How AI Transforms Dealer Engagement Using Data And Analytics

Introduction

In the fast-paced world of automotive finance, staying ahead of the competition requires innovation and efficiency. AutoFi, a leading player in the industry seeking innovation, embarked on a transformative journey to integrate cutting-edge technologies to enhance dealer engagement and satisfaction. This case study explores AutoFi's strategic implementation of AI-driven solutions and data orchestration, showcasing how proactive engagement revolutionized their approach.‍

Challenges Faced

AutoFi faced challenges managing its data from different sources like their CRM and support tickets, making it difficult to holistically understand dealer behavior. Proactive dealer engagement was essential, but existing systems lacked the ability to analyze and anticipate dealer needs before they became issues. AutoFi needed a way to gather all the data, understand it, and predict what dealers might need. The challenge was not just collecting the data but also making sense of it. Predicting issues, especially why dealers might leave, was essential to keep metrics like churn rate and retention at optimal levels. They needed a robust solution to understand the data and get useful insights that could be used by different business teams.

Solution Implemented

RapidCanvas implemented a solution to orchestrate AutoFi's data, centralizing data from varied sources, simplifying the process and making it ready for modeling. By utilizing intelligent data cleaning and feature selection and employing machine learning algorithms for predictive modeling, RapidCanvas leveraged existing data to build a customized solution for AutoFi's specific needs. The platform's transparency in modeling and its ability to deliver insights in a comprehensible manner were pivotal. AutoFi's team was provided with a 360-degree view of dealer behavior data, empowering AutoFi's business teams to make data-driven decisions effectively.

Feature selection 

The RapidCanvas team, collaborating with the AutoFi team, selected the most relevant features or data points. They also created new and enhanced ones from the existing data to train the churn predictive model. 

The model was trained using features such as: 

• Dealer data: Dealer tenure time, subscription tier and preferences
• Support tickets: Frequency and sentiment expressed in tickets 
• Platform usage metrics: Clicks, leads and inventory tracked daily‍

Predictive modeling

‍The churn model leverages a machine learning classification algorithm to analyse historical data and identify patterns related to churned dealers. The trained model is then used to predict the probability of future dealer churn, based on the learned patterns and potential churn factors. This provides valuable insights to the AutoFi team, both about the dealers at risk of churn as well as the specific reasons that are likely causing them to consider leaving.

Dynamic leaderboard

‍An interactive leaderboard generated on the RapidCanvas platform provided intuitive visualizations with insights into dealer behavior, with a particular focus on the dealers at higher risk of churn. 

Using a heuristic model, the RapidCanvas team created risk flags that indicate changes in dealer behavior that could be associated with a higher chance of churning. For example, metrics like usage of the platform compared to the average usage of all dealers or the business outcome i.e leads and submitted applications were closely tracked. If there was a fall in these, this was flagged as a potential risk factor. Recent support tickets with negative sentiment was another flag  

Combining these flags, a high churn probability was predicted by the model and the existence of and depicted on the leaderboard, ordered by the number of risk flags.

Continuous model monitoring

Since dealer behavior is dynamic and changes over time, the model is monitored and updated continuously with fresh data to ensure model accuracy remains optimal. This ensures that the model predictions stay effective in predicting dealer behavior and in supporting AutoFi’s successful relationships with its  dealer. 

Results and benefits

Reduction in dealer churn

With each dealer contributing $30,000 ARR on an average, retaining more dealers can save nearly $1.5MM in lost revenue opportunity

Increase in dealer satisfaction

‍The AutoFi team receives regular updates on at-risk accounts as well as the reasons for accounts being at-risk, helping to drive better dealer engagement and account management. The dashboards offered detailed notes on dealer account health and risk factors, equipping dealer engagement representatives for each account with granular insights ahead of each dealer interaction.
This personalized support based on predictive insights elevated AutoFi's engagement strategies.

Collaborative success

The insights from AI-driven dashboards were shared seamlessly across departments. Marketing and dealer engagement teams utilized the data for targeted strategies, amplifying the solution’s impact.

Explainable impact

With a transparent and explainable end-to-end data pipeline and a detailed breakdown of the different factors involved in categorizing dealers into various categories, the AutoFi team could easily understand the AI-led data analytics process and utilize the insights showcased on the dashboard.

Conclusion

RapidCanvas, with its innovative approach to data orchestration and machine learning, revolutionized AutoFi's operations. By simplifying complex data and delivering actionable insights, RapidCanvas became important for AutoFi's success in the competitive automotive finance landscape. The collaboration between AutoFi and RapidCanvas exemplifies how advanced technology can drive transformative change, making the future of automotive finance even more exciting and promising.