Here are key insights based on a Larry Kraus Tech-NEW-logy discussion moderated by Shaun Parkin, founder of Hall Road Investments, including Haik Sahakyan, co-Founder & CEO Arqa AI and Charlie Brown, sales director at Private Wealth Systems. You can access the recording and podcast here.
Key Insights:
Family office data pain points
- Challenges in selecting new technology: According to our audience, the top challenge when choosing new tech is migrating all the data. This is closely followed by the difficulty of getting everyone onboarded and trained. Another major hurdle is finding a system that integrates well with existing tools and meets the office’s needs.
- Biggest time-wasters in the current family office tech setup: Manual data entry stands out as the biggest time-waster by far.
Growing number of AI use-cases
- Automating Data Capture and Entry: AI handles tasks like capital calls, distributions, valuations, K1s, tax documents, and full custodial statements, reducing time and errors.
- Data Aggregation and Insights: AI aggregates and analyzes segmented data sets, providing real-time insights into portfolios and investments for informed decision-making.
- AI Chat for Portfolio Analytics: AI-powered chat allows users to query portfolios using natural language (e.g., “What’s my best-performing stock?” or “Which entities received large cash infusions?”).
- AI for Data Reconciliation: AI reconciles inconsistencies in data feeds from custodians, ensuring accurate financial reporting and correcting errors.
- Rebalancing and Trade Suggestions: AI monitors asset exposure, generates tax-efficient trade ideas, and facilitates automated portfolio rebalancing.
- Performance Analytics: AI automatically creates performance analytics and reports, eliminating the need for manual report generation.
- Document Translation and Processing: AI translates foreign language financial documents (e.g., Chinese) and processes them into the system for seamless integration.
- Mobile App Integration: AI-powered tools are integrated into mobile apps for real-time financial analysis and reporting on-the-go
- NOTE: Data security and privacy are crucial in the age of AI, especially when using large language models (LLMs). Many solutions expose sensitive data to public LLMs, but a better approach ensures privacy by keeping all data within a controlled system. This allows for leveraging AI technology without compromising personally identifiable information (PII), ensuring users maintain control of their data.
GIGO (Garbage-In-Garbage-Out)
- The quality of data aggregation and analytics is heavily reliant on the accuracy of the input data, which remains a major concern according to the panelists. PWS points out that custodial data feed errors can reach as high as 15-20%. Common issues include the duplication of corporate actions (e.g., stock splits) and discrepancies in recorded timing, securities mispricing, and partial transactions, all of which can distort account values, gains/losses, and risk/return analytics.
- Reconciling data from multiple custodians is necessary due to variations in reporting. While AI and automation help, some errors still require manual correction.