BlackRock's acquisition of a Silicon Valley-based robo adviser last August provides two distinct lessons for asset managers.
The first is about a willingness to accept limits: despite being the world's largest asset manager, BlackRock was ready to acknowledge it found a technology platform better than what it was developing.
Bringing it into the fold has quickly paid off, as BlackRock has struck deals to provide digital advice platforms for several large financial institutions, most recently LPL.
The second lesson is about the importance of technology: the acquired platform will aid BlackRock to further develop uses of data to understand and better serve institutional needs in the future.
Jon Xu, chief technology officer and co-founder of FutureAdvisor, says the financial industry has only begun to tap the data available to it, and that as firms manage and harvest data better, it will be tapped to make the online experience for investors deeply personal.
"We can already read data about people's incoming streams of money, and we can say, 'Oh, they took a sabbatical,' or, 'they're going back to business school,'" Xu says. "That type of hyperpersonalization is really about being able to build something that is really for the user. This reactive notion when you're managing money is a responsibility. People can't generally wake up in the middle of the night and do this analysis, every night. But a machine can do that and take all those data points to do that process better."
The industry need to get smarter about how it measures investor tolerance, Xu adds, and utilizing data to refine that process is one way.
But Xu is quick to caution that technology won't immediately make a provider better: "We as an industry need to evolve our thinking on this and not just put a slicker interface in front of a better survey, so to speak."
There's a lot of attention being paid to measuring client risk tolerance.
I think today, how people gauge risk for the user is a bit obsolete. There's so much data about the customer, about their behavior, about their existing finances, that you should be able to build a good risk profile off a lot of that data. Now, you may have to, and it's on us to be able to allow them to both understand what that risk profile is and modify that if they wish. But really, the smart default is something critical here. People don't know how to answer these questions; and they don't know what the result of these questions necessarily leads to, in terms of risk.
Even as you are able to use predictive models to help them understand their risk models, the hard part will be actually explaining it to them. That's ultimately where these solutions have to do a great job in order to get the engagement and have the right, suitable financial management for their situation.
Apps like TurboTax are helpful, but as you can aggregate a lot of that data, that explanation is going to be more important. As you manage client investments, it's even more critical, because you need to be reactive to people's situations. To not just be able to say, what is their tolerance for risk, but what is their capacity for risk? What is their ability to not just stomach risk, but to take that risk over time?
Being situational is one aspect of it, and building that custom profile based on all the data that we have. It remains to be seen how all that data translates - social media feeds, LinkedIn profiles -how that is going to result and feed into a particular model. But the industry is going there. There's an amount of data that we're not leveraging as an industry to better serve the customer.
When does the questionnaire model for risk determination and asset allocation become obsolete?
I'd argue that it's obsolete now. It's just that nobody has developed a solution that can replace that. At FutureAdvisor we have gone away from large surveys. Obviously we can still adapt to the needs of both the financial institutions and the end users, but this notion of having smart defaults based on a set of data is already our inherent approach. We're just going have to incorporate more data into that.
Regulators have recently raised doubts about robo advice partly on their scrutiny of how many questions a platform asks users about risk.
This is just a personal opinion, but I think asking more questions or a specific number of questions isn't necessarily going to make the suitable and fiduciary responsibilities get carried out any better. Even in the asking questions model, it's about what's relevant. How do you make people understand the risk that they're taking? That's why we spend a lot of time painting the end outcome as much as we can for customers, knowing that there's variability and volatility. Asking the customers a set of questions that they don't understand is ultimately not going to produce a better end result.
I do think we as an industry need to evolve our thinking on this and not just put a slicker interface in front of a better survey, so to speak. The way that you measure how well you are serving a customer can't just be how many questions you ask them.
What is the role for human advisors when data can already tell us so much about our behavior, and that insight is expected to become more sophisticated in time?
Fundamentally, there's both a trust-building exercise in this process of understanding one's own financial picture and a teachable moments exercise. People will always want to reach out to other humans to learn. There's a lot of value that people can add.
I would argue that waking up in the middle of the night to do a rebalance is not human-added value. But having that ability to explain and to offer that end-client experience as a human is still a very great connection.
At the end of the day, wealth advice is personal. That's why I think it's more nuanced than just saying, that younger people will always want this, and older people will always want this. It is going to be based on convenience; it's going to be based on how well the human-added value is on top of whether you can deliver me clarity.