Wyze’s Logan Dunn on Managing Petabytes of Customer Data

Logan Dunn

That's one of our five values — create too good to be true products.

Bryan Mahoney

The real question to me was not how much data is in the data warehouse, it's how much of it is usable? Because everything that isn't usable, to me, was a liability.

Logan Dunn

Every single brand — and I kid you not, every single smart home company — has messed up in the personal data space. I think we're going to get really good at spotting truth versus fake. I compared it to, if you go on Reddit, the Redditors will rip you apart if you're not real.

Bryan Mahoney

I think it's going to be brand-accretive — if you use too much of that, people are going to say, you're lying to me.

Logan Dunn

Maintaining the integrity of that as your source of truth for customer data, and you will be so much the more powerful marketer.

Bryan Mahoney

Here we go. Thanks for tuning in to the Brilliant Commerce Podcast, where I get a chance to sit down with the brightest minds in commerce and unpack the secrets behind some of the industry's iconic and soon-to-be-iconic brands. Today I'm thrilled to be joined by one of those smart thinkers, Logan Dunn, Head of E-Commerce at Wyze. Logan, welcome to the pod.

Logan Dunn

Yeah, thank you, I'm grateful to be here, thanks for having me.

Bryan Mahoney

Yeah, we've already been warming up here a little, and I think this will be a fun conversation, we'll take it in a number of different directions. But before we get there, maybe take one, two, three, four minutes and tell the audience a bit more about who Logan is, and ultimately how you got to Wyze.

Logan Dunn

Yeah, of course. So, Head of E-Commerce at Wyze, but before that I was Head of Growth, really helped start and grow our subscription business, and before that I was Director of Product Marketing, worked on ads, creative, launched products. Before my time at Wyze, I worked in e-commerce at a packaging company called ClearBags.com.

Bryan Mahoney

Oh, yeah.

Logan Dunn

We sold packaging to small businesses, so it really felt like B2C even though it was B2 small-B. My roots — I got an MBA at BYU, worked at Fortune-100 3M, did some work with Post-it and Scotch brands on their e-commerce, and had some of my own businesses too. I did an e-commerce incubator when I was in business school and shortly after, where we'd find small companies and try to accelerate them — we even had one make it onto Shark Tank, and helped the original founder do a Kickstarter beyond that.

Logan Dunn

So my whole life I've been really interested in commerce, and that's how I got led back to e-commerce at Wyze, from helping with subscription growth. Now I lead all of our direct-to-consumer business. It's exciting — every day is exciting.

Bryan Mahoney

That's awesome, and I've obviously done my homework, I know a bit about Wyze. I've been fascinated by the smart home space for more than a decade — a fun fact, I had the opportunity to meet the founding team at a company called SmartThings back in 2011, 2012. You mentioned Kickstarter — I think they were one of the early movers in IoT and smart home, and they built this amazing Kickstarter campaign for a product they were releasing, the SmartThings Hub, which was going to be kind of the brain of your smart home, with all these devices around it.

Bryan Mahoney

When I was doing research on Wyze, I saw similarities — a company with a content story, a hardware product, but also thinking about services and subscriptions. They went on to do great things, sold to Samsung and so on. But I think Wyze is doing things a bit differently. Maybe you can take a minute or two and tell us about the origin story of Wyze, and the unique place it holds in the market.

Logan Dunn

Yeah, so Wyze was founded by ex-Amazon employees — four of them. They left Amazon and decided to create a smart home company, and what they were doing was looking for a product that seemed grossly overpriced, that we could do better at. They had a unique skill set — my boss, Dave Crosby, our CMO and one of the co-founders, has expertise in content creation, he's got like 4.3 million YouTube subscribers, so he's busking on the side with this YouTube channel while also building this billion-dollar smart home brand. Then our CEO, Yun, is the connector — he had all these contacts with factories in China.

Logan Dunn

And then our chief product officer — they just came together and said, smart home, this is ripe for disruption. At the time, seven, seven and a half years ago, smart home cameras were about $200 a piece, main players were Google and Ring, that was about it.

Bryan Mahoney

I remember, boy, those things were good, but expensive.

Logan Dunn

Yeah, exactly, and you'd get them through an ADT or somewhere like that. They were just starting to experiment with cameras at the time. So day one, our first camera launched for $20 — competition in the market was $200, and the news was like, is this too good to be true? We actually adopted that — that's one of our five values, create too good to be true products. We've tried to create an immense amount of value for the price point, and we've stuck to that. We have other values too, like being friends with our users — they actually mirror the Amazon values, we talk about Amazon's founding values all the time.

Logan Dunn

But yeah, we just want to make an affordable smart home — our slogan for a long time was 'make great technology accessible to everyone.'

Bryan Mahoney

I love that.

Logan Dunn

We recently pivoted our brand to 'never wonder' — the idea that you don't have to wonder anymore, you can know. We're a smart home plus subscription business — main competitors are Ring, Blink, Eufy, among others. Super competitive space now, but we were the first, we were the ones who disrupted the space.

Bryan Mahoney

Yeah, and talk about disruption — the brand values piece, I was going to get to it later, but let's talk about it now. I use the term iconic brands to kick off every episode, and that's where I'm most interested. Wyze has obviously disrupted with price, but you're recording data, and this is a product where your users and customers are trusting you, putting it in their homes. So how do you think about communicating those brand values as a marketer, as a way to acquire customers and earn their trust? I think that's super important in this space, in this day and age.

Bryan Mahoney

I'd be so curious to hear your thoughts, as a storyteller and marketer, on how that's helped cement your place in the market.

Logan Dunn

So every single brand — and I kid you not, every single smart home company — has messed up in the personal data space, from a PII data breach to some sort of smear on their brand, or a leak, could be people brute-force hacking passwords and usernames, whatever it might be — but you can go look, every single company will have negative news about some sort of data breach. Some are small, some larger, whatever it might be, but we're no exception to that. We've had instances where we lost customer trust with how we handled their data. On the marketing side, that's not me who was handling the data, but it's my job to tell the story and decide how we recover from this.

Logan Dunn

And even more so our CMO, who's a brilliant storyteller and leader. What we've had to do is really focus on privacy and data — we hired a security team almost two years ago now, we do pen testing, we're rigorous about conducting tests, we did a bug bounty program. We have a new privacy feature launching next month that's basically, in a nutshell, end-to-end encryption. We require two-factor authentication — when I say end-to-end encryption, I mean your videos. What makes it bigger is we're not just grabbing your email and name and address like a normal e-commerce store, we're actually capturing videos of your life in your home.

Bryan Mahoney

Right.

Logan Dunn

That creates an extra burden of trust when it comes to privacy and security. In the last two years I think we've really seen how serious customers are about privacy and security, and how seriously brands who want to do well need to think about it. I look at Apple as the gold standard of honoring privacy and security, but we're trying to take it extra seriously. From everything I wrote down preparing for this — we use Shopify for our commerce platform right now, and we build most of our connectors ourselves.

Logan Dunn

We definitely use Shopify apps, but our apps have to go through the security team for review, especially when they request PII, and they have to fill out a SaaS checklist questionnaire about privacy. We're trying to really rebuild trust with our customers, make sure they feel like, okay, I can trust Wyze with this data — not just can outside people get in, but can inside people get in, making sure internally they can't access and view your videos. That's something we have to take seriously, and I think we've done a good job of taking it seriously. I think we've done a poor job at this point of marketing it.

Logan Dunn

Part of it is making sure we have all the pieces in place so that when we're not just telling a story that's not real, we're getting ready for a PR push and launch around privacy that showcases what we've done over the past few years to build security and privacy and trust with our customers. It's huge — customers care deeply about this.

Bryan Mahoney

In the security space, and they're getting smarter and smarter all the time. It's interesting, it's not always the sexiest thing to market — it's difficult to say, I'm going to run an acquisition campaign on privacy, although we saw Apple do it, to your point about them being the gold standard — those out-of-home ads with the iPhone as a vault, your privacy in there.

Logan Dunn

Right.

Bryan Mahoney

As I think about brand equity and building trust with the customer, I appreciate the honesty and transparency — saying, we've had our issues, same as everyone else. As a customer, I like doing business with brands that use my data the same way I, as a brand, might want to use that customer's data — I want fair play. I have an expectation that brands are going to screw up, and what I look for most is how clearly they communicate that transparently — we're not trying to sweep this under the rug, here's what happened, here's what we learned, here's how we're going to get better.

Bryan Mahoney

So when I hear you talking about being out front about the investments you're making, and that privacy matters, I think that's really great — I celebrate that Wyze is thinking that way. I'd be curious — that's all first-party data, obviously you're capturing this video data. As someone thinking long and hard about marketing too, how do you think about first-party data on the acquisition, retention, retargeting front? Do you apply the same standards there?

Logan Dunn

Oh, yeah, absolutely. Ironically, I knew we were having this discussion about data today, based on what Chord does — we do these data workshops hosted by our chief of staff, she's awesome, and our data team put together, kind of like a peer session on how to use data at Wyze. It's less about security — we always address security — it's more about how you can actually utilize your data to help you in your job, target your customers, other things. In that meeting they just told us our data warehouse, with first-party data — not talking about videos, that's totally separate—

Bryan Mahoney

Yeah, that's separate, yep.

Logan Dunn

Yeah, just customer data, like what you'd put into a CDP — I'm not uploading videos into a CDP, I'm putting customer data into the CDP, from events, what they're doing inside the app, all that. They just told us we have a petabyte of data in our warehouse, in our lake — this sea of data to wade through. So the problem doesn't become, can we target our customers, it becomes, how do we target our customers, just because there's so much. One thing we've tried to do over the last year, led by our CEO through our chief of staff, is create a more data-driven organization.

Logan Dunn

As a tech startup we've always been pretty data-driven, but we've been pushing hard on how to make data more accessible to employees, safely. One really cool thing is we have an AI tool — this is a buzz thing now, but I think every company will end up having this: what's your AI productivity tool? You've got your normal productivity suite — Teams or Slack, combined with Google Drive or Microsoft Office — and you've got Asana or task tracking. Well, I guarantee you'll see more and more of this at companies.

Logan Dunn

What's your AI productivity tool that helps you access all your data across everything, more quickly? Because you've got files you're using every day for business, and warehouses of data, and then, how do you do it responsibly. We use a tool called Glean.

Bryan Mahoney

Yeah.

Logan Dunn

We're hooking everything up to it, but from the security standpoint, before we could hook up any Git repos or Confluence, Atlassian stuff, our security team had to scour it for any keys, private or public, making sure nothing's there, so this AI can't go through documents and see those keys. But that's really cool to me, and it gets even cooler when you hook it up to your customer data pipeline. SQL has, for the longest time, been how you access your data — blending tables, getting it right. But I believe one of the big things for marketers—

Logan Dunn

—if you're a data professional, you're like, I have no problem with SQL, I can do this forever. But if you're a marketing professional looking at how to access and use your data, this is one of the big changes we'll see with generative AI — productivity tools accessing data for you. You already see it with Tableau, they've got an AI assistant, and the more interesting AI assistants are the ones understanding data. We've built a little inventory module where you can search product SKUs and the gen AI—

Logan Dunn

—just searches through our database to tell us the dims, the quantity, right now. We've got our whole inventory pipeline hooked up to this AI tool, we're hooking up our sales pipeline, and pretty soon forecasting tools too. That's where the cool stuff happens with data and generative AI — that's the fun stuff to talk about. But we still have the fundamental basics — this huge lake, our Kafka tool converting into downstream things, building cohorts out of it.

Logan Dunn

We've got machine learning on top of it, propensity models helping us determine, based on thousands of actions a customer's taking, which customer groups they're in — are they ready for subscription or not. We're doing as much as we can. The question isn't what do we do, it's — there's too much to do. So it is, I guess—

Bryan Mahoney

Where do you start? Yeah, where are the high-leverage things you can act on, the key insights that move the business in the right direction? It's funny, you said a petabyte of data, and we were joking about point solutions and the proliferation of the modern tech stack. It must have been a decade ago when data lakes and warehouses were becoming popular in commerce, and I used to talk with other technology commerce leaders, and it was almost a competition about who had more data in the warehouse — do you know how big it was?

Logan Dunn

I was like—

Bryan Mahoney

And I used to find it funny, because the real question to me was not how much data is in the data warehouse, it's how much of it is usable. Because everything that isn't usable, to me, was a liability — I'd rather not have it. One of the things I'm really excited about today, and not to talk too much about Chord, but we bring all of a brand's data together in one place, keep it secure, organize it, make it ready. I haven't seen Glean before, but I'm going to check it out — there's an awful lot of similarities with the approach we advocate for. We're working on a copilot, like a lot of folks are.

Bryan Mahoney

The reason I think our copilot is really exciting is it has access to all this data, and it's really neatly organized.

Logan Dunn

And if you want generative AI, you need more data, because that's where AI is powerful — making sense of a petabyte of data. You're exactly right about the problem — sorry to cut you off, but I get so excited about this — that is the problem, it's a liability with all of this, because I guarantee there's so much duplicate data in there too, because it's in this view or that view, or a whole copy of another table doing this thing. That's where you have an advantage over a copilot assistant inside, say, Teams with their branded copilot — they're limited in scope.

Bryan Mahoney

Yep.

Logan Dunn

To what someone who's built on a CDP has — and then your challenge is, how do I get the whole pie, not just a sliver of what the marketers want.

Bryan Mahoney

And that's why we're more than a CDP — we figured out it can't just be the CDP data, we need more data to deliver value. But then also, how do we make sure we have the connective tissue out — everyone loves to say AI, but a close second in buzzworthy terms is agentic workflows. I think a lot about, I don't just want an agent that can help me make sense of data, I want an agent I can ask to perform tasks on my behalf. That's what we've been working really hard on at Chord. You brought up Tableau, and they've got this chat interface now, and so do most BI tools.

Bryan Mahoney

But all of that value presupposes that the data in your warehouse is not a liability, that it's organized and high quality, which, of course, we're talking about that.

Logan Dunn

There's no one — in fact, let me see what I wrote. You asked a question, I can't even remember — oh, yeah, you said, many brands struggle with complexity, what's the key to making brilliant commerce feel effortless? And what I wrote was, know that you're not the only one struggling, have patience with yourself, because some might look like they have everything in order — which is so true. You think, my house is in order, comparing yourself to a neighbor or something. But really everyone is in chaos, because data is proliferating at a pace we cannot keep up with, and no one is good at it.

Logan Dunn

They'll say, we're good at this or that, but for every good thing, there's still a hot mess somewhere else, because you can't do it all. That's really where generative AI, I think, is going to be so powerful. I also wanted to add — where I see generative AI right now, let's look at Tesla for a second, the level one through five of self-driving vehicles.

Bryan Mahoney

Yeah.

Logan Dunn

Tesla came in and invented level two and three self-driving automation, where level one is adaptive cruise control.

Bryan Mahoney

Yep.

Logan Dunn

Where Teslas are right now is level three — you can put in a destination and it'll get you there, which is pretty advanced and still not widely adopted.

Bryan Mahoney

Depends how brave you are.

Logan Dunn

Correct. So to me, we still have level four and level five — level five being no one drives anymore, the seats face each other, you just tell it where to go and it gets you there. We're a long way from that, but making rapid strides. If I apply that to how we use generative AI tools, we're at like level one or two automation for generative AI, and I say that because no one can really break out of interacting with AI through a chat interface.

Bryan Mahoney

Yeah.

Logan Dunn

Generative AI becomes really powerful on all this data when it's doing things for you, presenting you insights without you asking, running processes that are actually useful — not just you interacting through a chat prompt. That's why, before you started recording, we were talking about this idea — I'd love to build a commerce platform on top of AI, rather than the commerce platform with AI on top of it. Imagine the AI serves the page to the customer, takes the credit card, the billing, and then the AI processes your payment because it's connected to a payment processor like Stripe.

Logan Dunn

And then the AI generates an email on the fly and sends it — hey, guess what, I know, because I've got this customer in the CDP full of info about what you've bought in the past, and I know my product models and what's frequently bought together, and I'm just going to build that email on the fly. That's when we start getting really cool commerce automation — we've got to get out of that chat-interface way of interacting with generative AI, but it's coming, for sure.

Bryan Mahoney

I'm excited to show you what we're working on. There are two things I want to make sure we cover based on what you've said, I don't even know where to start, but I'm going to stay on this autopilot thread because I think you're onto something. I hear the excitement in your voice, and as a technologist, yeah, that's super cool, I'd love to figure that out. But the brand purist in me, or the consumer, I should say — I want to still shop. I live about an hour outside Los Angeles, and when I have to go into the city, it's a huge pain — those are the times I think, give me level four or five.

Bryan Mahoney

I don't even want to think about it, I want to close my eyes and just be there. But then there are other times I love to drive, I enjoy it, I don't want anyone driving for me, I want to drive myself. My belief is we're going to see this bifurcation in shopping, where I'll be fine with an agent doing some things for me, for commodities I know I need — but that should leave me more time to discover brands I love, and that's where I want my hands on the wheel. I'd love your point of view — do you agree, or do you think I'm crazy?

Logan Dunn

No, no, I completely agree, because we're missing this whole side of commerce right now. I'm a marketer by trade, that's what I'm trained in, that's where my career's mostly been, even as Head of E-Commerce — I own the direct-to-consumer P&L, determining everything from sales generation to cost cutting.

Bryan Mahoney

Yep.

Logan Dunn

The AI thing is on this operational back-half, in my opinion — I agree there's a marketing-solution side, like sending email, putting products in, but to me that's still operational marketing. We're missing this whole other huge trend right now, which is the growth in social commerce. It's only days away before TikTok is bought, sold, closed down, exploded, who knows.

Bryan Mahoney

Yeah.

Logan Dunn

But there's no changing — one of your questions asked about trends in commerce, what trend is going to stick, and I really think — I've been drinking the Kool-Aid from my CEO, part of this is his opinion that's become mine — but I really think social commerce is coming, and we have a few hurdles to overcome in the U.S. economy before it can be as big as outside the U.S. — payment hurdles and other issues. But generative AI, we're seeing a lot of fake content come out too.

Bryan Mahoney

Yeah.

Logan Dunn

I wrote this while preparing for this podcast — I think we're going to get really good at spotting truth versus fake. I compared it to Reddit — the Redditors will rip you apart if you're not real.

Bryan Mahoney

I think it's going to be brand-accretive — if you use too much of that, people will think, you're lying to me, I know a real marketer that knows me and actually loves telling stories. So, sorry, go on.

Logan Dunn

I agree with you — this is what you're talking about, the bifurcation. You've got your AI doing tasks that need to get done, marketing's involved in that, block-and-tackling stuff—

Bryan Mahoney

I don't necessarily need to do, yep.

Logan Dunn

Yeah, stuff that's still good, that customers expect — you want it to get you the milk and eggs, so to speak, the stuff you know you're going to get. But over here you've got to have real, creative, awesome ways of interacting with customers, and what's exploding right now is social commerce — that's what I think is really here to stay, along with the fundamentals of a really solid brand message, sticking up for what's important to your customer base.

Bryan Mahoney

Yes.

Logan Dunn

One of the challenges I think we're facing at Wyze is, we came in at 10% of the market's price — $20 versus $200. That was our value prop, same tech, way better price. Now it's a bloodbath in the sub-$50 camera market — not even a red ocean, a straight blood ocean. What's happening is we have to sit here and ask, what's our actual value proposition — we can't just say value is our value proposition.

Bryan Mahoney

Yeah, right.

Logan Dunn

So that's why we're trying to beef up our privacy and security standards, creating end-to-end encryption of all your videos, so only you as an authenticated user can see them, and every time you look, it's authenticated again with a new key — because that's part of it. Part of it is making sure you're creating tools, from a software perspective, that reach out to your customers — we've created descriptive alerts using the Gemini model right now to describe what it's actually seeing, and it's getting so good — reducing notifications. What is your actual value prop that you're trying to add for customers?

Logan Dunn

That's foundational to marketing and commerce, and creates your love group — and then you've got to foster and nurture that love group like crazy. None of that will change, and that's where the bifurcation will happen. We're seeing more and more influencers helping accelerate getting your message out, creating trust quickly, and people buying on these social platforms. But that will always have to exist.

Bryan Mahoney

Yep, I agree, I think that's a great take. We only have a couple minutes left — I really want your take on this. You said this idea, too good to be true, let's create too good to be true products — and I don't know about you, but the first time I used Gen AI when it was really starting to develop, it felt too good to be true. It reminds me of, ten years ago, acquiring customers on Meta, or at the time Facebook — too good to be true, and we saw what happened there, we gave them a lot of data that became kind of weaponized.

Bryan Mahoney

I'm on the record saying I'm a little worried the same thing happens here — you're sitting on so much data, a petabyte, we talked about it. How are you thinking about transforming that data, or keeping it safe, so it isn't too good to be true — so you can get the most out of models like Gemini or what OpenAI makes available, without giving them all your data? Is that something you and the team are working on?

Logan Dunn

Yes. We have two sides of data — we have the CDP, but we also have a lot of insights we can garner from all these videos being captured. We don't sell any of those, but there's a ton of video capture data we can gain insights from and train our AI models on, and that's worth a ton of money. But I want to go back to something — having been on Shopify now for about three years, a little over three, we were on Magento before that — I'm just shocked at how crazy easy it is to give away your customer data. It's scary, if you think about how big Shopify is.

Logan Dunn

Every single app comes in, and Shopify limits their liability by showing you a warning that says, hey, you're giving this app access to your customer list, see your products — and that's mostly fine, that's public. But your customer list, shipping address, email — that's not public, that's PII.

Bryan Mahoney

Yep.

Logan Dunn

And so many people, especially small businesses, are just hitting approve, approve — I've seen Shopify sites, really small sites, with hundreds of apps.

Bryan Mahoney

Yeah.

Logan Dunn

All of which have been given access to that customer data. This is a real problem — how are we not just giving our customers' data away? You're not selling it, technically your privacy policy is accurate, but it's just scary. That's why our security team says, you can't even do those without going through security, checking the safety of what they're offering for their data — and why most of the time we're going to actually build that ourselves. I do kind of love the original Chord idea of headless commerce built on your CDP, because you're keeping more things in a central area, so you're not needing to proliferate applications with all your customer data everywhere.

Bryan Mahoney

We really believe in this idea of build then buy — there's so much tension, and I know you're kind of on the record about simplifying technology stacks, and I love this. We could talk for another two hours, but I believe more brands win when they can buy a platform they can build around, and there are parts where you don't have to become a full-fledged technology company. Wyze for sure is, and it makes perfect sense that you are, so you have engineering talent — but how do you strike the right balance between speed to market, innovation, the investments you're making, and security? This has been a big eye-opener for me.

Bryan Mahoney

How Wyze really thinks about customer data privacy — really, I guess all your data is customer data, and it's a first-class citizen within your organization. It makes me feel really good about the way you're building product and shipping services into the world.

Logan Dunn

Yeah, and there are so many ways we could continue this conversation, but one of the big things, when it comes to using the most of your resources, is I'd just start really early thinking about building to scale. The biggest thing with customer data — and I truly mean this, because I've had so many headaches — is I hate when the source of truth for this customer data is over here, and this source of truth is over here, and it's almost unavoidable, because your email system has its data, and—

Logan Dunn

—that's why we've pulled back into just a lake. I wish we would have done this from day one — just one location, one table that's our source of truth for our entire company, for everything about your customers. You're thinking, yeah, like a CDP, right? Yes, but it's really hard to bolt on a CDP after your company's already established and make it the central source of truth, because you've got all these teams doing their own thing with their databases here and there. If I could go back to early-day Logan at Wyze, I would just shake him and say, get a central location right now for all of your customer data.

Logan Dunn

I'd say, just put it in a table in Snowflake or something, for right now, and make everyone put it there, and focus on — don't proceed down this lane without making sure you can link identities back to that central location, and write to that central location. You can just do so much more if you're putting it all together in one place, and starting really early with that foundational piece. I can't tell you how much identity resolution I've had to do between Shopify IDs and Braze IDs and all these different IDs everywhere — email is no longer a source of truth for the customer. You've got to figure out how to stitch it all together.

Bryan Mahoney

Yeah.

Logan Dunn

So, if you're looking for the easiest win of all in executing on your customer data, it's that — have a single source of truth. Because it's so much easier to go create a cohort, analyze your cohort, especially with AI, because bolting AI on top of it — that's why you're going to see an AI productivity tool coming to every organization, because that tool is going to hook into all your data sources, and your AI is going to be your access. But early on, especially because those are going to be expensive—

Logan Dunn

—especially early on, the best you can do is get it together into a single source of truth, and just religiously, zealously focus on maintaining the integrity of that as your source of truth for customer data, and you will be so much the more powerful marketer.

Bryan Mahoney

Yeah, everyone who knows me knows I can't resist a good pun, and those are exceptionally wise words to leave this episode on — invest early, invest often in a single source of truth. And if you're a company that's seven years old, pretend it's day zero and do that, because in order to take advantage, to have the level of excitement Logan clearly has for AI, AI is only as good as the data you can give it, and you're really holding yourself back if that data lives across a hundred different systems. Investing in that single source of truth gives you optionality as we figure out where the real value is going to lie in AI. This has been an amazing conversation.

Bryan Mahoney

As promised, Logan, I already want to invite you back for version two and version three.

Logan Dunn

I'd love to, I could talk about this all the time, so — cool.

Bryan Mahoney

Well, I'll wrap us up there, so we have more to talk about next time. Thanks for being on the pod, looking forward to keeping it going.