial Insight

Survival analysis by cohort was first conducted to understand the baseline performance of the product

Univariate and multivariate analysis was subsequently completed to identify the key drivers of customer attrition, and segment the population for targeted retention strategies

The mix and stability of the predictive variables over time was assessed, as it is imperative to understand any underlying shifts in performance

Graph 01

Attrition & Value Model

Deploying Vestigo’s Machine Learning expertise, a model to predict customer attrition was constructed

The net returns at a channel and score segment level were modelled, highlighting value-eroding segments

Continuing to analyse the lifetime value of customer segments in this way will be revolutionary for the client, enabling the product to be refined, and business profitability enhanced

See how we can help

Contact the
Vestigo team

By submitting this form, I agree to Vestigo’s Privacy Policy

Experienced risk and analytics consultancy

Benjamin davies Oja2ty 9 ZLM unsplash

Building analytical solutions that are focused on your real-world objectives

Get the latest news from Vestigo

Sign up below to receive future Vestigo news and updates to your inbox.
© Copyright 2024 Vestigo Partners Ltd Registered Company Number 10826517 | T&Cs | Privacy Policy