Fidelity Investments is one of the world's largest providers of financial services, with assets under administration of $4.7 trillion, and it didn't get there by using yesterday's technology. In fact, the company has a standalone group, Fidelity Center for Applied Technology – more commonly referred to as Fidelity Labs -- whose job it is to take the long view, to examine technologies years before they become mainstream and help the firm get a leg up.
Recent examples of the Labs’ work includes the new Fidelity Watchapp for the Pebble smartwatch. Network World Editor in Chief John Dix recently caught up with Sean Belka, Senior Vice President and Director of the Fidelity Center for Applied Technology, to learn how the financial giant drives innovation.
What role does the Fidelity Labs play in the organization?
If you look at the company’s business and technology groups, they’re very focused on today’s products and customers and tend to have a one to three year roadmap. So Labs was created to have more of an external orientation and a different time horizon. We try to work on things that are three years out from mass adoption.
We have offices in San Francisco, North Carolina, Europe and in China, and in each of those offices there are people looking for new ideas from startups, academia, venture capitalists and big company labs. On the academic side, for example, we have a relationship with Berkeley and Stanford in San Francisco, MIT in Boston, and Trinity in Dublin. And on the startup scene we’re really looking for things that would be useful for Fidelity to bring in to research, pilot and potentially deploy.
But we get internal requests as well. So we’ll have a business unit come to us and say, “We’re trying to figure out cloud computing or big data, can you do some research? Can you connect us to the thought leaders in the space?”
Do people work specifically for this group or do they wear multiple hats?
We have about 75 people globally. A lot of times we partner with a business unit on a project and they will bring in developers, QA people, etc., so any project might have a combination of people working on it. Fidelity has about 12,500 technology associates and we spend more than $2.5 a year on technology across the enterprise, so having an R&D group focused on what’s new makes sense in our context.
What kinds of things do you examine?
Five years ago it was a lot of mobile and social, but now we’re spending time on things like wearable computing, artificial intelligence, and gamification of content. So our job is to put together an R&D portfolio and do the research, and oftentimes build a 1.0 version and then transition that to the appropriate business unit. So our job is always to be pushing the envelope in terms of what’s new and emerging, taking it to some level of tangibility. It’s not just research. It’s also development and then enabling the business units to do what they do best, which is get it into all the distribution channels and handle the integration with our businesses.
How is the group organized?
We have a number of different functions. We have technology researchers, and their job is to identify new and emerging technologies and bring them in-house, do research, and then recommend an action. We have development teams who are responsible for building early prototypes and pilots. We have people who are responsible for managing various, what we call, innovation ecosystems. For example, we have a woman who is responsible for San Francisco’s Silicon Valley, and a gentleman in Dublin who is responsible for Europe.
+ ALSO ON NETWORK WORLD Fidelity looks to virtualize, standardize data center resources +
And then we have some horizontal capabilities. So, as an example, we have a big practice with our partnership with Stanford University’s D.School around design thinking. So we hire people who either were students at the D.School or have a design thinking profile that may help our businesses by approaching a project and asking things like, “What is the real customer problem here?,” or “How can we create lots of prototypes to iterate with customers?” So they’re almost like customer researchers.
We also have folks who help scale innovations. So we have a patent office, as an example, with people who help fill out patent applications and decide where we want to protect our IP. And we have a group that runs what we call Idea Management. We use a platform to be able to run idea campaigns across the firm and externally. Their job is really to catalyze innovation throughout the rest of the firm’s 40,000 employees.
So it’s a range of skill sets.
How do you work with enterprise IT?
It depends. We have enterprise IT and we also have business unit IT. In business unit IT we have one group called Personal Investing, which is the thing most people know, like Fidelity.com or an Investor Center. We also have Workplace Investing, which is 401(k)s. And we have Fidelity Institutional, where we work with banks, brokers and insurance companies to be their back office and middle office. So in those cases we work directly with the technology and the business folks to build out pilots and applications. With the enterprise IT folks, we work on things for the broader firm. So, as an example, we did a pilot of Jive Clearspace six or seven years ago which turned into our social platform. We’ve done some work around video that turned into some of our video strategy. We’ve done some stuff in cloud computing using public and private cloud.
Can you give me some specific examples?
Take Google Glass. We initially were looking at that as a kind of consumer application. But then said, “Hmm, are there applications in the enterprise? Would it be useful if we could have one of our associates be able to watch what’s happening through the eyes of a remote contractor through Google Glass? We would try to take any emerging technology and say, “What are the use cases that really make sense here? Is there a consumer case, a B2B use case, an enterprise one? Sometimes there is, sometimes there isn’t. Sometimes we think there is and there isn’t.
When we get to the pilot stage we need to have a use case that makes sense. So we’ll look at something like digital currencies or mobile wallets, and it’s the job of the researcher to say at the end of the research phase whether they recommend doing a pilot. If so, what is it? And then prioritize a set of use cases or scenarios that they think might make sense.
We look at big themes and within those themes. So as an example, in the past few years mobile, social, cloud, big data, pervasive video have been really big themes. Now we’re adding on top of that wearable computing, artificial intelligence, including digital assistants and gamification of content to drive engagement. With the latter, for example, we might consider using a game-like interface to teach people how to do asset allocation, or on the enterprise side, use gamification for training modules. So we typically start with a team connected to a use case and that’s where the development agenda emanates from.
How about big questions about things like cloud and use of white box devices?
I also report to our CIO, as does the enterprise infrastructure group. All of the CIOs meet every Wednesday for three hours to coordinate agendas and talk about topics, and I lead some of the future-facing strategy discussions. So, as an example, we would talk about cloud computing. What’s our compute strategy moving forward? How do we see that? Or talk about storage, or DevOps as a model, or Platform-as-a-Service.
And then we typically set up a team that might have a CIO from enterprise infrastructure who could primarily sponsor that and we’ll put FCAT researchers on that team to help do the external research, to bring in companies to talk to, and then work with that team to come up with the future-facing strategy.
All the business units have to be part of that process, but the enterprise guys have huge day jobs. They’re running big development operations or big infrastructure operations so in some ways we supplement them with people who can spend 50 hours a week doing an architectural review and writing a white paper.
Can you talk about where you folks stand in terms of adopting technologies like cloud?
We have many things we’re always going to hold very near and dear to our hearts, and, as you might have seen, we just opened a new data center in Omaha, Nebraska, and I think we’re always going to have data centers. Always is a long time, but for the foreseeable future. But Fidelity Labs, as an example, is hosted in a public cloud. We have a hybrid strategy. We use public cloud where it makes sense and proprietary data centers where they make sense. Over time “where it makes sense” changes as those capabilities get better and security protocols get better. So it’s an evolving model for us.
How would you assess the speed of change on the technology front? Are things accelerating, or leveling off?
That’s a good question. In the late 90s you had everyone driving to the Web, which was a really big thing, and it was obviously transformational. At Fidelity today 97% of our trades take place online, and 90% of customer interactions. So it really was quite profound, and something that really changes the entire way services are delivered doesn’t come along every day.
Having said that, now there is more than one thing and they’re each pretty big. Mobile is huge. Social is huge. Cloud is big. Big data is big. Artificial intelligence I think will be big. So in some ways there’s more breadth, and we don’t feel any of those areas can be ignored. So it’s not like in the late 90s when everyone did a massive student body left out to the Web. Here it’s like -- What’s your mobile strategy? What’s your social strategy? What’s your cloud strategy? What’s your big data strategy, etc? So there’s a dynamicism today, and almost a complexity, because some of them are not discrete.
In other words, mobile gets better and better because services in it are cloud-enabled and socially complemented. So they’re discrete in one hand but they’re also converging. So I think it’s a little bit more complicated than it used to be and I think that makes people feel like the pace of change is faster because there’s so many different things happening.
There are some mathematical underpinnings to why progress is faster. I just finished Andy McAfee’s book “The Second Machine Age.” If you think about Moore’s Law calling for capacities doubling every two years, it went X to 2x to 4x to 8x, 16x. But now one doubling can actually have as much performance improvement as all that has preceded it.
And similarly, look at how many people are involved in innovation these days. In the early 90s if you wanted to do a startup you had to raise a lot of money, buy servers, have a physical location to put those servers, buy a sales force, do expensive brand building. Now if you have a great idea and a credit card you can be on Amazon in 10 minutes. You can have an app store as your distributor in 15 minutes. You can have social media as your amplifier. None of those things existed.
So it is easier for startups to come out of nowhere, which I think also drives people to feel the pace of change is a lot quicker because the barriers to entry are much lower. They’re much lower capital-wise, they’re much lower distribution-wise, and I think people haven’t really fully absorbed that yet.
You mentioned wearables a few times. A lot of the early stuff seems fairly goofy, but do you expect that to be a big thing?
Right now some of it seems a bit gratuitous and I think the jury is still out. I wear a Pebble watch and I find it useful, because I can see text and other information in a very convenient way just by looking at my wrist versus pulling my phone out of my pocket and entering my password in. So when I’m driving it’s more convenient. These things will succeed to the extent that they actually make people’s lives easier and better and they won’t if they don’t.
It’s hard to imagine any of them being as broadly accepted as a mobile phone or a laptop, but that’s what Fidelity Labs is here for. We’re here to do the experiments, to push the envelope on these things to see if we can create a better customer experience. And if the answer is no, that’s OK, it was an experiment. We learned something from it. And if the answer is yes, great, we’ll have a head start and we’ll just keep building on that.