Diarmuid Thoma, VP of Fraud & Data Strategy at AtData on the power of email address data

Enjoying the podcast? Don’t miss out on future episodes! Please hit that subscribe button on Apple, Spotify, or your favorite podcast platform to stay updated with our latest content. Thank you for your support!

Consider the humble email address. Most of us don’t give this part of our identity a second thought but there is a wealth of information that can be gleaned from this one piece of data. And when it comes to fraud the email address is the most consistent predictor, more than any other data point.

Diarmuid Thoma, VP of Fraud & Data Strategy at AtData
Diarmuid Thoma, VP of Fraud & Data Strategy at AtData

My next guest on the Fintech One-on-One podcast is Diarmuid Thoma, the head of fraud and data strategy at AtData. He has spent most of his 23-year career focused on fraud prevention and today he leads a team that has built the most extensive email database in the industry.

In this podcast you will learn:

  • The history of TowerData and the merger that resulted in the creation of AtData.
  • The two areas of the company that Diarmuid leads.
  • Why banks and fintech should care about email data.
  • Why email is the most consistent fraud predictor.
  • What other data points they take into consideration in their fraud models.
  • The staggering percentage of every email in existence that are in their database.
  • The number of new high risk domains that are being created every day.
  • What is returned to the client when they provide an email address to AtData’s API.
  • How banks and fintechs are using this data today.
  • Why they are included very early in the funnel for lenders.
  • How the AtData quality score works and what it can tell their clients.
  • How they have incorporated AI/ML into their fraud models.
  • Details of their recent white paper on balance customer experience and fraud prevention.
  • The trends that banks and fintechs should be paying attention to when it comes to email data.

Read a transcription of our conversation below.


Peter Renton  00:01

Welcome to the Fintech One-on-One podcast. This is Peter Renton, Chairman and co-founder of Fintech Nexus. I’ve been doing this show since 2013, which makes this the longest running one-on-one interview show in all of fintech. Thank you so much for joining me on this journey.

Peter Renton  00:27

Before we get started, I want to remind you that Fintech Nexus is now a digital media company. We have sold our events business and are 100% focused on being the leading digital media company for fintech. What does this mean for you? You can now engage with one of the largest fintech communities, over 200,000 people, through a variety of digital products, webinars, in-depth white papers, podcasts, email blasts, advertising, and much more. We can create a custom program designed just for you. If you want to reach a senior fintech audience, then please contact sales at fintech nexus.com today.

Peter Renton  01:09

Today on the show, I’m delighted to welcome Diarmuid Thoma. He is the head of fraud and data strategy at AtData. And what we’re going to be talking about today is a little bit different. We’re talking about email data. Now we’ve had fraud people on the show before, and I’ve we’ve done lots of webinars on fraud as well here. This is the first time I’ve ever really dived deeply into email data and how rich that data can be, and what information you can garner from it. It was truly fascinating. I learned a lot in this episode. We talk about obviously, the use cases for email data, where it fits in the funnel when you’re doing like, for example, a loan application, we talk about how many email addresses they’re tracking, the percentage of all emails out there that are part of the AtData system is staggering. And then we talk about good emails and what information you get from good emails, and of course, the bad emails, and those that have really been set up for fraud, how you can detect that. It was a fascinating discussion. Hope you enjoy the show.

Peter Renton  02:19

Welcome to the podcast, Diarmuid.

Diarmuid Thoma  02:21

Thank you, Peter. Glad to be here.

Peter Renton  02:23

Okay, so let’s kick it off by giving the listeners a little bit of background about yourself, why don’t you give us a quick introduction of yourself so we know a little bit more about you.

Diarmuid Thoma  02:32

Well, I guess you can probably tell by the accent, but I’m Irish and based and live here in Ireland in Cork, but in terms of my, suppose my background, career wise, it’s pretty much exclusively fraud prevention, actually. So I started out in banking, thinking, you know, that was going to be my, my future. And did you know financial services in college and so on. And I think within about a year of working in actual banking, I discovered No, I didn’t enjoy this much. So segued into credit, but very quickly, and I turned into kind of fraud prevention, but this was, this was 23 years ago in the early 2000s, right. And to be frank, nobody had a job in fraud back then it wasn’t, it wasn’t really a thing. So kind of figured out as we went along back then. But I liked it. It was it was very kind of interesting. It changed a lot and kind of suited my, I think personality, but so working in Hewlett Packard actually back then and the early days in their fraud prevention stuff. But so when I kind of started building, I was like, Okay, this is getting interesting. So went from e-com, commerce fraud to telecoms fraud, and a few years there. And then in the very early days of Facebook, joined them. Actually, I think there’s about few 1000 people in Facebook now in Ireland, but there was five, I think, when I went there, so we were five or six on the floor, big empty floor. And so yeah, I was there for a couple of years. And then we started our fraud company, it was kind of one of the founding members of complete fraud company called Trustev, and we built that up in a few years. And we sold that to TransUnion for $44 million, I think it was, and then was the kind of global audit fraud side for TransUnion for many years, and funnily enough, a customer of AtData, or TowerData as it was then.  And that’s kind of how I ended up here, actually. So a couple of, about two and a half years ago, after speaking with Tom Barker CEO and kind of convinced me that to build out the fraud side here so, so yeah, so 23 years kind of goes by quick.

Peter Renton  04:39

Interesting. Okay, so tell us a little bit about AtData and TowerData, I guess it used to be called, but why don’t you to tell a little bit of the history of it, and how you describe it today?

Diarmuid Thoma  04:51

Yeah. So going back it was actually founded by, as I mentioned there at Tom Burke, our CEO or currency in 1999. So TowerData was and so, primarily starting in like email validation, you know, checking that, pinging that, make sure the email is valid, deliverable, etc, mostly for marketing services and stuff like that. That company just grew and grew. And so bringing it into 2022, to the point where, you know, it had 1000s of customers already, global kind of reach already processing billions of emails. But to bring it even further, we merge TowerData, as it was known then, and FreshAddress. So two of the kind of pretty, pretty big players in the email validation space, merged in 2022. And that’s effectively when the rebrand happened as well. So you know, as it was kind of a, I suppose a rebirth, almost like it instead of a rebrand, because it was the signaling of a very different change of direction for both companies as well, you know, when I joined and to heavily invest in fraud, and then also into kind of new technologies which would cover areas like AI and the ML technologies. And we’ve really doubled down on that, and expanded is probably our fastest and biggest expansion is on that side. So yeah.

Peter Renton  06:13

Okay. So you sort of lead up the fraud activities, although anti-fraud, I probably should say, activities at AtData.

Diarmuid Thoma  06:21

Don’t worry, I do the same thing is like I work in fraud, but ya know, fraud prevention. So yeah, primarily, it’s the fraud and data science side of things is what I look after, because the two are pretty tightly coupled. That’s been the biggest expansion, as it were, for the company, right, is that the company has always had huge resources from a data perspective, right? You know, and again, wealth of customers, and both companies did I guess, and but the change has been to use that and use the AI and use the ML tech capabilities that I brought, you know, and to bring it into a new, a new kind of age. And that’s exactly what we’ve been doing for the last couple of years. It’s been very, very exciting. But myself, what I do is largely  develop the new fraud products and the new fraud features. And we’ve been, we’ve been rolling them out pretty fast in the grand scheme of things, I suppose is that, you know, like, we have new technologies around domain risk, and you know, and new ML models, new, you know, all of this. It’s basically just trying to keep up with the industry, but keep up with the fraudsters and keep up with customer needs and demands. And then yeah, and talking to sales. One of the things I do, try and do a lot of actually. It does take up a lot of time, but it’s definitely worth it is speaking to customers. Speaking to the, you know, within the sales process, the post sales process, you know, seeing what their challenges are, seeing what’s new in the market, seeing what’s not there, and what would maybe what we can do. So  that’s a lot of my time and the odd podcasts and webinars.

Peter Renton  07:51

Right. Okay. Well, maybe you could tell us what, what are the industries that AtData serves? So I imagine this is, you know, some of this you’d mentioned marketing services before, I mean, this would go across industries, right?

Diarmuid Thoma  08:04

Yeah. Like, and probably the simplest way to explain it is that in its current form, you know, we have kind of two main divisions within the company. And that is the the martech side and the fraud side, right. And they’re distinct from each other. In like, obviously, we’re both doing kind of the excellence of email, but they have two very different purposes. So on the martech side, where we’re very focused on deliverability, quality of customers, including, you know, cleansing databases, enhancing databases with names, address and stuff like that. And it’s a huge part of the business, you know, obviously, with swipe pressure tested as well. Then on the fraud side, it is all on the fraud prevention side, very, very different demographic in terms of the clients and different, the cost, obviously, of what we’re dealing with is a lot different in the fraud side as well. So the variety of customers is expansive. I mean, you’re talking, you know, some from the largest fortune five hundreds down to the smallest, to kind of mom and pop shops on the street. So you know, so it’s, it couldn’t be more diverse from from a client base perspective. But yeah, from a use case, it’s primarily those two.

Peter Renton  09:15

And so within financial services, I’m curious about kind of the use cases there. And we obviously know that emails are a critical piece of of anyone’s identity these days. But looking at the financial services use case. Why should banks and fintechs care about email address data?

Diarmuid Thoma  09:36

They don’t have to take my word for it. It is the most important PII element when it comes in the digital world, right? You know, we’re used to very fit, you know, our names and our addresses and stuff like that being important, you know, in the physical but like, particularly post COVID. Everything, everything from my perspective, I certainly do, the vast majority of my dealings are all online now. So for digital emails, one of the most consistent elements within that. I’ve had mine since I was about 16. And I haven’t changed it, and it would be a big pain to change the email, you know. So it’s very consistent, very sticky. It’s also very, it’s very useful, you know, in that I need it every day to do my insurance, I might need it to do my banking, but I need always an email. So it’s always there. But from a fraud perspective, that is where it gets interesting, right? When you think about, yeah, you said lending, right? So in a modern world, right, so if I’m going for a loan, I’m not gonna go into a branch, I’m gonna go and do the application online, right? And that’s for the vast majority, that’s the way banks and fintechs obviously are doing it. What’s the point in doing a name and address check? I’m not sending the money there. I’m not sending any check to the house and the name. Well, as a fraudster, I can write down any name and address I want, you know, so that really doesn’t have any bearing, you have to think about from a fraud perspective, what are the elements of that application that they need to be there? In their control? So one, the receiving bank account, right? They need, they need to be in control of that, because that’s the goal is to get the money to that account. The other thing is, well, they need to be in control of the communication, which is primarily email. So they’ll either have compromised it, or just most cases just created a new one. And that’s where we come in, you know, we say like, No, this email, this email was only created yesterday, despite what they say, and it’s coming in from this location, it’s coming, it has no activity until now, this application. So that’s just one kind of nuanced one for, you know, in terms of lending, but like, there’s so many applications. And it is, even when I was a customer of TowerData and using this data in TransUnion, it was by a good margin, the most consistent fraud predictor and correlator in terms of our models that we were running then.

Peter Renton  11:59

So just because you created an email yesterday doesn’t necessarily mean it’s fraud, right? You could just say, right, I’m done with my email, that I’m switching to a new service, and you have to do something for the first time. So, you know, obviously, it’s only one data point. How do you kind of, dig a little deeper here?

Diarmuid Thoma  12:19

Yeah, so it’s not just like, as you said, some at some point, every email is new, right? Now, I will say for the vast majority, the frequency of change is low, right? You do generally hold on to your emails, and there is kind of new versions of that coming. But for the most part is low. And we’re not saying we never like in inner product, we always say email risk. We’re not saying email fraud. We’re not saying just because you have, has it increased the risk associated with this application? Or this, whatever, transaction? Yes, it has, right. But you’re right, in that we can’t just base it on that. And we don’t, we have other variables, like I’d say we take into account the domain, we take account into the location coming in. So we also take in like, IP data and stuff like that, that’s coming in as part of that application. And it’s all kind of weighed up as part of of that, but critically for us is that we are an AI ML house, right? And we we can actually tune to our customers, our individual customers demographic, what would their good profile customer look like? And we train the model to that demographic uniquely. So our accuracy is pretty exceptional in because of that. So there’s false positives. Yeah, you know, there’s always false positives in everything that you know, in this industry, but they are certainly very limited.

Peter Renton  13:38

Okay, so then, what’s the size of your reach? I mean, like, how many email addresses are you tracking?

Diarmuid Thoma  13:45

You know, when we talk about the count okay, it is billions, right? Every month, we’re processing about billions of emails. However, I never kind of, you know, there’s a lot of companies around say well our database is 16 billion or whatever. And it’s like, that’s kind of meaningless to a customer, or to a company. Because how do I know that those billions apply to my demographic? So the way I explain it to most of our customers, I said, Alright, let’s say, let’s say they’re operating in North America and say, right, when you send us your emails, 99% of the time, we’ve already seen them, that’s our reach. So only 1% are considered new. And we’ll never be 100% because there is, as you said, a certain demographic that genuinely are new. Now, a lot of them are fraud, but our reach is pretty expansive. And that applies globally as well. So for the most developed countries globally, we’re in the 90s, and high 90s in terms of our recognition rate, where we’ve seen them and that goes back decades.

Peter Renton  14:43

And so, like, how are you building that database? I mean, I imagine like every time a client will send you their emails to do some verification, that like the 1% that are new, there might be someone who’s you know, turned 18 and never seen the email address before and they’re doing…

Diarmuid Thoma  14:59


Peter Renton  15:00

Is that how you’re building your database?

Diarmuid Thoma  15:01

Yeah, basically, yeah. So the benefit of being quite large and having a large customer base who send you a lot of traffic is that you can derive. So if the product effectively, the fraud product, is a byproduct, you know,  that’s what it came from. So from our validation company, it’s the, you know, being able to see all this activity and you’re like, No,  and a couple of early fraud companies started adopting it without necessarily that being its use or intention, but saying, I’ll use that, and it just exploded. And now it’s almost an industry in itself.

Peter Renton  15:33

Right. Right. So are there certain high risk domains when it comes to email addresses? I mean, I imagine everyone’s got a Gmail address it feels like these days, and you’ve got Yahoo, and the old AOL and that sort of thing here. How do you know when this is a disposable email address, or are there high risk domains that you actually deal with?

Diarmuid Thoma  15:55

Oh, yeah, I mean, we published a blog on this actually, because this is one of the biggest changes in the industry over the past few years, right. So back in when I was managing kind of fraud platforms, you knew there was always disposable domains, you know, and then they’re, they’re not created for fraud. They’re created for legitimate purposes. And you said earlier about, you know, I just don’t want spam, I don’t want all that. So I’ll just use this disposable email. And that’s fine. And that’s their, that’s their use case. However, they are adopted for fraud, and massively so but they’re still manageable, like, in terms of, you know, how many domains were out there, you know, you’re talking, whatever about a couple of million, maybe a year, new ones created. And that sounds a lot. But in COVID, you know, I mean, everybody knows that mass migration to online, but also mass increase in fraud, universally across every industry. And we trend, obviously, the numbers and stuff like this, and that went from about 900,000 domains being created a year to 2.2 million in space of a couple of months. That’s not domains all. That’s just the high risk domains, right? So it’s 2.2. But today, that figure stands at about three point, nearly 3.5 million every year. So for context, that’s 100,000 new domains today. 100,000 more everyday, high risk domains. And that’s been the change, right? It’s a very interesting kind of behavior, because what it means is now the domain itself is disposable. So, you know, you think 100,000 is a shocking number every day, what’s even maybe more shocking is that about 40, or 43%, to be specific, of those are only active for about two weeks, three weeks max, and they’re gone, they’re never seen again. So the industry problem like was that by the time most companies have classified them as high risk, or disposable or whatever, they’re gone, they’re done, because it took about six months to apply that classification. So that’s why they’re so prolific. We spent last year working on kind of new technology that would basically have real time classification, immediately. The first email of the first millions of everything, and it would be classified there and then so we released that actually a couple of weeks ago. So we’re quite excited about that. But yeah, it’s been, that one has been the I think, is quite a shocking in terms of the industry impact it’s had, and I say, I don’t say like all disposable are fraud, but it certainly is a big, big factor in the fraud industry.

Peter Renton  18:28

Right, right. So I want to get back to sort of the use cases for financial services. And, you know, it sounds like you have an API now. And you’ve released a real time updates to that, which obviously that’s fantastic. How do you actually work with financial services firms? You talk about email risk. Do you provide a score for each email? Or what are your clients getting back?

Diarmuid Thoma  18:52

It’s primarily almost exclusively API based, right. So it’s a real time API, generally. Like we can go, it’s incredibly fast. For the most part, it’s about 500 milliseconds, there abouts, or faster if we need to be. What’s returned to them is quite a lot of data, actually. You know we only need an email, but we can return back a huge amount. Because I mentioned at the start, we have a very wide base of customers, right? Some of them are just regular stores that need some help on their online shop, you know, so they’re small volume, and some of them are super mega, you know, fraud platforms or payments platforms. And so at that, so and but they don’t want just a score, they want the score, yes, but they want every piece of data that supports that decision. So we return it all as part of that. And that score is basically our customizable models running and producing that profile risk back to the customer. In terms of the actual banks and how they use it, so the deployment generally speaking, so if we think of the loan application, I suppose is the most common one or that could be card, or it could be anything but it’s the most common use case for the fintechs or banks. We’re quite early in the funnel, because not only are we quite effective in terms of the fraud prevention side of things, and quite accurate, we’re incredibly easy to install. It’s a very, you know, it’s quite a small product compared to some of the big platforms out there. So it’s quite easy to install. But most importantly, I suppose for them, it’s cheap compared to a credit check, for example, or doing that, because again, logically, why would you do quite an expensive credit check on a fraudulent application? It’s just a waste of money, you’re going to kick it out anyway. So that’s why we’re, generally speaking, quite early in that funnel. And we get back a lot of you know, for those companies who are really maximizing, we give back quite a lot of information for their client demographic. You know, what type of emails, what type, where are they coming from, are they coming in on mobile, are they coming in, you know, location wise? All that kind of data is returned in the API response. So it’s quite a rich dataset for not just fraud, but for general, you know, KYC type stuff.

Peter Renton  21:00

I was reading about a piece of news you did. You recently launched the AtData quality score. Is that the score we were just talking about? I mean, what is what is that exactly?

Diarmuid Thoma  21:10

Interestingly, no, no, it’s uh, so this is a kind of personal favorite of mine, because it is the polar opposite of what we were talking about, right. So yeah, so it has similar DNA, right? It is a full, AI/ML model based product, right. But we’ve been talking, let’s say, you know, in terms of the fraud, how bad is an email? And we have a score, saying from zero to 100, and 100 is really, really bad. And that’s high, high risk. Well, the quality score is the other end of the spectrum, the quality score is designed to tell you how good, potentially a customer is. So if you imagine them, and that’s been like, that’s been the AtData kind of mantra, and it’s been what we’ve been trying to achieve is that when it comes to email, we’re it. You know we can tell you everything that could possibly be told about an email. And that’s what this is. So you have the fraud. So well, let’s say, you know, if you’re looking at it from a flow point of view, the fraud has kind of come in and said, Yeah, okay, we’ve gotten rid of all the bad actors. And that’s it. So now we have all these ones that didn’t score badly for fraud. But I’d like to know, how good are these guys? So, for example, how engaged are they? Do they, you know, do they read all their emails? How they, how active are they? And maybe they haven’t been seen in three years, you know, or before now. And then I suppose, basically, and this is where it gets kind of really interesting is, how likely are they to spend a lot? You know, are they big spenders, when they do come in? And we have this, this is all still AtData data, believe it or not. So you’re, again we’ll use the lender, which is the consistent theme here. If you’re a lender, and you have a credit limit, let’s say a short term lender, and you’ve credit limit of new loans for $1,000. But this, we can run the quality score and said, Well, they’ve very high activity, very good, very good email, propensity to spend is very, very good. So maybe we make that 5000 For this, you know, so it’s a pre-qualification kind of tool as well. And that’s on that side. But we also, because the one of the reasons I liked about it so much is it’s so flexible, we can deploy it in a marketing context, and then it helps deliverability it helps, you know, sales enablement, it helps all of that for that context. So we can actually change the meaning of good for the product the clients require. So yeah, it’s cool.

Peter Renton  23:25

You’re gathering a lot of that from the transaction data, right, and you’re getting a lot of payments data from just from your database of customers?

Diarmuid Thoma  23:32

Yeah. And like the product is designed to evolve, it’s in that, so we’ve designed it to be flexible as it evolves into the future. So for example, like, a definition of good is different, depending on the industry. So a fintech versus industrial versus e-commerce is that they’ll all have different versions of good. And so we’ll effectively end up with different versions of that, and that data is going to grow and grow.

Diarmuid Thoma  23:57

Okay. I want to go back to the data science hat. I’m just curious about, you’ve talked a little bit about you know, AI, but maybe you can just sort of tease that out a little bit. How is AtData using AI to build your deliverability and your fraud models?

Diarmuid Thoma  24:13

You know, it’s funny having, as I explained, having come from kind of not using it much at all. If I was to say today, it’s pretty much now in our DNA. It’s almost embedded, like our the modeling and the feedback, and all that, is now pretty much embedded into every product we have. Like I was talking earlier about like our high quality data, and we’ve had that and that’s great and you know, but what the AI has allowed us to do is take high quality data, because remember, it doesn’t matter how good your AI is, if you’ve got crappy data, your AI is going to be pretty crappy too., right? So because we had that from the get go, we were already having a headstart. And so by being able to apply all of this, we were able to bring our products into kind of places wh,ere nobody else had, or even us had conceivably done. So when we look at like the fraud, you know, the fraud customization in the modeling and like the data can be got, you know, there’s other places maybe, but what the levels of accuracy that we achieve now, all because of that customizable modeling, and all because the uniqueness of that, those decision elements is unmatched. There isn’t anybody that can come close. And that’s what it can do. And we’re only, we still only think we’re getting going like, as I said, it’s in every product, but it’s still, it’s just getting going.  And we’re expanding. I think we have a new data scientist starting today or tomorrow, actually.

Diarmuid Thoma  25:11

Okay, so you released a white paper fairly recently, which I will link to in the show notes, on the balance between customer experience and fraud prevention. It’s always a dance, right? So what are some of the key takeaways from that white paper?

Diarmuid Thoma  25:58

And you say dance? I think I’d probably describe it as a battle half the time because growing up in the fraud industry, it’s always like fraud versus sales. And the two are like that all the time. Because sales look at the fraud department as a barrier, as you know, they’re the Sales Prevention Team effectively, you know. But to be honest with you now, fraud has evolved, fraud prevention has evolved so much, if that’s still how you see it, then you’re doing it wrong.

Peter Renton  26:28


Diarmuid Thoma  26:28

Because a good fraud strategy, a good fraud systems, yes, they should be able to effectively and accurately take the risk out of the transaction or the whatever you’re doing. But because they are there, they should be able to enable new revenue, right? And that’s what the article is, it’s a long enough article or white paper about it, but effectively that’s what it’s about is that the balance between it, and you need to be able to deliver that to customers, right. You need to be able to, the customers expectations have, are rising. Everything is rising, right, fraudsters technologies are rising, the customer’s expectation of what their friction  they tolerate is, is getting smaller and smaller every day. And the businesses, they expect those services to get more advanced and cheaper every day, you know. So everything is going up. But, you know, the thing is that effective fraud strategy should be able to open up new doors to a business, so markets maybe that they thought were too high risk, products that they thought were too high risk, etc. If you have a manageable fraud to you know, system, you can open that. So it’s it really is changing and it has, not everywhere yet, but it certainly is looking that way now.

Peter Renton  27:39

Right, right. What are the trends that banks, fintechs and others should be paying attention to here when it comes to email data?

Diarmuid Thoma  27:48

So the big one for the last question is it? Two, I think come to mind, like the first one is kind of what we’ve been talking about, right? Because it’s current, but it’s emerging, is the high risk domains. Like I said, that wasn’t as big as it was, a couple of years ago, that was manageable a couple years ago. It’s exploding now. And they have to keep in mind, these aren’t just like, high risk domains, and for disposable. These are like, these include like spoof domains, you know, so, for example, it’s a simple one, but we have just over, I think it is, 100 versions of Gmail, on our active. So we have Gnow, Gmal, Gmil, so many different combinations you can possibly imagine. But we have spoof banks as well, you know, versions of, you know, the big banks that are out there, and they try and replicate it to make it look like that. And we see those emails. So that side of it is increasing at a rapid, rapid rate. Every customer I speak with today, is saying, Yeah, we struggle with this. So that’s something that they need to have, make sure the measures are there, make sure they can, they can deal with it, and make sure they understand what they’re dealing with when it comes to that, because it’s still gaining momentum on that side. The other is a bit more future states, I suppose and long-term. But, you know, we talk about it, we see it in the news as the, you know, the advent, and we talked about it here with the advent of AI, and its, you know, what it’s doing to all the different industries and how it’s changing what we do. But keep in mind, fraud itself is an industry and I don’t mean the fraud prevention, I actually mean the fraudsters in this case, right, and they’re using it and they’re, they’re using it for, to increase their scalability because that’s what they do, right? They’re all about scalability efficiency, right? They’re just like a business. So they are also leveraging it now. We don’t know exactly, yet, how that’s going to manifest itself in different, we’ve seen it already. I’ve seen you know, we’ve seen much greater scale in terms of like email tumbling, which is like the changing of an email to appear different and things like that. We’ve seen that being scripted, automated, massively automated, so you know, you see thousands and thousands of them come in every couple of seconds kind of thing. So we’re still seeing that evolution now. But it’s just the beginning. You know, so for the banks and for the finance industry, making sure you’re able to cope with that, that shift, that scale, that your system that you have in place isn’t just built to deal with your current fraud problems. It needs to be future proofed, it needs to be able to adapt quickly, because those trends will happen. And you’ll be surprised how quickly you get hit by one of them. So if they have that flexibility to take in a new data feed, or you know, just maybe change their existing one to cope, then great, fantastic, or they have alerting so I think they’re the two biggest ones. There’s probably 15 others, I could probably list, but those are the two that come to mind now anyway.

Peter Renton  30:51

Okay. Well Diarmuid, we’ll have to leave it there. That was a really interesting discussion, I learned a lot today. I’m sure the audience did as well. So thank you very much for coming on.

Diarmuid Thoma  31:00

That was a pleasure, really. Really was.

Peter Renton  31:04

Well, I hope you enjoyed the show. Thank you so much for listening. Please go ahead and give the show a review on the podcast platform of your choice and go tell your friends and colleagues about it. Anyway, on that note, I will sign off. I very much appreciate you listening, and I’ll catch you next time. Bye.