Kurtosys spoke recently with Matthew Oakeley, chief technology officer at Investec Asset Management about the dangers of believing technological hype, how regulation may be a necessary evil but it shouldn’t define your business as an asset manager and the opportunities currently being explored around robotic process automation (RPA).
What is the biggest tech challenge facing you at the moment at Investec?
Some of the biggest challenges facing any investment management firm belong to relate to what I call ‘internal digital’. It is not enough for IT teams just to supply trading or compliance systems anymore – people across the business want technology to help them access data services in a flexible way. They are increasingly used to the way we use our smartphones and access services as consumers outside of the workplace, yet that digital transformation has been quite slow in investment management firms.
When people tend to speak about ‘digital’ they mean external digital, such as websites and platforms but I’m interested in that internal digital journey. How do you find and access data and services, how people work flexibly and – as a result – how do you increase scalability?
Most investment managers don’t look particularly scalable – they tend to add headcount as they add assets. Investment management has always been very much a people business and I’m not sure technology has made a massive difference to that.
Why do you think that is, when the banking sector seems to be further down the path of using these new applications?
We’re a different kind of beast to banks. There’s a lot of hubris in the investment management industry about technology becoming the competitive advantage. But our job is to look after our clients’ money. Technology has its part to play, but we are not technology firms, we are investment firms.
We’ve got tools that help various parts of the business – managers, traders, middle and back office, client services and management – but a lot of the processes that enable the people ‘on the ground’ to access data and services are still relatively immature and I think we can do better.
Whether it’s about using data analysis to drive insights, or better positioning teams to answer client queries or just increasing the efficiency of operations, I can’t help thinking that this is an area where there is a lot of opportunity to make life better. If nothing else, my job as a technologist is to make people think Investec is a great place to work so the best people want to work there.
How are you approaching that challenge?
Strategically, our two big initiatives are the Data Hub Programme and the Digital Workplace Programme. Data Hub is about making sure that the left hand and the right hand of the firm join up: all the information about what we do with the money flows through, allowing us to craft the narrative for our clients, and join up as seamlessly as possible.
Digital Workplace is about mobility, access to information, tools people use and the training they receive, through to things like robotics, which we are starting to look at. They are not specific projects with milestones and deliverables but effectively vehicles to get everybody thinking about the topic.
On a day-to-day basis, you’ve got to try and focus on how to help people make use of what they’ve already got. People love implementing new systems and big vanity projects, but if we can use what we’ve already got better, we can get an awful lot more out of it, which is often about improving internal communication. So if you can find a way to increase the knowledge in a company, you can do an awful lot without spending millions of pounds on shiny toys.
Where has your department been spending money recently?
As well as investing in our data architecture, we’re investing in a strategic platform that supports our marketing content management. We’ve also been putting money into our investment quant teams, who are starting to play with more analytics tools. I don’t think that technology is just the preserve of the IT department anymore. They might be the ones to implement structures and processes but if you’ve got mathematicians and data scientists elsewhere in the business, it would be foolish to force IT to take over work they can do better themselves, which can be quite uncomfortable for the IT department to let go.
Before smartphones were so ubiquitous, when Nokia brought out their first phone with games on it, I’d bet it wasn’t the IT department who created any of that. Often you have to let people have tools to play and experiment.
With all these newer technologies we’re seeing touted around asset management – AI, blockchain, machine learning – how much is hype?
People choose to tell their stories for a reason and fall into the trap of believing their own PR. Blockchain is a good example. It could be an effective technology for creating new market infrastructure, although I don’t really think fund managers probably need to embrace it right now. But if someone could sell me a new product – based on blockchain – that helps deliver funds cheaper, faster and more effectively – I’d look at it.
Take something like robotic process automation (RPA), which we are looking at. The standard case for RPA is a process repeated over and over again at scale and volume. A classic example is a mortgage processing company – you get thousands of application forms, so many that you’ve outsourced to India for them just to be done manually because it was much cheaper. You could automate the single process of checking a form, bring it all back to Alabama and have Donald Trump tick the box that “America brought the jobs home” – just it would be done by a robot instead. I can see that business case: take something which you do thousands of, automate the process, and save loads of money.
But asset management isn’t like that. We have lots of little manual processes, rather than a few big manual processes. So is there actually a business case if you’ve never spent the money to automate it before? RPA is not free. You still have to pay someone to develop and maintain it. So, like lots of people, we’re having a look at it. Everyone is having a play, but I don’t think anyone is quite sure yet.
Is it all upside? Where are the risks you’ve identified?
With RPA, we recognise you could easily get into trouble if you are not careful. If you had a robot that could trade, and a robot that could settle, does that mean that whoever controls the robot can now do two roles not previously allowed to be done by the same person – someone could trade and settle?
If you had a robot that could open a bank account, a robot that could move money and a robot that could close the bank account, if one person controlled those robots they could steal all the money. So our approach on this is to treat these as if they are people – our robots are on the HR system, with a role and line managers. Employees can’t ‘double up’ – if they both need a robot they each have to request their own. We decided to treat them very much as digital employees from top to bottom.
So where are you testing the water?
We’re testing them in the operational team, the compliance team, HR are quite interested. Just for some of the processes used for moving data around. Wherever there are spreadsheets, you can usually do something with them.
Do you think the perceived impact is much bigger than the actual impact – good or bad?
There are things on the horizon in this industry such as AI, machine learning and data analytics with lots of potential uses. If we can come up with decent data analytics or AI tools for intelligent prospecting, the marketing team are really interested in that. It’s not just about knowing whom to call, but discovering when to call. If I have exactly the right contact but if I call them before the decision to allocate new assets that might be a lot more useful than phoning them two weeks later.
How have you been responding to the regulatory changes that we’ve seen in recent years?
These things drain everybody’s time, energy and resource. We’ve met our various deadlines, we’re compliant – it was all quite dull in the end. But our business isn’t defined by the regulatory topics – these are things you have to do as well as the rest of your business work. It’s not as though we launch a product and then afterwards we find regulation has rendered that product redundant. Regulations are a drag on cost but could bring about opportunity.
Can you offer an example where that change has opened up opportunity?
Look at investment research under MiFID. After a summer of everybody being confused about what the directive would mean about paying for research costs, almost the whole industry decided we would absorb the costs.
In one sense, this has improved value for the client. But it will also have far-reaching implications on the structure of the research world. What are people willing to pay for? How many will decide to seek alternative sources of research and not use the big brokers?
That disruption will force research brokers to become more competitive so a lot of the transparency brought to those markets will also change the way those markets behave.
Wherever there is change there is opportunity.