As we look at the technological landscape nothing in recent times has been as impactful as the buzz created by ChatGPT just over a year ago.
The interest in advanced data and generative AI in the investment management community has been one of the most discussed topics at industry events. There are many use cases for this technology layer but thus far, outside the use case of ChatGPT in the retail world, very little has been developed or applied. Everyone is trying to do something revolutionary but real tangible progress still seems firmly stuck in thesis mode.
In this blog series, we will discuss the different requirements needed to realise the promise of generative AI. We will also conclude with how we use AI in the form of HUBGPT.
The first step, which is now easily achievable due to new technologies, focuses on speed – allowing data to be streamed in real time across your organisation and available when needed.
Breaking batch
We live in the world of instant everything, where cravings are satisfied with a few taps, and the world is at our fingertips. Fancy a meal? Your smartphone can summon any cuisine, arriving at your door in minutes. Want to watch a movie? Your smart TV allows you to see anything your heart desires.
Our reality thrives on the immediate, the now — except, it appears, when it ventures into the realm of data processing within investment management. There - batch is still king.
Investment managers continue to struggle to process and access data in real time
It is clear that investment managers need no convincing about the power of data, with 93% of asset managers planning to extend the use of analytics and insights.
But there are significant barriers to overcome first.
49% of asset managers struggle to access real-time, high-quality data, while 46% complain about the reliability, completeness, and freshness of the data they do manage to access.
In a world that thrives on immediacy, the irony lies in the inability to embrace real time data processing in one of the most critical sectors where the landscape often resembles a slower-paced marathon rather than a sprint.
An evolving stream of data and insight
The benefits of real-time data streaming are profound and the ability to harness its power is the key to driving greater business insight and efficiency, accelerating innovation, and protecting against risk.
After all, the right information is powerful. But the right information when it’s needed even more so. The quicker an investment manager can access and absorb data, the quicker they can act on it and deliver value to their clients.
With data constantly streaming and available to all permissioned to use it, across different business functions and use cases, each one is able to further enrich and in turn create richer data sets for downstream consumers.
For example, trade data as it is captured from Order/Execution management type systems needs to be enriched for accounting, confirmation, settlement, and regulatory reporting. Evolving this data through streaming with a state reflecting its processing through each step of the lifecycle makes it easier to meet all business and regulatory requirements.
If data is discoverable and available at differing stages of quality and maturity in real time it turns large enterprises into nimble entities allowing them to respond to challenges as they happen and act on opportunities as they arise.
Even the Securities Exchange Commission (SEC) is pushing for faster data processing. On 15 February 2023, it released a paper "Shortening the Securities Transaction Settlement Cycle," which states that "the Commission continues to believe, as it stated in the T+1 Proposing Release, that the transition to a T+1 settlement cycle can be a useful step in identifying potential paths to T+0 settlement ‘Implementation of T+0 now lowers compliance and operational risk in the future’”.
By replacing batch processing, firms can validate data the moment it comes in, ensuring that there are no errors and it’s able to move to downstream systems without friction. Exceptions are identified and dealt with immediately, so when trades close for the day, firms can send out their data to brokers, clients, regulators, and other stakeholders safe in the knowledge it is trustworthy.
Evolving your data platform to access new technologies
Embracing data streaming is step one and an important one in the journey to remove the barriers that have existed for decades within current systems.
HUB offers a streaming data platform that provides a single lens of well governed data and focuses on data from the point of ingestion all the way to maximising that data’s value by serving it in a fast, reliable, and in a decentralised way to a broad range of consumers.
In my next blog, I will outline further steps of applying data mesh practices and strong governance to your organisation which will make AI a step change rather than a transformation.
