Sign up to save your podcasts
Or
Share How Featurespace Is Helping Fight Fraud
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
How Featurespace Is Helping Fight Fraud
This episode was recorded at the Money 20/20 event in 2019. On this episode, PaymentsJournal’s editor-in-chief, Ryan McEndarfer, sat down with Dave Excell, founder of Featurespace.
PaymentsJournal:
Dave, thank you so much for joining me on today’s episode.
To start off, can you give us a little background on Featurespace and how you
help prevent fraud for your financial institution clients?
Dave Excell:
Great. Well, thank you for having me on the show today. I
started Featurespace while I was studying at the University of Cambridge over
in the UK, and was really fascinated in the application of statistics to
understand behavior in the context of how people interact and behave in
different circumstances. We’ve used those same ideas, thoughts and research in
how we help our financial institutions prevent fraud. That’s mainly around
building up a unique, distinctive profile, which helps us understand what good
customer behavior looks like. We use those profiles to look at anomalous
activities, or changes in behavior, that are suspicious then we then use in
algorithms to detect and prevent fraud in real time. Then, importantly, we are
able to take feedback into the system so that it continues to learn and evolve
new data sources or information being fed back into the platform to make sure that
the performance of the system is optimized.
PaymentsJournal:
Excellent, thank you for that overview there. So, I’d like
to kind of get into the topic of money laundering here. Money laundering is
often thought of as a separate form of another type of fraud. Why do you think
money laundering in particular gets that different “bucket” from the
public’s viewpoint, if you will?
Excell:
One of the things that we frequently talk about is whether it is fraud is and there’s often money laundering afterwards. Often when we think about those two concepts, fraud is the activity of essentially stealing something or taking money from someone else, like an instance of credit card fraud where maybe a fraudster has acquired stolen card details from the dark web, then using those details to purchase something. They can then sell the item that they’ve purchased for cash, so they end up with a pile of dirty money that they then need to transfer into a good source of funds. This is where money laundering comes in. So, we often see money laundering in the act of taking those proceeds of crime and trying to convert them into a sort of legitimate currency that fraudsters can use in their day-to-day lifestyles and activities.
PaymentsJournal:
Interesting! So money laundering itself is kind of unique,
in that it can kind of be seen as post-fraudulent activity. Do you think that
money laundering can be prevented in the same way as other types of fraud?
Excell:
Definitely. I think the way we built our platform enables us
to really understand what good and legitimate activity looks like by customers
of financial institutions. We can use those same profiles to look at specific
types of behaviors that are indicative of money laundering. One of the challenges,
though, is that with fraud, we often get very good determinations – or in the
machine learning concepts, labels – that define when fraud has taken place.
Whereas when we look at any money laundering papers, those are usually referred
into suspicious activity reports, into the regulator. So, getting the
definitive confirmation that money laundering has taken place is not as
frequent as what we see in fraud scenarios.
PaymentsJournal:
Interesting. Now if we could, I’d like to get down to brass
tacks here. I think you alluded to this a bit in your previous answer, but I’d
like to flush it out a bit more. What are some key components that must be
included in technology used to fight both fraud and money laundering?
Excell:
One of the key elements is around data sources and being
able to pull together a good picture of what a customer or business at that
financial institution is doing so you have a well-filled profile and
understanding. One of the key things that we’ve done at Featurespace is to be
able to do that at an enterprise level. Rather than looking at the activity of
a customer or commercial entity, when they’re doing one particular type of
payment, whether it’s a card payment, ACH, wire, or check, today those are
typically sort of monitored in independent solutions. So one of the key things
that we’ve done is pull all of that together into a centralized enterprise
system to have a complete view of what the customer is doing. Outside of that
is not just looking at the movement of money, but also at how they are
interacting with the financial institution, how they’re looking at digital
activity in terms of behavior on-site or on the mobile device, and how they’re
potentially contacting the call center. This gives us much richer context in
terms of understanding how that person is interacting with the bank, which
gives us additional signal to know if it’s criminal or if it’s legitimate
activity from a genuine customer or business.
PaymentsJournal:
Great. Now, you brought up the centralized system that
Featurespace has. What are the benefits of a financial institution using just
one provider to fight both fraud and money laundering?
Excell:
One of the key benefits is to be able to have that consolidated view of a customer and enable one place where you have your financial strategies rather than needing to go through and optimize different systems. When there are different systems in place, the gaps or weaknesses between the systems is often what criminals try to exploit – where that data isn’t carried over. They try to get between the cracks of those systems, essentially get their feet in the door there, and then continue to pry or open once they’ve established that little crack. Joining all those systems together, leverage is also reduced as the potential entry points for the fraudsters and criminals to be able to access. Ultimately, it will help in the fight against crime, but also in enabling genuine customer activities. By having a picture of what the consumer does, and focusing on knowing when we are seeing good legitimate activity, we can ensure that those transactions and interactions continue without introducing more friction to the customer journey.
PaymentsJournal:
Well, Dave, thank you so much for speaking to me today about
Featurespace and the intersection of fraud and AML. I hope to have you back on
the podcast soon.
Excell:
Ryan, it was great to be on. I look forward to next opportunity as well.
The post How Featurespace Is Helping Fight Fraud appeared first on PaymentsJournal.
View all episodes
By The PaymentsJournal PodcastThis episode was recorded at the Money 20/20 event in 2019. On this episode, PaymentsJournal’s editor-in-chief, Ryan McEndarfer, sat down with Dave Excell, founder of Featurespace.
PaymentsJournal:
Dave, thank you so much for joining me on today’s episode.
To start off, can you give us a little background on Featurespace and how you
help prevent fraud for your financial institution clients?
Dave Excell:
Great. Well, thank you for having me on the show today. I
started Featurespace while I was studying at the University of Cambridge over
in the UK, and was really fascinated in the application of statistics to
understand behavior in the context of how people interact and behave in
different circumstances. We’ve used those same ideas, thoughts and research in
how we help our financial institutions prevent fraud. That’s mainly around
building up a unique, distinctive profile, which helps us understand what good
customer behavior looks like. We use those profiles to look at anomalous
activities, or changes in behavior, that are suspicious then we then use in
algorithms to detect and prevent fraud in real time. Then, importantly, we are
able to take feedback into the system so that it continues to learn and evolve
new data sources or information being fed back into the platform to make sure that
the performance of the system is optimized.
PaymentsJournal:
Excellent, thank you for that overview there. So, I’d like
to kind of get into the topic of money laundering here. Money laundering is
often thought of as a separate form of another type of fraud. Why do you think
money laundering in particular gets that different “bucket” from the
public’s viewpoint, if you will?
Excell:
One of the things that we frequently talk about is whether it is fraud is and there’s often money laundering afterwards. Often when we think about those two concepts, fraud is the activity of essentially stealing something or taking money from someone else, like an instance of credit card fraud where maybe a fraudster has acquired stolen card details from the dark web, then using those details to purchase something. They can then sell the item that they’ve purchased for cash, so they end up with a pile of dirty money that they then need to transfer into a good source of funds. This is where money laundering comes in. So, we often see money laundering in the act of taking those proceeds of crime and trying to convert them into a sort of legitimate currency that fraudsters can use in their day-to-day lifestyles and activities.
PaymentsJournal:
Interesting! So money laundering itself is kind of unique,
in that it can kind of be seen as post-fraudulent activity. Do you think that
money laundering can be prevented in the same way as other types of fraud?
Excell:
Definitely. I think the way we built our platform enables us
to really understand what good and legitimate activity looks like by customers
of financial institutions. We can use those same profiles to look at specific
types of behaviors that are indicative of money laundering. One of the challenges,
though, is that with fraud, we often get very good determinations – or in the
machine learning concepts, labels – that define when fraud has taken place.
Whereas when we look at any money laundering papers, those are usually referred
into suspicious activity reports, into the regulator. So, getting the
definitive confirmation that money laundering has taken place is not as
frequent as what we see in fraud scenarios.
PaymentsJournal:
Interesting. Now if we could, I’d like to get down to brass
tacks here. I think you alluded to this a bit in your previous answer, but I’d
like to flush it out a bit more. What are some key components that must be
included in technology used to fight both fraud and money laundering?
Excell:
One of the key elements is around data sources and being
able to pull together a good picture of what a customer or business at that
financial institution is doing so you have a well-filled profile and
understanding. One of the key things that we’ve done at Featurespace is to be
able to do that at an enterprise level. Rather than looking at the activity of
a customer or commercial entity, when they’re doing one particular type of
payment, whether it’s a card payment, ACH, wire, or check, today those are
typically sort of monitored in independent solutions. So one of the key things
that we’ve done is pull all of that together into a centralized enterprise
system to have a complete view of what the customer is doing. Outside of that
is not just looking at the movement of money, but also at how they are
interacting with the financial institution, how they’re looking at digital
activity in terms of behavior on-site or on the mobile device, and how they’re
potentially contacting the call center. This gives us much richer context in
terms of understanding how that person is interacting with the bank, which
gives us additional signal to know if it’s criminal or if it’s legitimate
activity from a genuine customer or business.
PaymentsJournal:
Great. Now, you brought up the centralized system that
Featurespace has. What are the benefits of a financial institution using just
one provider to fight both fraud and money laundering?
Excell:
One of the key benefits is to be able to have that consolidated view of a customer and enable one place where you have your financial strategies rather than needing to go through and optimize different systems. When there are different systems in place, the gaps or weaknesses between the systems is often what criminals try to exploit – where that data isn’t carried over. They try to get between the cracks of those systems, essentially get their feet in the door there, and then continue to pry or open once they’ve established that little crack. Joining all those systems together, leverage is also reduced as the potential entry points for the fraudsters and criminals to be able to access. Ultimately, it will help in the fight against crime, but also in enabling genuine customer activities. By having a picture of what the consumer does, and focusing on knowing when we are seeing good legitimate activity, we can ensure that those transactions and interactions continue without introducing more friction to the customer journey.
PaymentsJournal:
Well, Dave, thank you so much for speaking to me today about
Featurespace and the intersection of fraud and AML. I hope to have you back on
the podcast soon.
Excell:
Ryan, it was great to be on. I look forward to next opportunity as well.
The post How Featurespace Is Helping Fight Fraud appeared first on PaymentsJournal.