What’s Driving Asset Management Technology Trends?

Asset management technology is evolving at breakneck speed. While automation, data workflows, and analytics have always been crucial, technology is now a powerful differentiator for clients and employees.

Clients demand real-time access to more detailed investment information, moving beyond traditional month-end reports to interactive self-service. Among younger staff, there is an expectation that everyone can write code; they seek an innovative environment where coding skills add value without impacting broader operational processes.

Keep your eye on these four critical asset management technology trends:

1. Public cloud: Over the last few years, public cloud adoption has shifted from a niche concept to a widely accepted inevitability, though the price is still high. Furthermore, cloud capabilities are now recognized as foundational for firms seeking to deploy AI solutions.

2. AI: Asset managers now leverage deep learning and, more recently, Generative AI (GenAI) for a range of tasks with use cases continuing to expand, including wealth advisor efficiency and recommendation, investment compliance, and ESG analysis, to name just a few.

3. Distributed Ledger Technology/ DLT: While DLT has been frequently criticized as a solution looking for a problem, the adoption of tokenized assets shows promise in select areas.

4. CI/CD and APIs: Continuous integration and deployment (CI/CD) practices are crucial for keeping pace with industry change. APIs are pivotal in seamless integration and enable asset managers to innovate without disrupting core workflows.

Most asset managers have experimented with deploying their technology to the cloud, but very few have fully embraced adoption. Accenture’s report reveals only 8% of asset managers have successfully completed their cloud migration. Despite crossing barriers to entry, such as basic information security and compliance, significant work remains to transition their daily operations technology stack applications to the public cloud. This shift would enable them to reap the well-documented benefits of scalability and redundancy.

While market fears and basic barriers to public cloud have been reduced, other obstacles have evolved.

• Security faces heightened scrutiny as increasingly sensitive data is migrated to the public cloud.
• Persistently high costs are under the spotlight since taking full advantage of the public cloud’s scalability (and hence cost savings) requires software development work. These foundational tasks can include transitioning from a traditional database into a cloud database service or rearchitecting an application into microservices for scalability.
• Operational Transformation: Moving to the cloud isn’t just about technology—it’s a cultural shift. Asset managers must rethink processes, workflows, and organizational structures. Embracing agility and automation is essential.

When machine learning garnered significant interest a few years ago, and more recently, Generative AI captured attention, asset managers were excited , ‘What can GenAI enable?’ Answering this question has proved challenging. It requires finding the right blend of machine learning expertise, business, data, and process knowledge—a task more complex than initially anticipated.

At Linedata, our exploration of AI has expanded significantly as we have built a business delivering practical and bespoke AI solutions for asset managers. We swiftly increased the number of operational use cases, successfully completing proof-of-concepts (PoCs), advancing in over 30 AI scenarios and bringing solutions to production. These encompass a wide range of front-to-back use cases, from automatically interpreting compliance rules to handling KYC (Know Your Customer) and AML (Anti-Money Laundering) processes, as well as providing intelligent suggestions for managers' ‘next actions’.

Now that we have brought relevant use cases to market, Linedata’s challenge is to produce them at scale so that all our clients can benefit. This includes deploying required AI infrastructure, such as appropriate retraining to ensure models are kept current and ‘guard rails’ to assist with model errors.

The big unknown for GenAI is how much better large language models will become over the next few years. If they continue to improve, current solutions that optimize user performance while relying on user oversight may eventually reach full automation. One advantage will be that as models do improve, they can be easily swapped out and adopted in our use cases at Linedata. This means that when there are material improvements in the core models, they can be instantly recognized and implemented in our use cases.