The buy-versus-build debate for OEMS platforms was a key discussion at TradeTech, with a panel unanimously agreeing that buying off-the-shelf products, in which certain functionalities can be built on top of that, is the most appropriate approach.
Oskar Wantola, Man Group
Buying was viewed as a better option if available particularly thanks to the speed in which you can implement in comparison to when building. Various panellists agreed that building should be secondary to buying, and used only if specific needs are to be met.
A similar view was shared by Lynn Challenger, global head of trading and order generation at UBS asset management, in a separate panel this week.
When asked to assess how to ensure future resilience in an ever-changing market, he suggested buying before building. “You don’t need to spend years and a lot of money trying to build […] these technologies exist already, they’re out there. You’re better off buying something,” he said.
Read more: ‘Buy before build’ says UBS AM’s global trading head Lynn Challenger
“Ideally what I would like from multi-asset OEMS is to have the best functionalities. You have specific functionalities for different asset classes, but you also want to have unified core functionalities, so it’s easy for traders to trade multi asset classes,” said Oskar Wantola, head of listed execution technology at Man Group.
“You would like to have separate modules and from those modules you can pick whatever is the best, either build it internally or buy it if there’s one off-the-shelf and connect all those modules through interoperability functionalities.”
Looking at the continued evolution of OEMS platform, Andrew Kovacs, director of product, EMEA, at Charles River Development, highlighted that the firm wants clients to lead the discussion to ensure OEMS platform cater specifically to their needs, adding that any good provider should be able to offer interoperability.
“When I hear about traders doing things like coding in Python that is music to my ears because I feel like that’s the future direction for a lot of systems,” said Kovacs.
“So being able to have a really sophisticated UI is wonderful, but ultimately, having a system that is completely flexible and open to programmatic interaction is going to be really important for the future.”
Giving a buy-side perspective on the expectations of EMS platforms, Martin Hendry, deputy head of trading at Liontrust Asset Management, noted that accessing improved efficiencies for the trading team is a key area of focus.
“We’re looking to automate. We’re looking to get better data. We’re looking for the best connectivity with our brokers so we can enhance our performance through that constant feedback loop,” said Martin Hendry, deputy head of trading at Liontrust Asset Management.
Hendry highlighted that automation has been a journey firms have been on for years, with the shift in automation today more focused on artificial intelligence and better data driving those decisions.
“We’re also looking at automation across asset classes, so rules we’ve been using for years in equities, we can apply to FX and we can apply to elements of fixed income as well – creating those efficiencies. And if traders can input those rules, build those different functionalities within the system, then we don’t need EMS providers to build things for us and we can do all that ourselves in house – those elements are really key for us,” added Hendry.
The issue of data quality – which has been discussed in various panels across the conference this week – was reiterated within this panel as a key challenge that has to be addressed in order to meet growing OEMS requirements.
Equities were noted as having a good dataset by panellists, however, it was emphasised that it is difficult to build rules for automation in other asset classes. The foundation of a good OEMS is good data, various panellists concluded.