Impact at a Glance
Higher accuracy predicting Customer Churn
Actionable insights into churn reason tied to churn stoppage
Greater predictability for customer success employees
Enhanced customer satisfaction due to early innervations
About the client
The client is a software as a service (SaaS) company that enables groups to share and collaborate information for board meetings. Their software is used by one million users and more than 700,000 board members across governance, risk, compliance, audit and ESG departments in various industries, globally.
Business Challenge
Being a SaaS company, the client faced similar challenges as by all global leading SaaS players- unpredictable customer behaviour leading to discontinued product usage, unexplainable product consumption trends across geographies and an impact on revenue as a result of it all. It is at this stage that Diligent began looking at turning things around on customer success side and got into partnership with Cognida.ai.
Cognida.ai proposed predicting customer churn with the help of its Zunō accelerator, which used AI to understand the product usage data and uncover patterns and give insights as to the reason behind why a customer can churn- all before a quarter of the predicted churn date.
Before adopting Cognida.ai’s solution, the client faced challenges which hindered their market growth and penetrability:
- Revenue loss due to unpredictable churn
- Low Customer Lifetime Value (CLV)
- High support and Retention costs
Customer Churn Prediction using Zunō.Predict
Cognida.ai proposed customer churn prediction solution by conduction various consulting led Discovery workshops with Diligent, to figure out where exactly the problem lies in the value chain and what’s the best way to solve it. Zunō.predict was used as the accelerator of choice which makes prediction simple and scalable. Zunō.predict, with it’s ability of ingest copious amounts of data and come up with actionable insights for preventing customer churn, became the corner stone for bringing predictability in what is an otherwise chaotic sales cycle.
Intuitive and powerful dashboards was front-ended to the entire AI layer which spoke to the customer success executives in a language that they could understand and design client interventions around. This shot up their productivity along with customer satisfaction many folds.
This was executed in 2 key phases:
- Building ML models: Product clusters were made to group data pertaining to similar product groups and analytics run on it to see the themes present. Zunō.predict’s AI was deployed as a top layer to enable predictions, which enabled identifying the reason behind churn for each product at a granular level: across products, geographies and BUs, giving executives control over preventing customer churn at least a quarter before it can happen
- Enhancing ML models: Once predications started flowing in, and executives were able to effectively act on them for a duration of 2-3 quarters, phase 2 was launched with an aim to understand how the real-world scenarios are shaping and to tweak the models to an higher levels of accuracy based on algorithm performance and model feedback. This resulted in fine tuning of all the ML models to push the accuracy of churn prediction by >18%- with resulted in deeper insights and targeted account management and greatly enhanced customer success executive’s productivity.
