How to Benchmark Fraud Performance and Find Hidden Gaps

WEBVTT

1
00:00:03.040 –> 00:00:19.950
Jerry Hoff: Welcome, everybody, and welcome back to the Blueprint series. My name is Jerry Hoff, and this is how to benchmark fraud performance and find hidden gaps. So, if you want to find out all about benchmarking on fraud, you’re in the right place. I want to introduce

2
00:00:19.950 –> 00:00:28.690
Jerry Hoff: our… our guest for today, Maria Benjamin. So, Maria is a long-time fraud expert. How you doing today, Maria?

3
00:00:28.690 –> 00:00:31.769
Maria Benjamin: Good. It’s sunny here in Chicago.

4
00:00:31.770 –> 00:00:44.839
Jerry Hoff: Well, it’s sunny down here in Fort Lauderdale as well, as usual, so that’s great, that’s good for Chicago. Maria has an amazing amount of experience, and I think most notably, Maria, you were at Venmo.

5
00:00:45.190 –> 00:00:50.600
Jerry Hoff: We’re also at, we were… you were at Venmo, and then you were also at a.

6
00:00:50.600 –> 00:01:11.789
Maria Benjamin: Yeah, as a teen, banking, app, not a bank, called Step Mobile, which actually recently got acquired by, MrBeast, because it’s Gen Z, but it’s a secret credit line, so I was right there when they started, I think, a year on Market, to help them with their fraud solutions.

7
00:01:12.080 –> 00:01:23.919
Jerry Hoff: Which is amazing, because I can’t even begin to imagine the fraud scenarios you must have seen between Venmo and between StepMobile. I guess it’s MrBeast’s headache now, but

8
00:01:23.920 –> 00:01:33.980
Jerry Hoff: So, hopefully, we’re gonna take a lot of that experience today, and then we’re gonna, you know, use that for the benefit of our audience on benchmarking. By the way, for folks who are on.

9
00:01:33.980 –> 00:01:44.630
Jerry Hoff: Again, welcome. We want to hear from you, so we’re going to have some polls that are going to be completely anonymous, and then feel free to put in questions and answers, and we’ll work those into the presentation.

10
00:01:44.630 –> 00:01:48.200
Jerry Hoff: Maria, anything else we should add before we dive right into the material?

11
00:01:48.570 –> 00:02:04.309
Maria Benjamin: Yeah, I’ll just say at SIFT, my main thing is working with customers at their various points to think about how are we going to improve workflows, what’s your strategy? It’s not always necessarily SIFT’s implementation, sometimes it’s just advice on other areas as well.

12
00:02:04.660 –> 00:02:06.840
Jerry Hoff: Excellent, excellent, wonderful.

13
00:02:06.870 –> 00:02:21.299
Jerry Hoff: Well, let’s get into benchmarking, and I’ll say up front, when I hear the word benchmark, I feel like it’s homework or something, but I know that it’s not, and I know that in anything with fraud or security, or really anything, health, etc.

14
00:02:21.300 –> 00:02:28.780
Jerry Hoff: it always is compared to what, right? So, raw numbers don’t really tell us anything unless we can make a comparison.

15
00:02:28.780 –> 00:02:34.069
Jerry Hoff: So, Maria, walk us through the benchmarking concepts that we have here on the slide.

16
00:02:34.340 –> 00:02:59.040
Maria Benjamin: Yeah, it’s like, it’s not unlike math, but I hopefully… a little math homework, but it’s more interesting than that, at least I hope so. It’s the idea that you’re taking these raw numbers, right? You know somebody’s purchased something, you know you’ve got a certain amount of users, you can see what they’re doing on your application over time, so you’re taking those metrics, and you’re putting context around them. You’re understanding

17
00:02:59.040 –> 00:03:21.570
Maria Benjamin: hey, what’s our acceptance rate? What’s our chargeback rate? When do people spend the most money? Is it on Friday when they get paid, or is it Saturday when they’re going out? So you’re taking that and you’re contextualizing your behavior and your patterns for your company. Then you’re thinking about, the gaps there. So you’re like, great, I have no chargebacks.

18
00:03:21.570 –> 00:03:38.880
Maria Benjamin: Well, if you have no chargebacks, are you leaving money on the table? Have you made your product really unusable for people? So that’s where understanding how they play together then can help you think about, where are my blind spots? What am I… what are my trade-offs here?

19
00:03:39.210 –> 00:03:40.609
Jerry Hoff: Yeah, go ahead, I’m sorry, go ahead.

20
00:03:40.830 –> 00:03:54.059
Maria Benjamin: I was gonna say, I’ll move on to the next one unless you got, a part to add, and then that’s where you start thinking about balance, where you’re thinking about revenue, friction, your cost. I know a lot of people in the fraud area, it feels like you’re always

21
00:03:54.060 –> 00:04:11.340
Maria Benjamin: fighting product, where they want to launch something, and they’re like, you’re crushing my dreams, and I’m like, no, I’m trying to make sure you don’t lose us a million dollars. So that’s where… it’s always just about trade-offs here. It’s never thinking, like, there’s one perfect solution, you want it to be at zero.

22
00:04:11.340 –> 00:04:26.290
Jerry Hoff: Very well said. Yeah, you know, it’s interesting, because I’ve been in these kind of meetings where they’re like, hey, our chargeback rate is 0.4%, and I’m thinking to myself, is that good, or is that bad, right? Is it… is that something we need to improve, or is that, like, we’re, you know.

23
00:04:26.570 –> 00:04:40.859
Jerry Hoff: are we above industry average? So, absolutely, absolutely. Yeah, I like what you said. I feel the same way when I go to the doctor, right? If they’re like, well, your blood pressure is, I don’t know the numbers, 80 over 120. I’m like, is that good, or is that bad? So…

24
00:04:40.860 –> 00:05:05.400
Maria Benjamin: Right? And thinking about if it’s in range, right? Like, you can be a little high, because you’re like, oh, you just ran a mile, it’s okay for it to be up, and then you’re like, it’s a little low, maybe you need to think about doing something else. So that’s definitely very much where you can have sustainable levels, and it’s also reminding people that your business is also special. Of course, you’re part of an industry, and that’s where you think about tracking against your industry, like.

25
00:05:05.400 –> 00:05:10.660
Maria Benjamin: Oh my gosh, food service is just very fraudy,

26
00:05:10.660 –> 00:05:15.960
Maria Benjamin: But it doesn’t necessarily mean you don’t have the ability to try and control that problem.

27
00:05:16.430 –> 00:05:27.649
Jerry Hoff: Maria, do most organizations benchmark, or do a lot of them skip, or do they… are they benchmarking the wrong stuff? I know we’re gonna get into it, but just as a temperature check, what does that look like?

28
00:05:27.650 –> 00:05:52.619
Maria Benjamin: I was gonna say, everybody usually has a performance metric that they’re at least looking at, and we’ll get into this a little bit later, but you kind of sometimes get moving targets, where you say, oh, I took care of that fraud problem, and then you’re like, great, team’s doing awesome, and then somebody chooses something else, or they decide to look at it through another lens, and you’re like, well, by fraud standard, we’re doing great, but maybe by product standard, you’re like, oh.

29
00:05:52.620 –> 00:06:02.109
Maria Benjamin: we’re very under. So, that’s… most places are tracking something, or… but I have seen people say, if it’s not…

30
00:06:02.110 –> 00:06:08.519
Maria Benjamin: if I’m not hearing about it, it’s not something to be concerned about. And that’s not always the greatest tactic, because then…

31
00:06:08.520 –> 00:06:09.689
Jerry Hoff: Out of sight, out of mind.

32
00:06:09.690 –> 00:06:17.230
Maria Benjamin: Yeah, then you get the fire drills, then you get the visa remediation, then you get people saying, I need to do everything now in the kitchen sink.

33
00:06:17.880 –> 00:06:19.449
Jerry Hoff: Makes sense, makes sense.

34
00:06:19.620 –> 00:06:33.259
Jerry Hoff: Very good. Well then, that brings us to the next slide, which is what… okay, so you’ve now just convinced me most organizations are doing some kind of benchmarking of some kind, but they might not be doing the right benchmarking, so we want to get into what undermines

35
00:06:33.260 –> 00:06:42.910
Jerry Hoff: most organizations’ benchmarking efforts. And this brings us to our very first poll. So this is an anonymous poll. Would love for participants to fill this out.

36
00:06:42.910 –> 00:06:54.520
Jerry Hoff: So what is your fraud organization’s primary KPI? It also popped up for me, too, so let me move it out of the way. Maria, walk us through what these are, you know, these terms here, if you don’t mind.

37
00:06:54.940 –> 00:07:18.590
Maria Benjamin: Yeah, and I was just thinking about… you can kind of choose whatever KPI you want, you just want to make sure that what you’re building your foundation on is going to be very solid, as in, you aren’t double counting things. I’ve seen people go, oh, a wallet transaction and a credit card transaction both came through, and I’m like, but that’s one transaction, but they counted it in two buckets, and then they had.

38
00:07:18.590 –> 00:07:19.120
Jerry Hoff: venue.

39
00:07:19.120 –> 00:07:43.909
Maria Benjamin: very strange things. So you just want to make sure you’re… like I said, you can pick any of these, or other things, but you just need to make sure the foundation is accurate over time. So we’re thinking acceptance rate, like, what is the authorization? Is it going all the way through to the bank? Are you getting, accounts taken over? What’s your security looking like? Are you getting blocked a ton? Is somebody who’s

40
00:07:43.910 –> 00:07:49.600
Maria Benjamin: It’s been allowed to go through continually getting issues, even after they’ve been approved.

41
00:07:49.600 –> 00:08:09.529
Maria Benjamin: chargeback rate, we’ll come more in depth on this one, but just the money getting ripped back from a company because of, the bank, or decline rate is just how often somebody’s getting, like, said no to, and then false positive rate, which is how often did I say somebody was fraud, and they weren’t.

42
00:08:10.600 –> 00:08:14.700
Jerry Hoff: Does anybody ever say, oh yeah, we’re measuring all of these KPIs? Does that ever happen?

43
00:08:14.990 –> 00:08:27.449
Maria Benjamin: I have definitely seen people where, like, they’re measuring a lot, even more than this, actually, but it ends up getting a little lost in that, how often are they actually reviewing them? They often will

44
00:08:28.020 –> 00:08:41.289
Maria Benjamin: around, where they might say, like, this is the one that we’re focusing on right now, and then, oh, something else happened over here, so then they’re, like, running to scramble and, like, squash that problem. So I’ve definitely seen people measuring lots of things.

45
00:08:41.480 –> 00:08:44.820
Jerry Hoff: Yeah, you know, it’s… I’m… I… I’m not a fraud…

46
00:08:45.120 –> 00:09:04.780
Jerry Hoff: I don’t work in fraud per se, obviously, but I do cybersecurity, and it’s very similar, where, you know, either organizations are measuring too much, they’re measuring too little, or they’re just not taking… they’re not ingesting and building a virtuous cycle with the data they’re collecting to improve the situation, so…

47
00:09:04.910 –> 00:09:15.819
Jerry Hoff: All right, very good, very good. So, let’s get into, let’s get into the Know Your KPIs, and we did get our, our, our quiz, our poll results back.

48
00:09:15.930 –> 00:09:18.040
Jerry Hoff: So I’ll mention what those are.

49
00:09:18.340 –> 00:09:29.169
Jerry Hoff: Chargeback rate came in at 46%, so it looks like the majority of folks are doing chargeback rate, followed by block rate, which is 38%,

50
00:09:29.500 –> 00:09:46.199
Jerry Hoff: Nobody, 0% on account takeover rate, and then we had 8% for acceptance rate and decline rate, and then 0 for false positive rate. Maria, how does that stack up with your experience? And then maybe if you could walk us through some of these other KPIs, that would be great.

51
00:09:46.570 –> 00:10:06.479
Maria Benjamin: Yeah, I was like, this is an interesting area, because I’ll say, yep, chargeback sounds so familiar, that’s kind of, I think, what most fraud teams are accountable to. But it’s a little interesting that nobody said false positive. Usually that’s where people are thinking about that insult rate, that’s saying, like, hey, how often is somebody who

52
00:10:06.480 –> 00:10:31.220
Maria Benjamin: is a legitimate customer, and we’re leaving money on the table there, is not something that’s getting measured anywhere. So that’s a little unusual to me, because that’s often what I hear from executives are really concerned about that one. So it could be the different audience in the room. So I did want to, highlight what you mentioned here was the friction and lifetime value of customer. I know we talked about some of these others of

53
00:10:31.220 –> 00:10:56.029
Maria Benjamin: block and decline rate, and these are usually based around transactions, things that you’re normally, concerned with, with money getting blocked, or it going straight into your bank account, and it goes on the bottom of the register, and turns into profit. But friction and lifetime value of a customer are things that often product is really concerned about, and depending on what area you are in the business, sometimes

54
00:10:56.080 –> 00:11:09.110
Maria Benjamin: this is what everybody is concerned about, because, you’re just trying to survive. Every dollar is a good dollar, which is not necessarily true, but I do like to think about this where people, like I said.

55
00:11:09.110 –> 00:11:23.450
Maria Benjamin: think of fraud as something that’s siloed, and, oh, it’s crushing the product, it’s crushing our dreams, but these are ways that if fraud is also measuring these KPIs, and they’re aware of it, then they’re also making strategic decisions.

56
00:11:23.450 –> 00:11:25.309
Maria Benjamin: In line with the company.

57
00:11:26.270 –> 00:11:39.199
Jerry Hoff: You know, again, very similar to cybersecurity and information security. If you do too much, you’re losing customers. If you do too little, you’re losing to the criminals. So, that balance is really hard to…

58
00:11:39.410 –> 00:11:52.579
Jerry Hoff: I guess that’s that tightrope, which brings us into the next slide, which is the North Star. This was interesting, Maria, and I’ll give a little sneak peek to everybody. It’s a different

59
00:11:52.580 –> 00:12:04.539
Jerry Hoff: North Star, depending on what stage of the evolution of your company you’re at. So this is early stage and startup stage. Maria, walk us through what the North Star for those folks should be.

60
00:12:04.920 –> 00:12:19.920
Maria Benjamin: Yeah, so this is… every business is its own thing with what they’re thinking about, but early stage, you’re kind of thinking about, I’m just trying to get my company far enough along that, it doesn’t matter if I have, really high

61
00:12:19.920 –> 00:12:35.980
Maria Benjamin: chargeback rates, and I go into visa remediation if I have no customers. So, oftentimes, people don’t think about this as it evolves, but you’re like, this is what I’m really focused on. But it also helps fraud teams think about not only trying to stop

62
00:12:35.980 –> 00:12:59.030
Maria Benjamin: money that’s gonna turn into a chargeback, but also, how can you create those premium experiences? So you’re thinking about, like, hey, can I reduce friction for people? And you flip it, you say, alright, I’ve seen this person for 3 months, we’ve been watching them, like, they own this credit card, I know for certain, which means I can get this money back if they’re actually going to charge back.

63
00:12:59.030 –> 00:13:23.309
Maria Benjamin: They care a lot about their account, which means if something happens, I know who to… is accountable to it. So these are things where you think about turning fraud into the area of, like, can I promote user growth? Can I actually think about reducing friction? Where is that going to be without it tumbling off the hill, right? Like, you still need to be able to process, and you do want to think about.

64
00:13:23.540 –> 00:13:47.809
Maria Benjamin: maybe I don’t want to get investigated by the FCC or something like that. But you are going to allow more chargebacks, your margins are going to be thinner, you’re thinking about that payment fraud. So, this is one that I think people are a little surprised that we say, are you measuring this? But it might be something that’s very, very important. And not necessarily only at early stage, but it’s the one I’ve seen very prominently at early stage.

65
00:13:48.740 –> 00:13:57.939
Jerry Hoff: it makes a lot of sense, and, you know, in development, you know, building web applications and SaaSes and so forth, the, you know, the cardinal sin is

66
00:13:57.950 –> 00:14:16.669
Jerry Hoff: trying to optimize too early, right? And it sounds like there’s a parallel here. When you’re in an early-stage startup, or you’re an early-stage or a startup, whatever it happens to be, you… yeah, chargebacks are a good problem to have, in the sense that if you don’t have any customers, then you don’t have any business. So, very, very interesting parallels.

67
00:14:16.680 –> 00:14:27.169
Jerry Hoff: Maria, drawing from your experience, and obviously don’t say anything that you can’t say, but what are some of the more interesting fraud scenarios that, I guess, you’ve seen

68
00:14:27.580 –> 00:14:44.070
Jerry Hoff: I don’t know, is there something, like, something memorable that jumps to your mind when you think about, like, an early-stage startup, and I don’t know if it’s, you know, benchmarking or something, but, you know, does any story jump to mind? Because I would imagine you’ve been in several, and you must have seen some pretty

69
00:14:44.410 –> 00:14:47.739
Jerry Hoff: outrageous things, but I don’t want you to say anything that you can’t say, obviously. We don’t want Mr.

70
00:14:47.740 –> 00:15:07.300
Maria Benjamin: Yeah, I was like, I’ve seen some things, but for this one, actually, I’ll kind of call it one of those more net positive examples, at early stage, in that I have seen… it was… we were suddenly getting all of these transactions from this very, like, particular area. It looked like a fraud attack, they’re all at this, like.

71
00:15:07.300 –> 00:15:30.389
Maria Benjamin: $200 range, they’re all originating from, I think it was, like, Georgia or something like that, and we couldn’t understand why suddenly we had so many people on our platform, and I remember having to make that call of, do we want to actually accept all these transactions, that are just coming in en masse, or are we going to reject them?

72
00:15:30.390 –> 00:15:54.439
Maria Benjamin: And in that moment, we… I did actually decide to accept them, and that’s because of a handful of our first few users. We actually got some additional contacts from them, and we did some research online, and we found out, oh, there was, like, some sort of payout that had happened in this particular area in Georgia. Think about, like, you get your tax return, right? And you’re like, alright, what am I gonna do with this?

73
00:15:54.610 –> 00:16:13.820
Maria Benjamin: $100! So it was essentially like that, and so we found out, oh, everybody had just found out that through online, it was an easier way to actually manage it, because it’s like a digital, kind of return, and so instead of it thinking through direct deposit, they’d use some of our

74
00:16:13.960 –> 00:16:26.420
Maria Benjamin: online resources. So that was the moment we were like, great! Something we definitely could have turned very fraudulent, very fast, ended up actually being something that we wanted to accept over time, because we did some more research.

75
00:16:27.460 –> 00:16:47.350
Jerry Hoff: you know, I had an experience like this over the weekend, and I’d love to, you know, tell you it real quickly, and it has to do with, I guess, the difference between what you’re saying here, an early-stage startup versus a more mature organization. I was trying to get some server space, so this is gonna get a little bit geeky, but I was trying to get some, you know, virtual private server space

76
00:16:47.350 –> 00:16:49.040
Jerry Hoff: From a vendor.

77
00:16:49.040 –> 00:16:58.980
Jerry Hoff: And I signed up, put my credit card in and everything else, and then it told me, hey, you need to do additional authentication of who you are, and it was like a know-your-customer type thing.

78
00:16:58.980 –> 00:17:11.100
Jerry Hoff: And then they started asking me for my passport, and this is a very well-known organization, so it wasn’t that I didn’t trust them, but I just didn’t feel like it. I’m like, you know, I just don’t feel like doing that right now. So I didn’t do it.

79
00:17:11.140 –> 00:17:26.530
Jerry Hoff: And I wound up making an account with a smaller company that had less friction. There was no friction. I just put my credit card in, I got the services that I needed, and I was off to the races. And it was just interesting that the more mature company must be dealing with unbelievable amounts of fraud.

80
00:17:26.569 –> 00:17:45.820
Jerry Hoff: But had they just let me at least get my first server running, and then maybe later, you know, as I expanded the account, maybe then, you know, asking for more information, I would have been more willing to do it. But I guess that’s the kind of tightrope, and that’s the dynamic that we see between, you know, these different size organizations. Is that right, Maria?

81
00:17:46.170 –> 00:18:11.120
Maria Benjamin: Oh yeah, and this is definitely something I think about when, I’m telling people where to put their fraud, checks, and where you’re actually going to decline somebody, is put it at the place where they’re the most motivated, which is going to be, hey, I… now I have my one server, and I need to add another. Now you’re really motivated to keep that account going, rather than being like.

82
00:18:11.120 –> 00:18:13.199
Maria Benjamin: This whole thing down and start somewhere else.

83
00:18:13.200 –> 00:18:30.399
Maria Benjamin: Similar idea of, okay, they did all the transactions, and now they want to withdraw it. You’re like, I think I want to see a passport before you withdraw all your money out of this account. You go, okay, that’s when they’re the most motivated, because who’s going to leave that money on the table, rather than the idea of.

84
00:18:30.400 –> 00:18:45.200
Maria Benjamin: there’s 5 different gambling apps. If I can’t log into DraftKings today, I’ll log into FanDuel tomorrow, or something like that. So I think that’s where you need to just think about, like, where are you putting the friction, not just what you’re introducing.

85
00:18:45.200 –> 00:18:52.599
Jerry Hoff: Exactly. I mean, you make a great point. Like, in my personal situation, it was setting up a virtual private server, and…

86
00:18:53.190 –> 00:19:08.640
Jerry Hoff: I just was doing an experiment. I wanted to see if something would work, right? So then, for me, the trade-off was, alright, submit my passport or, you know, driver’s license, take a picture of it and everything else, and send it over to them, which, you know, it’s always a little bit…

87
00:19:09.500 –> 00:19:12.330
Jerry Hoff: You know, it’s not something you want to do lightly, right?

88
00:19:13.430 –> 00:19:22.239
Jerry Hoff: just for an experiment, I’m like, no, if the experiment works, then I can understand as I build these things out. But we’re talking about, like, $8 a month. Is this really…

89
00:19:22.270 –> 00:19:29.639
Jerry Hoff: you know, at this phase, but that account would have grown, because I do so much stuff online, so they might have missed out on

90
00:19:29.670 –> 00:19:42.649
Jerry Hoff: thousands, or tens of thousands, or hundreds… I don’t know hundreds, but tens of thousands of the lifetime account value that I would have brought that was kind of overlooked. So, I guess that’s… that’s important.

91
00:19:42.730 –> 00:19:45.650
Jerry Hoff: Yep, okay, very good. Real quickly, let’s get…

92
00:19:45.650 –> 00:19:48.559
Maria Benjamin: When you’re giving customers the ick.

93
00:19:48.560 –> 00:19:49.969
Jerry Hoff: Yeah, right.

94
00:19:50.150 –> 00:19:50.770
Maria Benjamin: Yup.

95
00:19:50.770 –> 00:19:57.069
Jerry Hoff: In all fairness to that organization, I’m not going to mention who they were, I happen to have been doing all of this

96
00:19:57.070 –> 00:20:11.829
Jerry Hoff: from a airport lounge, so it probably looked a little bit strange. The IP address, like, I’m in a… I’m not in my normal city, you know, I was in some random lounge, and I’m doing this kind of work, so that probably triggered something on the fraud side.

97
00:20:11.930 –> 00:20:21.750
Jerry Hoff: So, let’s get into enterprise mid-market. Now, this is interesting, because this is different. Walk us through why enterprises have a different North Star than startups.

98
00:20:22.140 –> 00:20:29.220
Maria Benjamin: Yeah, this is what I was thinking about when everyone said chargeback, or, like, not everyone, but a lot of people said they measured chargeback. 65%.

99
00:20:29.220 –> 00:20:53.250
Maria Benjamin: Yeah, right. This is… because it’s very traditional. It’s what fraud teams usually have the most control over, it’s the thing that they’re going to most likely be accountable to, it’s going to be where they’re actually, like, I have enough control, I’m putting in the fraud rules, this is our domain for chargeback, and it’s just so tangibly measurable when you talk about the bottom line.

100
00:20:53.250 –> 00:21:18.250
Maria Benjamin: So, you’re doing finance, I’ve often reported to, chief financial officers, so they’re gonna be the ones doing the books, saying, like, yep, money in, money out this month, what’s happening here with these chargebacks. So, that’s pretty common, so who am I accountable to? It’s often chargebacks, when you’re thinking about fraud. And then, the other half, too, which makes it a little difficult, is, like.

101
00:21:18.250 –> 00:21:43.209
Maria Benjamin: why isn’t this a perfect metric, despite everybody being like, that’s what we go for, is not all chargebacks are fraud, right? Like, if somebody got cold chicken tenders, that’s not my problem. Or at least not a problem within my control. It’s my problem as someone who works for this company and wants it to do well, but it’s not something that I could actually help with. And then it doesn’t really account for leaving money at the table, right?

102
00:21:43.210 –> 00:21:51.749
Maria Benjamin: Like, a lot of people, if they don’t understand fraud well, have the misnomer that they think that zero chargebacks is the goal.

103
00:21:51.750 –> 00:22:16.699
Maria Benjamin: And I’ve heard a ton of companies tell me, like, we’re trying to get to zero. And I’m like, do you… you could make the most strong system that you’ve ever seen in your life and still get a chargeback, because somebody’s willing to lie. And the other half is you’d be leaving so much money on the table that way that there is the cost of doing business. So what your really goal is, is to think about revenue and your body.

104
00:22:16.700 –> 00:22:29.299
Maria Benjamin: line, where your profit is. That’s what you want to grow at the end of the day. When you’re implementing a fraud solution. It’s not to actually get rid of all fraud, it’s to increase the revenue at the bottom line.

105
00:22:30.660 –> 00:22:36.589
Jerry Hoff: That’s interesting. What is… so where does that… break happen between…

106
00:22:37.000 –> 00:22:43.750
Jerry Hoff: startups and smaller businesses to, like, the mid-size. When is that inflection point where, you know, you change your North Star?

107
00:22:44.150 –> 00:23:08.310
Maria Benjamin: I’m gonna say you’re thinking usually, with… you’re measuring all of these, usually, but the Northstar is going to be the one that’s gonna be the most important for your performance, and so usually chargeback gets in there that so long as you aren’t in visa remediation, or have some other processing problems, I have seen companies that are new lose their payment processor. When you’re switching over from the

108
00:23:08.310 –> 00:23:29.399
Maria Benjamin: not losing your payment processor to, like, okay, we actually want to optimize these, usually tends to be somewhere where you have enough customers that you’re proven out, right? Like, you’re not an MVP anymore, minimally viable product, you’re not just trying to get your foot… you have a foothold, and that’s when you’re now thinking about, alright.

109
00:23:29.400 –> 00:23:36.970
Maria Benjamin: Getting customers is not my problem. It means I have product-market fit. That’s when everything else becomes the problem.

110
00:23:37.310 –> 00:23:47.199
Jerry Hoff: That makes sense, that makes sense. Does this hold true as well for, like, I guess, restaurants, restaurant chains? It doesn’t matter the industry, it applies kind of universally, just depending on

111
00:23:47.550 –> 00:23:49.990
Jerry Hoff: Your level of… You’re level up?

112
00:23:49.990 –> 00:24:03.290
Maria Benjamin: I would say, actually, it really depends… you can have any different ones. I know we use these two examples here, but if you remember that slide backwards, like, there’s lots of different reasons why you might choose these. You might even not have an online business and still have

113
00:24:03.290 –> 00:24:28.280
Maria Benjamin: problems with shrinkage, or thinking about, a whole van stolen of chocolate is not a chargeback problem, but that’s certainly a business problem. So you might be thinking about other elements, like physical goods is very different, high-value goods can also be very different. If I lose one Louis Vuitton bag, that might be very detrimental. So, it changes, on your industry and your area.

114
00:24:28.280 –> 00:24:30.110
Maria Benjamin: of concentration.

115
00:24:30.590 –> 00:24:45.269
Jerry Hoff: Interesting, interesting. Well, this brings us to our second poll, and that is, what is your preferred… your preferred benchmarking model? So, Maria, I think we’ve… maybe walk us through, you know, the pros and cons of these models.

116
00:24:45.600 –> 00:25:00.050
Maria Benjamin: Yeah, so this is… now that we’ve decided to learn chargeback, what are we actually looking at here? And there is the MasterCard version, which is this month’s chargebacks over this month’s transactions.

117
00:25:00.050 –> 00:25:24.919
Maria Benjamin: Then, there’s the Visa model, which I also call the finance model, which is the… this month over last month’s transaction. So, that would be getting an idea over time. And then the last one is chargebacks tied to the actual origination of the transaction. So that means I charged back in May, but that transaction

118
00:25:24.920 –> 00:25:26.960
Maria Benjamin: is actually in December.

119
00:25:27.280 –> 00:25:52.259
Maria Benjamin: that back, and this I think of as more like the data science model, because it’s going to be way more accurate for understanding your, if what you did actually had a cause and effect, right? Like, I put in a rule and it changed, but it’s going to be very hard for you to actually ever tell anyone else your performance, because you have to wait 90 days, 180 days to kind of

120
00:25:52.260 –> 00:26:01.859
Maria Benjamin: fully bake if your solution was there. But the other two models allow you to say, alright, I did something, it looks like it had an effect, I can do something immediately.

121
00:26:01.860 –> 00:26:23.070
Maria Benjamin: That’s also kind of the reason why I prefer Visa model, but that’s just because most places I’ve worked for use Visa as… like, most credit cards are through Visa that I’ve worked on. So we’re like, we use Visa because that’s the one that we have the most often, but if all your traffic’s MasterCard, do that one. So there’s a couple different trade-offs here.

122
00:26:23.070 –> 00:26:34.780
Jerry Hoff: Well, we got the numbers, and we only have just a few minutes left, so let me just go through it. The chargeback tied to the month the transaction occurred, which is the data science model, that came in as the winner at 46%.

123
00:26:34.820 –> 00:26:48.889
Jerry Hoff: then this month’s chargeback slash last charge… last month’s chargeback, which is the Visa model, was in second place, and then the third place was apparently the MasterCard model, which is this month’s chargeback

124
00:26:49.170 –> 00:26:55.129
Jerry Hoff: Over this month’s transactions. Is that kind of normal, Maria, from what you’ve seen? That breakdown?

125
00:26:55.130 –> 00:27:17.400
Maria Benjamin: Interesting. I guess I usually see it flip-flopped, that I’ll usually see, Visa first, then I’ll see Data Science, and then I’ll see MasterCard at the end. And that’s because Visa and MasterCard don’t… it’s not really that different when you think about your measurement, but I’m a little surprised that the transaction tied back is the number one, because it takes so long

126
00:27:17.400 –> 00:27:17.870
Maria Benjamin: to bake.

127
00:27:17.870 –> 00:27:18.410
Jerry Hoff: Yeah.

128
00:27:18.410 –> 00:27:20.459
Maria Benjamin: If you know if it’s correct or not.

129
00:27:20.680 –> 00:27:30.119
Jerry Hoff: Right, very well said. Well, we only have about 90 seconds left, so real quickly, benchmarking considerations. Maybe just, tell us about Know Your Super Bowl. What is that?

130
00:27:30.120 –> 00:27:55.059
Maria Benjamin: Yeah, this is just thinking about knowing where your events happen the most. So, if you’re in iGaming, your Super Bowl is your Super Bowl, which means you might have different considerations of what’s happening then, versus… and food service industries, I’ve had a lot of people be like, Mother’s Day is our biggest transaction day, or Taco Tuesday, that fell on Cinco de Mayo is gonna be our biggest day, because

131
00:27:55.060 –> 00:28:19.860
Maria Benjamin: of all these promos, and that’s when you think about changing your fraud strategy, your risk strategy, your benchmarking is going to be very different. So… Love it. Oh, yeah. Quickly, you don’t have to trust us on this. If you have SIFT, you can actually look at our fiber benchmarking, and then there are actually MRC and community tools, which also provide some stats. If you’re like, I don’t even know what a good number looks like, these are two ways you can really understand

132
00:28:20.020 –> 00:28:22.899
Maria Benjamin: what… what am I trying to shoot for here?

133
00:28:23.200 –> 00:28:35.039
Jerry Hoff: Love it, love it. We packed a lot of information into that 30 minutes, so I want to thank everybody for joining. We had great participation. Thanks for everybody for filling in the, the anonymous poll.

134
00:28:35.090 –> 00:28:43.310
Jerry Hoff: We will be giving out a recording of this. If you want to watch it again, we’ll be sending that out, and please join us next month

135
00:28:43.310 –> 00:28:56.469
Jerry Hoff: We’ll be continuing the Blueprint series with how to produce friction without compromising fraud security. So, thanks to everybody. Maria, thank you, and looking forward to everybody joining us again next month. Have a wonderful rest of the day.

136
00:28:56.470 –> 00:28:58.190
Jerry Hoff: And… thank you all.

137
00:28:58.390 –> 00:28:59.450
Maria Benjamin: Thank you.