Personalized service and close relationships with customers are critical to the success of wealth managers and private banks. Often demoted as challenging to digitize, as personal connections require trust and human interaction to grow, client advisory significantly benefits from better and more efficient data leveraging artificial intelligence.
Wealth managers increasingly adopt AI, especially Generative AI, to provide advisors with up-to-date client information, summarize meetings, prepare portfolio briefings, and provide tailored investment advice. The banks and asset managers who embrace AI and modern data stacks don’t just upgrade their tech stack and processes today but gather critical insights and experience in AI adoption and customer feedback for tomorrow, providing a significant competitive edge towards those who haven’t started their AI journey (seriously) yet.
Adopting AI means more than registering a ChatGPT account. Banks must focus on managing their data, especially client data, to get the basics right first. Still, many organizations run siloed data systems, often with redundant and sometimes diverging information, which are not readily available for analytics, AI models, and business use cases, including personalized advisory. A solid, modern data stack and high-quality data are essential for any (semi-)automated back office, advisor, or client-facing digital service.
There are several key questions to address:
- How do we drive data culture and become more data-driven as a business?
Data Mesh has gained recognition as an approach to drive data culture, literacy, and availability across organizations without putting even more burden on the shoulders of IT and data analytics teams. In line with many digitization strategies and value-based product thinking, organizations applying data mesh embrace data sharing and data quality as fundamental for data products that create value for their organization and beyond.
Ownership is a key component – those responsible for a business process also claim responsibility for the data and derivatives they create. Then they offer such data as products to the organization while maintaining high quality, good documentation, and real useability.
Self-service is another key aspect – advisors, relationship managers, and marketers who are self-enabled to access the most recent data ad-hoc without needing to request and review BI reports are more likely to benefit from the information available.
From a more technical point-of-view, a modern data stack is critical to becoming a data-driven organization. A data mesh or a new generation of cloud-native data systems powering data warehouses, data lakes, and data lakehouses provide a solid, future-proof foundation to capture, process, analyze, and use data. (You can find some leading vendors here: cid.com/partners)
- How do you create trust in data, AI-based or AI-enriched results?
Data quality is critical for trust. Transparency is essential to inform data users what level of quality they may expect when interacting with data sources or results from AI.
Some organizations opt for a three-tier classification (e.g., bronze, silver, gold) or a scorecard that considers dimensions such as source (internal or external), level of vetting, or processing or aggregation steps taken with a potential impact on the quality of results. It enables users to make an informed decision about how to use the data and ensure proper documentation for downstream use.
Transparency about what type of AI is involved is important, too. While automated classification based on Machine Learning is probabilistic, it can still leverage proprietary data and training in a controlled environment. Organizations might want to limit the use of Generative AI to particular, controlled-environment applications, leverage customization capabilities such as RAG (retrieval augmented generation) to increase relevance and reduce erroneous output, or simply flag results for the user to validate as they could be inconsistent or include hallucinations.
Private AI and private cloud are essential frameworks to maintain data security, critical to ensuring clients’ and advisors’ data is processed rightly using trustworthy proprietary information.
- How can we create value for relationship management and business development and increase user acceptance?
Relationship managers, advisors, and business development professionals must already cope with abundant information across various sources. Instead of adding even more, AI-powered solutions help streamline information processing and create value by simplifying access to actionable intelligence.
Aggregating data such as (U)HNWI dossiers, household data, company information, internal lead databases, and proprietary CRM data is a critical driver of adoption and user acceptance for intelligence tools in relationship management. When advisors no longer need to log in to several systems and care about mapping and joining the dots, information becomes much more useful and easier to benefit from.
Artificial intelligence, powered by a modern data stack, plays a significant role in aggregating data:
- It blends information about individuals and organizations across disjoint data sources.
- It adds timely insights from news, social media, and more by automatically filtering, tagging, and mapping it.
- Generative AI can summarize and present actionable information seamlessly and integrate it into CRM workflows, the ultimate game changer.
This approach supports business applications such as:
- Client monitoring enables advisors to provide more personalized and proactive advice (Read more)
- Prospecting intelligence to win new business and tap into new markets efficiently (Read more)
- Support investment decisions leveraging qualitative data turned into structured, useable insights (Read more)
Author © 2024: Alexander Stumpfegger – www.linkedin.com/in/alexander-stumpfegger
