Businesses constantly look for ways to keep customers coming back to their products. Two of Vidora’s media partners recently used Cortex to power automated systems to reduce churn. Cortex went through a two-step process:

  1. Predict which users are at risk of churn
  2. Automate personalized outbound communication to re-engage at-risk users

The net result? Cortex helped significantly reduce churn, translating to massively increased revenues in both cases.

Case Study 1: Subscription Video on Demand Business

Our partner in this case study is a Subscription View on Demand business, with a large catalogue of film and television titles that users can access online as well as in their app. For this business, a churned user is one who unsubscribes from their service.

To re-engage users, this business had a series of ‘win-back’ email campaigns designed to go to at-risk users. They wanted to use Cortex to identify the users most at risk of churn. Then, they would send them win-back emails with titles most likely to re-engage them and prevent them from leaving.

The Cortex Solution

Using our self-service machine learning platform Cortex, our partner was able to quickly integrate machine learning into its business.

Firstly, our partner used Cortex Models to build models which predicted each user’s propensity to be retained in the platform over 30 days.

Cortex Predictions then built a cohort of users in the “danger zone”. These are users at risk of churning, but still impressionable enough to potentially come back. This prediction automatically ran every day to find users with 40-70% likelihood of churn. Our partner then automatically downloaded this list along with personalized content recommendations. They used these for weekly email newsletters, and these lists plugged directly into their CRM tools.

Together, Cortex Models and Cortex Predictions allowed our partner to build a system which automatically identified at-risk users and engage them with personalized outbound communication as soon as they entered the “danger zone”.

Results

Cortex ultimately delivered a 5%+ reduction in churn. Beyond direct increases in revenue, our partner also saw value from analyzing the behavioral profiles of loyal vs. at-risk users, providing key insights such as which types of content build user loyalty.

Cast Study 2: Transactional Video On Demand

Our partner in this case was a Transactional Video on Demand business, where users can rent or buy TV or film titles. For this business, if a user went 60 days without having rented or purchased a title, they were considered “churned”.

To keep users engaged, this business had a weekly email newsletter promoting a different set of titles. They wanted Cortex to identify the users most at risk of falling into inactivity, and personalize email engagement to those users, promoting titles that were most likely to encourage them to transact.

The Cortex Solution

Using our self-service machine learning platform Cortex, our partner was able to quickly integrate machine learning into its business.

Firstly, our partner used Cortex Models to build models. The models predicted each user’s propensity to make a purchase or rental within 60 days.

Next, Cortex Predictions built a cohort of users at risk of churning, but who could still be brought back to make a rental or purchase given the right offer. This prediction automatically ran every day. In doing so, found users with 50-80% likelihood of churn, and our partner was able to automatically download this list along with personalized content recommendations for weekly email newsletters directly into their CRM tools.

Results

Cortex increased the number of users making a transaction on their platform by over 20,000 unique users per month. Beyond direct increases in revenue from increased transactions and reduced rates of churn, our partner also saw value from analyzing the behavioral profiles of users who transact frequently vs. those that transact less frequently.

Looking to reduce churn for your business? Cortex lets anyone in your organization quickly and easily predict which of your users are at risk of becoming inactive, and develop personalized re-engagement strategies to win them back. Talk to us today about using Cortex for your retention marketing campaigns!

Want to know more? Contact us at info@vidora.com!

Want to Learn More?


Schedule a demo and talk to a product specialist about how Vidora’s machine learning pipelines can speed up your ML deployment and ultimately save you money.