Prevented Churn: $2.1M AUM Saved with AI Prediction

Prevented Churn: $2.1M AUM Saved with AI Prediction

Executive Summary

Vanguard Point Advisors, a growing RIA firm, faced unexpected client attrition that threatened their projected growth targets and revenue streams. Golden Door Asset deployed a sophisticated AI-powered churn prediction model, analyzing client behavior and engagement to identify at-risk clients. By proactively addressing client concerns and strengthening relationships, Thomas Adeyemi and his team prevented the loss of $2.1 million in AUM and significantly improved client retention rates. This case study highlights the power of AI in proactively managing client relationships and safeguarding assets under management.

The Challenge

Vanguard Point Advisors, managed by Thomas Adeyemi, had established a solid reputation for personalized financial planning and investment management. However, in the second quarter of 2023, they experienced an unexpected surge in client attrition. Initially dismissed as normal market fluctuations, the trend continued into the third quarter, raising serious concerns.

Specifically, the firm observed a 3.5% increase in client churn compared to the same period in the previous year. This translated to the loss of approximately $3 million in Assets Under Management (AUM). While new client acquisitions helped offset some of this loss, the firm recognized the critical need to address the underlying causes of the attrition.

Further analysis revealed that a significant portion of the churn was concentrated among clients with between $500,000 and $1 million in assets, representing a core segment of their client base. These clients, generally within 5-10 years of retirement, cited concerns about market volatility, rising inflation, and a perceived lack of personalized communication as reasons for transferring their assets to other firms.

The financial impact was substantial. Losing $3 million in AUM resulted in a projected $30,000 decrease in annual revenue (assuming a standard 1% management fee). More importantly, the unexpected churn strained resources, diverting attention from proactive growth strategies and negatively impacting team morale. Thomas Adeyemi realized they needed a proactive solution to identify at-risk clients and address their concerns before they decided to leave. Simply reacting to client departures was no longer a sustainable strategy.

The Approach

Recognizing the limitations of traditional reactive approaches, Vanguard Point Advisors partnered with Golden Door Asset to implement an AI-powered churn prediction model. The strategic decision framework centered around three key pillars:

1. Data Integration and Analysis: Golden Door Asset worked closely with Vanguard Point Advisors to integrate various data sources into a unified platform. This included client demographic data, investment portfolio information, historical transaction data, communication logs (emails, phone calls, meeting notes), and even external market data. The goal was to create a comprehensive 360-degree view of each client.

2. AI-Powered Churn Prediction: Leveraging machine learning algorithms, Golden Door Asset developed a customized churn prediction model tailored to Vanguard Point Advisors’ specific client profile and business context. The model analyzed hundreds of variables to identify patterns and predict the likelihood of individual clients churning within the next 6 months. Key indicators included declining engagement with the firm’s communications, reduced trading activity, withdrawals exceeding a certain percentage of the portfolio, and negative sentiment expressed in client feedback surveys.

3. Proactive Intervention and Relationship Management: The AI model provided a prioritized list of at-risk clients, enabling Thomas Adeyemi and his team to proactively engage with these individuals. This involved personalized outreach, addressing their specific concerns, offering tailored financial planning advice, and reinforcing the value proposition of Vanguard Point Advisors. The approach was not simply about retaining clients at all costs, but rather about ensuring they received the best possible service and felt confident in their financial future. For example, if the model indicated a client was concerned about market volatility, the advisor would schedule a call to discuss hedging strategies or adjust the portfolio to a more conservative allocation.

The strategic shift was from a reactive to a proactive relationship management approach, powered by AI-driven insights. This allowed Vanguard Point Advisors to anticipate and address client needs before they escalated into attrition.

Technical Implementation

The churn prediction model was built using Python and several key machine learning libraries, including scikit-learn, pandas, and TensorFlow. The technical implementation involved several stages:

1. Data Preprocessing and Feature Engineering: Raw client data was cleaned, transformed, and preprocessed to ensure compatibility with the machine learning algorithms. This involved handling missing values, normalizing data ranges, and creating new features from existing data. For example, a "Portfolio Volatility Score" was calculated based on the standard deviation of daily portfolio returns over the past year. Other key features included:

   • Communication Frequency: Number of emails, phone calls, and meetings with the advisor over the past 90 days.
   • Website Engagement: Time spent on the firm's website and specific pages visited (e.g., market updates, investment guides).
   • Withdrawal Ratio: Percentage of the portfolio withdrawn over the past 12 months.
   • Net Promoter Score (NPS): Client satisfaction score based on periodic surveys.

2. Model Selection and Training: Several machine learning algorithms were evaluated, including logistic regression, support vector machines (SVM), and random forests. A random forest model was ultimately selected due to its high accuracy and ability to handle a large number of features. The model was trained on a historical dataset of client data, with 80% of the data used for training and 20% used for validation. Hyperparameter tuning was performed using cross-validation to optimize the model's performance.

3. Integration with Existing Infrastructure: The trained churn prediction model was integrated with Vanguard Point Advisors’ existing CRM system using a REST API. This allowed the model to automatically score clients on a regular basis and provide alerts to advisors when a client was identified as being at high risk of churn. The integration was designed to be seamless and require minimal manual intervention.

4. Model Monitoring and Maintenance: The performance of the churn prediction model was continuously monitored to ensure its accuracy and effectiveness. Key metrics included precision, recall, and F1-score. The model was retrained periodically using updated client data to maintain its accuracy and adapt to changing market conditions.

The entire process adhered to strict data privacy and security protocols to protect client information. All data was encrypted both in transit and at rest.

Results & ROI

The implementation of the AI-powered churn prediction model yielded significant results for Vanguard Point Advisors:

• Improved Churn Prediction Accuracy: The model correctly identified 78% of clients who were likely to churn within the next 6 months. This allowed Thomas Adeyemi and his team to focus their efforts on the clients who were most at risk.
• Saved AUM: Proactive intervention based on the model's predictions prevented the loss of $2.1 million in AUM. This represents a significant return on investment, given the cost of implementing and maintaining the AI model.
• Reduced Overall Churn Rate: The overall churn rate decreased from 3.5% to 1.8% within the first year of implementation, bringing it well below the industry average.
• Increased Client Satisfaction: By proactively addressing client concerns, Vanguard Point Advisors improved client satisfaction scores and strengthened relationships. The firm saw a 15% increase in their average Net Promoter Score (NPS).
• Improved Advisor Efficiency: By prioritizing outreach to at-risk clients, advisors were able to focus their time and resources on the most important relationships, leading to increased efficiency and productivity. They reported spending 20% less time on reactive problem-solving.

The financial ROI was substantial. Preserving $2.1 million in AUM translates to an additional $21,000 in annual revenue (assuming a 1% management fee). This revenue stream, coupled with the reduced costs associated with client acquisition, made a compelling case for the value of AI-powered client retention.

Key Takeaways

1. Proactive Retention is Key: Don't wait for clients to churn. Use data-driven insights to identify at-risk clients and proactively address their concerns.
2. Data is Your Asset: Leverage all available client data to create a comprehensive understanding of their needs and preferences. Integrate disparate data sources for a complete picture.
3. AI is a Powerful Tool: AI can automate the process of identifying at-risk clients, freeing up advisors to focus on building relationships and providing personalized service.
4. Personalization Matters: Tailor your communication and advice to the specific needs and concerns of each client. Avoid a one-size-fits-all approach.
5. Continuous Monitoring is Essential: Regularly monitor the performance of your churn prediction model and retrain it as needed to maintain its accuracy and effectiveness. Market dynamics change, so your model must adapt.

About Golden Door Asset

Golden Door Asset builds AI-powered intelligence tools for RIAs. Our platform helps advisors proactively identify opportunities, mitigate risks, and deliver a superior client experience. Visit our tools to see how we can help your practice.