Latest / Talk Commerce / Live from Shoptalk: How Signifyd Is Redefining E-Commerce Fraud Protection Beyond the Checkout
Transcript
- [00:00.2] Hello and welcome to a live episode of Talk Commerce here at Shop Talk in Las Vegas. And I'm here with Nicole from Signifyd. Nicole, tell us a little bit about your company and what it is that you do. Sure, yeah. No, thanks a lot for having me. So Signifyd is think of ourselves as an E commerce protection platform. [00:18.3] And what that means is we stop the bad guys who are trying to buy things at people's stores. And when we do that, when a lot of people do that, you stop one bad guy and you'll stop like 10 good guys. So it could be you when you get declined, when you're trying to check out. And so we've got machine learning models that help us figure out exactly who is trying to abuse some kind of a, promotion or actually stole someone's credit card and is trying to make a purchase. [00:43.2] And we work with all these merchants to stop those transactions from happening. So that's our core really, is to protect those transactions. What we've really been working on lately is growing beyond the transaction. And, and that means when you go to create an account online, when you go to like, add a credit card to your account online, you go to log in all the way through to purchase. [01:03.1] And then if you go to make a return, making sure that your best customers, you can give great return experiences and stop the guys who are trying to ship rocks back on the returns. Okay, there's a lot to dig into here, but I think let's start with letting the good guys through because I think that's an experience that most consumers can relate to at some point. [01:22.1] Having an annoying captcha or having your card declined. Tell us how you kind of filter out the good from the bad and what that means to merchants. It's a great, question. So I'd say we have like three layers of magic. So the first layer of magic is that we work with a ton of customers. [01:37.3] So we work with Lenovo, Walmart, Vuori, Aloe, big companies and small companies, and we collect a ton of data. So, so that's kind of first layer of magic is just this consortium model so that when we see a bad guy try to do a transaction at one merchant, we're able to stop it. [01:52.7] Or we see a. We've seen a lot of really sophisticated attacks where someone goes and buys a bunch of stolen credit cards online and then goes and tries to, like, we call it a bin attack. They'll just go and try to hit it up a bunch of credit card numbers. So we have this consortium model that's kind of layer one. Layer two is the data that we collect across those merchants. [02:09.8] We know you're. And it's creepy, but it's for fraud, so it's not super creepy. So we know your device, we know your, you know where you're sitting. We know that it took you an hour to build this thousand dollars cart or it took you like five seconds to build $1,000 cart. We know that we've seen this email address with this payment credential before. [02:26.6] So we collect a lot of data. That's kind of miracle number one. Or miracle number one is like the consortium. Number two is all the data that we collect. The hardest part of data is actually doing something with it. So we have machine learning models that take all that data, spin it up on every transaction in less than 500 milliseconds and decide if it's, if we, if we see enough about the transaction that we think it's good or there's some kind of a flag that we think we should, we should stop it. [02:51.2] Cool. And what does this mean for these kinds of merchants? Like could you tell us maybe the impact that we had on Lenovo or on one of these companies from implementing the solutions you offer? So customers come to us usually with one or, or two problems. One is they have a chargeback problem. [03:06.4] So that means too much bad's getting through and customers are calling and you're like, hey, that wasn't my purchase, I didn't buy $1,000 laptop. So there's a chargeback problem. Or, and, or they have a problem where they're not getting enough approved. So because they're trying to catch all these bad transactions, they don't have a high enough approval rate and which means that they're over declining with Lenovo, I'd say when they came to us they had a bit of both. [03:26.8] And what we've done for them is really reduce their fraud problem and allow them to increase their sales by 4 to 5%, which is pretty meaningful to have that kind of top line growth. And a lot of companies are shocked when they come to a fraud provider. But we can not only help them save money, but actually drive revenue. [03:43.7] And what I think it ultimately means to people walking around shop talk is that they can create all these great experiences and the bad guys don't ruin em. Yeah. Do you have any new features coming out or recently released that you're excited to share? Sure. So I don't think you could do a podcast here without the word agentic. [03:59.5] Right. Like I think I think you, we have to say, we have to. You know, it used to be that we could just stop the bots. Right. Like, a bot was bad. Yeah. And that's like captcha like you talked about. But not all bots are bad now. And so now it's part of figuring out like, who are the good bots that you might want on your site? [04:16.3] Who Nicole has sent on my behalf to go shop at your site? Versus the bad guy is doing things. So, you know, we're doing a lot of work around the protocols, around making sure that for our customers we can create those good experiences. And you know, as they're trying to go out and create discoverability using agentic or personalization using agentic or even a full checkout experience that we're not gonna stop it. [04:38.8] Yeah, because that'd be the worst. Right. Like you create this awesome personalized experience. I'm gonna buy this shirt from this brand I've never met before and then I get declined. Yeah. So that's what we're, you know, when we think about protect, we think about protecting good experiences. Yeah, Very cool. Well, to follow with this AI theme, my next question was gonna be, what's the buzz here at Shop Talk? [04:57.5] What are people asking you about? And, I suspect that may be part of the answer. Yeah. And I think I, I'd almost say we're almost through, like maybe the first hype cycle on agentic that everyone's like, oh no, the agents are coming. Right. We gotta prepare the checkout. And I think, I think where we're seeing a lot of agentic is in discovery. [05:13.5] Right. I'm sure you're using it. I'm using it. I'm using it to help me find, you know, make decisions, become educated about products, learn about products. I've not sent an agent to buy something on my behalf yet. And so, you know, I think a lot of that discussion is here saying, okay, we've kind of got to that first really big hype and the agent, you know, it's all coming and we're kind of settling in to like, what's it actually going to be like? [05:34.1] And how are we going to set our sights up to be discoverable and for agents to tell the right story of our brand through their platforms. So I think that to me is what we're hearing and talking a lot about. Yeah. So a lot of our audience here at Tok Commerce is maybe small and medium businesses, companies that aren't quite to Lenovo level yet. [05:54.0] Part of the like Growing up process for them is starting to think seriously about things like fraud mitigation and, you know, security. And for the new customers who are just getting into, you know, looking into getting serious about this, what, what would you say to them? What's your message to. [06:10.0] To that kind of business owner or, or employee? Great question. I talked to a woman yesterday and I was like, she's a smaller company. They're doing, I think, getting close to $50 million a year in revenue, which, you know, they're rapidly growing. And I was like, who does your fraud? Like, who. How do you manage fraud? She said, katie. Yeah. [06:25.0] And there's a woman, Katie, who's literally reviewing suspect transactions and doing their fraud. And so at some point, Katie's great. Right. But at some point, Katie is like, as you're growing, it's gonna. You're gonna have to hire a lot more Katies. Katie's not gonna get to sleep. And so I think that's where, you know, we work with a lot of companies. [06:42.3] We worked with Vuorian Aloe before they were Vuorian Aloe. Right. I mean, they've always been Biori and Alo, but they were much smaller. And so as they're scaling, you can't, you can't just throw people at the problem. Yeah. One, it's not as efficient. Obviously you're gonna spend a lot of money. But two, humans are awesome. [06:58.8] But our machines are pretty great to be able to figure out who those good and bad transactions are when the bad guys are using bots. Exactly. Yeah. They're coming at you. They're not a small shop coming at you with less sophistication. They're going to come at you with their full force. So we have a plugin with companies like Shopify, Woocommerce, Magento. [07:17.2] So we make it really easy for these companies as they're growing and scaling to be able to plug our services in. Yeah. And let's dive into returns a little bit more. Could you tell us about how signifyd helps with. With that process and what that means for merchants? Sure. So, as you know, the bad guys, like, they don't just go away. [07:34.3] Right. So, like, as you, as you start to stop someone at the checkout, they find new and clever ways to defraud a company. And so returns, as we were talking to our customers and saying, you know, what are some of the pain points that you have is returns? And returns is really a moment of truth with a customer. [07:50.7] Right. We all have something that doesn't fit. It's the wrong size. It was a different color than I wanted. And when you ship something back, it really is a moment of truth of how. How much do I love this brand? How much am I willing to shop here again? And, And so merchants are wanting to offer, like, instant refunds. [08:07.9] What a great experience, right? Like, hey, these jeans didn't fit. I'm going to send them back. Great, here's your hundred dollars. And I've got it instantly. And what am I probably going to do? Maybe go buy something on their store as opposed to I ship it in three to five days later, after they scan it in, I get a refund. And so, plus, merchants are wanting to offer these instant refunds, but are scared of the guys who are going to ship rocks in the box. [08:28.5] Right. Or their old jeans. On the return side, we are offering what we call instant refunds. And so allowing our customers to figure out who are those good customers that they want to have those really great experiences. And then, you know, Isaac comes in and you're like, I don't know about Isaac. Like, he's shipped rocks in the box over at this merchant. [08:46.1] So I think you should wait till you actually see the merchandise to give him his refund. Okay, great. So if I have a good track record of, you know, I've done a couple successful returns in the past, you can help give me my money back immediately. If I haven't yet, then maybe it'll have to build that reputation to protect the merchant and improve the customer experience along the way. [09:05.0] So I think we're really trying to help our merchants find their best customers and give them really great experiences the first time. Not just like, oh, this is Nicole. I've seen her like 20,000 times. I know I can do this. It's. Oh, geez, I've never seen Nicole. How should I treat her? Right, because you have that data from other. [09:21.1] We've seen her. Yeah. You've seen her. Yep. Awesome. Well, that seems like a pretty nifty feature. And I think flows are an especially good example where returns are super important. Okay, so a couple of the kind of stock questions I like to do as we're wrapping up. One is do you have any hot takes about the industry or any big opinions? [09:41.2] Hot takes on the industry? I don't know, like, my background, I'm an engineer and a shopper. So I feel like that's kind of like why this is the perfect role for me. And working here, you know, I'm excited to see just, like, all the personalization that. That I think that's one of my hot takes, is just what AI and technology can do to drive personalization, not only in my shopping experience, but my marketing experience. [10:04.3] So I'm excited to see how, how we really can bring that together, even in how we talk to our customers at Scala. So I'm excited about that. The other one, as we talked about with Agentic, is just, you know, I, I'm excited to see how it plays out. Like I said, I think discoverability and using LLMs to help us get more information faster and more relevant information faster. [10:25.9] I think the checkout's gonna take a minute. Yeah. For me to like trust an agent with my card and say, yeah, just go find me these black pants when you see them at a good price. Yeah, I would agree with that. Definitely a little nerve wracking to give up control of the actual. I don't even wanna give my card to my kids. Yeah. [10:41.4] Let alone a box. Yeah. Okay. So another one of my questions, I like to give people a chance to just kind of rifon is what's your shameless plug like? What's your elevator pitch for? Why someone should use Signifyd. Oh my goodness, that's a hard one. Why Someone should use Signifyd. I think it's, it really is unique that a company can help you save cost and grow revenue. [11:00.0] Yeah. And we, more often than not, we'll do that for a customer to help them actually stop, you know, save money on their bottom line, grow their revenue and allow them to create these great experiences without having to worry about the bad guys. Mm. Anything else that you're especially excited to share with our audience or to promote? [11:18.5] You know, I think Signifyd, really our, our last couple of years has been about growing beyond the checkout and just this idea of how powerful it is when I can see you the first time you create an account, to, you know, logging in, to modifying your account, to buying things, to returning things, it creates this really commerce platform that can help our merchants offer great experiences across the lifecycle.