The data management challenge
What lies beneath: the importance of a robust data model to capture, store, and then feed the client portal.
After discussing the current status of client portals in the British wealth management industry we covered more backend-related topics. We wanted to understand the complexity of databases and data models of firms in order to better understand how new portals can be built upon these legacy systems.
Key points:
- The devil is usually in the details regarding a good data model. Wealth managers need to pull in data from disparate vendors and systems, and an integration layer to deal with different data types and structures.
- Most, if not all, wealth managers will need professional advice in some form to make robust data management a reality.
Being able to serve up data to a client portal relies on having a smooth backend architecture capable of taking in, storing, and then sending on data and information in a clean and readable format. What complicates things is that wealth managers need to pull in data from disparate vendors and systems, and the end result can be fragmented and patchy if the process is not carefully managed.
Wealth managers need an integration layer to deal with different data types and structures in order to bring together a meaningful offering. This would still retain the integrity of the original data source and present it to the client portal.
Research found
- For most of the respondents, suitability information and portfolio performance metrics are the most difficult data to link to a client portal.
- Almost half of the databases connect/ would like to connect to over four systems.
- Respondents perceive portfolio look through information and performance data to be the most significant aspects for display on their portal. They also value having news and insights pages, and notifications about events.
- The main challenges of client portal creation are the complexity of data that is scattered across numerous systems and data quality issues.
Data storage
Data storage is an essential process in data management. All types of data need to be pulled in, cached, and stored to be ready for the portal to access. This creates a more secure backend, but it also creates complexity because a layer to integrate and combine data is required.
For example, there might be different systems looking at discretionary and model portfolios. Still, there is a need to combine all that data and have an aggregated view too. An integration layer is very valuable here so that it can deal with each of the backend systems and take in data to aggregate it, which then can be used as a base to model and see how things look.
Caching does come with a few issues, however. In particular, complicating the processes of tweaking data, so sometimes more flexibility is required. In practice, this means getting to the data after it comes from the API, but before it gets to the warehouse, to ensure that the data and other content are optimised in terms of accuracy and wording.
The need for real-time information is pressing, too. So, within the data management process, data needs to be sorted according to whether it should be available in real-time or at the end of the day. Document sharing is one thing that comes up repeatedly as a real-time requirement.
While real-time availability sounds like the best solution in general, its resource intense nature is cause for caution. Also, most systems collect data at the end of the day, so real-time availability could be unnecessary if no other relevant back-office systems support it.
A focus on the data architecture to make sure firms can take the data and leverage insights more systematically is now largely underway. Wealth managers need not only to serve up data, but also to know how and when clients are looking at it and take note of their engagement level.
Making this a reality
Most, if not all, wealth managers will need professional advice in some form to make robust data management a reality.
Being able to make sense of user behaviour is key, too. Knowing how clients use a portal, mobile or desktop, how long they spend on it, and where they hover, can inform the portal’s future direction.
This is as much about behaviours as it is about data. Indeed, pushing information at clients is about so much more than just showing them their investments. It is about knowing how and when to approach and with what. Often, it is not around valuations or performance, it is around insights and news, about being informed about markets, what is going on, and how all that impacts over the medium and long term, etc.
Collaboration with a vendor is obviously a bonus in this regard, and firms are leaning more towards this model. Having some sort of in-house knowledge is also an intelligent play if firms are looking to understand and glean practical and strategic insight from their data.
Data science is also well regarded. Here, the win is that the scientists bring basic level expertise, which can then be moulded to the business.
Indeed, the possibility of someone who sits in between the business and digital innovation, who can translate and explain how data works and its benefits to those making the strategic decisions, can only be a good thing.
Data science is something that, conceptually, has come a long way in the wealth management industry within a relatively short time. In addition, wealth management can learn from other sectors.
Some of the first digital offerings brought in people from the gaming industry resulting in their frontends becoming user-friendly, intuitive, and thus, engaging.
All of this is essential if wealth managers are to drive up portal use. Realistically, no client will choose a firm just based on a portal. However, not having a well-thought-out portal in terms of data model or UX could mean being filtered out of the selection process. Good data management and a good frontend are thus vital.
