Austin Hankwitz (00:00):
Hey everyone, I'm Austin Hankwitz.
Katie Perry (00:02):
And I'm Katie Perry. And this is After Earnings, the show from Morning Brew and Stakeholder Labs that brings investors up close and personal with the executives behind the world's most interesting companies.
Austin Hankwitz (00:13):
And today we're talking with Bill Magnuson, the CEO and founder of braise. Braise is a well-known customer engagement platform that allows the companies you all know and love to best acquire and retain new customers. I'm talking about KFC, burger King, FanDuel, chime, Etsy, and countless other consumer facing brands. We dig deep into the evolution of their product suite over the last 13 years, their recent path toward profitability as well as how they're leveraging artificial intelligence in 2024 and beyond.
Katie Perry (00:44):
And as Austin said, Bill's a co-founder of Braze and was the company's first CTO. So it was really interesting to get his perspective on AI in particular. We also got into the science of when you get that email or push notification from a brand you love and why something that could seem like a black box for the average person. So let's get into it. Phil, welcome to the show. You're fresh off of reporting your Q4 and year end earnings and we got a lot of ground to cover. Thank you so much for being here.
Bill Magnuson (01:13):
Absolutely. It's great to be here with you.
Katie Perry (01:15):
So we want to dive right in. Starting with, let's set some context. The state of marketing as it exists today and where Braze fits you are a customer engagement platform. I think marketers are very familiar with that sort of verbiage, but how would you break that down in terms of the value you're serving to your customers and what your product does at the end of the day for someone who's not as inside of the marketing world?
Bill Magnuson (01:41):
It's really great to hear you say that you feel marketers are generally familiar with the customer engagement category. When we IPOed a little bit over two years ago, I think we were the first company to ever IPO and describe ourselves as a customer engagement platform. And when we talk about customer engagement, it includes a lot of activities, many of which are sending messages, collecting data about customers, using that data to understand them better and drive additional relevance. And I'm excited to talk more about what I refer to as the craft of customer engagement as we dive deeper into this conversation because I think that what characterizes customer engagement as separate from merely the activity of sending marketing messages to people is really about being focused on long-term goals on doing so in a way that's agile, that's integrated into the product cycle that's respectful to the customer and really understands that the customer is in control of their journey.
(02:33)
But we'll get into all that a little bit later. Just to kind of start with the brass tacks of it all, Braze is a customer engagement platform as you said, and what we literally do is that we work with usually B2C brands and we integrate into their usually digital products and services, we'll integrate into their mobile apps, their websites, connected fitness products, connected tv, et cetera. What we do when we integrate into them is two things. The first one is that we're there in the background paying attention to what the customer is doing, what are the parts of your product or service that they're most interested in, where are they exploring around what leads to them being more engaged or less engaged, those kinds of things. So that we're trying to learn their affinities and their interests and then we open up the ability to message them.
(03:21)
And so within a mobile app, we might be able to deliver a survey, we might have a slide up that shows up, we'll have a content stream that can live persistently in the application. We can dispatch push notifications, we can send SMS, we can send email, we can deliver WhatsApp messages or really a whole bevy of messaging options across all of these different touch points that the customer has. Same thing on the web, same thing if we're on a connected fitness product or we're on a smart TV or what have you. And we have customers that span across all different verticals because ultimately what we're trying to do is something that all businesses should do, which is that if you want to build a good durable, long-term business, you should invest in having strong relationships with your customers. And there's a very human fundamental there, which is that you want to build a strong relationship with someone, you should strive to understand them better as you interact with them and use that understanding to make it so that the interactions that you have are more valuable to them so that you build a stronger bond over time.
(04:21)
And what we're really trying to do is help B2C brands achieve that very human goal of building strong relationships or deeper understanding but at a massive scale. And so we can get into some more scale numbers later, but Braze actually on behalf of our customers is interacting with over 6 billion monthly active users across all the different brands that we work with. And it's not uncommon for an individual brand, especially today given how global the app stores are, how smartphones have actually deployed as a technology more widely than any technology that's come before them, including even the ability to read what's on the phone, which is just an incredible testament to their accessibility and their reach. And of course what that's done is it's created opportunity for digital businesses in particular to have massive global audiences. When an audience is at global, though it's diverse, it has a lot of different personas in it, it has a lot of different cultures and languages. And so that challenge of being able to build a strong connection with them is one that requires a very sophisticated approach. And that's why I started this by saying the craft of customer engagement is distinct from merely the activity of sending messages. And it's really that tension of the opportunity of massive global scale but actually servicing it because of how diverse and complicated it is.
Katie Perry (05:41):
Just thinking about the role of technology in advertising, back in the day it was mass communication and it was sort of the spray and pray approach and basically what you built it sounds like is a way to do more targeted communication that's personalized but not lose the scale. You mentioned 6 billion people, so I'm curious in your view, the company's been around since 2011. What did you see happening in the marketing consumer landscape that you felt the need to build something different and what's, what trends are driving the momentum for your business currently?
Bill Magnuson (06:20):
So it's kind of interesting to go back to our history in 2011 because we didn't actually start out knowing that we were going to become a marketing company. The original vision was really predicated on the change that we saw coming because of mobile. And so there were two parts of it. First was that we felt that huge businesses would be bored and built to be mobile first and second that the wide scale adoption of mobile by consumers would lead to transformation in the enterprise in terms of how they did business and how they interacted with customers. And then what we wanted to do was work on a problem space that we felt the advent of mobile and the spread of it would fundamentally change and create new opportunities in. And so if we actually think about not just the technology like ours that has created the ability to personalize communication, the fact that everyone carries around a device with them that they have a very kind of intimate connection with means that you actually have the opportunity to understand every individual customer and to communicate with them as an individual in a way that you didn't before.
(07:23)
And so I think that's a really great testament or another example of that tension of the opportunity that exists because you have the ability to communicate with someone wherever they are at any time. You can beam your message right into their pocket or to their phone sitting on the table or what have you. But of course there's a responsibility for that because it is a private device, it's one that can permeate every aspect of your life. And so if you're going to be sending a message, it needs to be relevant. And I remember in the early days having the flashback of remembering telemarketers calling when you were trying to sit down to dinner with your family, which all of us above a certain age remember and how interruptive that was in your life and just how terrible it was. And I remember talking to customers in the early days, which was like if you are being given permission to buzz your customer's phone with a push notification or with sending them an SMS or what have you need to take that permission seriously, you've been given a right, but there's also a responsibility there.
(08:26)
And so to your point about mass undifferentiated communication, if I'm driving down the highway and I drive by a billboard that isn't personalized to me, I'm not offended by that. And similarly, if I'm watching a broadcast like an NFL game on a Sunday on CBS over the air or what have you, that's something that I'm going to expect that advertising to be targeted to the general demographic that's watching it and it's not personalized. But even to take that TV example, as soon as I'm streaming that when I start get advertising that's not personalized to me, I start to get annoyed by it. I'm like, come on, I'm streaming a little bit about who I am. Let's make sure these ads are at least somewhat relevant. And then if we take that further and you say, alright, what about when I'm getting an email or a push notification that's coming into my inbox, the push notification is actually buzzing my phone, my bar for how relevant that needs to be to me is even higher before I turn it off.
(09:19)
And now let's jump one step further. When you're literally in the product experience, you're in the mobile app or you're on the connected fitness product or what have you while you're actually using it, that not only needs to be relevant and personalized, it needs to actually fit into the product experience. And so as we get deeper into the first party experience, the bar for what a consumer actually expects continues to get higher, but of course our ability to drive impact also gets higher. And so it's a great connection point between the opportunity that exists but also the expectation from the customer that you need to get over the bar of in order to have the writer, the permission to continue to communicate with them there. And that's where the sophistication need comes into play.
Austin Hankwitz (10:05):
I think it makes a ton of sense, bill. And I think what's so cool about Braze in general is when I think of a technology company, your customers generally speaking are boring companies that no one's ever heard of, but you are working with companies that everyday people know and love. I know you guys just announced you're working with DraftKings and FanDuel. Can you kind of break down for us how they are using your suite of products to not just acquire new customers but also engage those customers consistently to bet more on their platforms?
Bill Magnuson (10:33):
Yeah, I mean I think there are really great examples of why real time matters. Braze actually was built as an event driven stream processor and we can maybe dig into that story a little bit more, but I actually most recently had come from the hedge fund industry when we started this company. And as I mentioned when we first started, we didn't know that we were building a product that would eventually be used by marketing teams. We were trying to build something to allow mobile app owners to be able to build stronger connections with their customers. And if you think about that problem space, what we're trying to do is take data as it's being generated. So someone is actively using your product and as they're navigating through the product, they're taking certain actions and the context around them is evolving. You want to be able to respond to that evolving context in the moment.
(11:20)
And unlike even dispatching an email to someone where you might go to your inbox and refresh it and if it takes five seconds, you're probably blaming Gmail or your internet connection or whatever anyway. If something stutters in a product experience for even a couple of seconds, you probably have already assumed it's frozen and you've, you've flicked it away. And so that performance requirement was super important for us and therefore we actually built our backend not to use the batch processing and relational databases that basically everything in the marketing technology landscape was doing, but we instead used what at the times brand new technology of an event-driven stream processing engine. And so when we look at something like DraftKings and FanDuel, they have interesting situations where the odds that they can provide to a player or the actions that are happening throughout a game are all evolving in real time and they have a huge number of simultaneous games that are happening all at once and they need to be able to communicate both betting lines or other opportunities or changes that are happening in the game or the odds or what have you directly to a customer in a way that is relevant to them and they need to do it nearly immediately because these are things that are constantly changing.
(12:36)
And so that's a really great example where actually there's deep integrations between a lot of the activity going on out in the real world as the games are changing and also as the available odds and betting lines and everything else are changing. And then that actually then needs to go into Bray's engine to understand as the world is changing, how do we match that up with the context and the understanding that we have about each individual consumer? And then out of that combination comes a bunch of messaging that we then need to dispatch rapidly and get it out to people while it is still relevant. And so that's actually a really great problem space and we see that in a bunch of other verticals as well. If you consider on demand, this is one of my favorite categories because the marketing can be used not just to drive additional activity but actually to improve the quality of the product.
(13:26)
If you consider what happens when you've got on demand, whether it's food delivery or it's rideshare or what have you, when you have a mismatch between the number of drivers available and the demand coming from the consumer side, that can be a bad experience in either direction. If there's too many drivers on the road, they're sitting idle, they're not making money, they're not happy about that. When there's not enough drivers on the road, you absolutely don't want to be sending out passenger incentives when you don't even have any drivers to match them to. But on the flip side, if you actually have too many drivers, you could send out passenger incentives and there's interesting things that you can do with things like sports games and events and weather. If you for instance know that severe weather is coming into a certain area to be able to send out a notification ahead of that and say, Hey, there's going to be severe weather in your area in the next hour or two, if your plans are flexible, we recommend taking a car home. Now it helps them kind of flatten the curve in terms of demand spikes. It helps their network operate in a more healthy way. And it's a really interesting use case where because you're able to do that realtime demand balancing, you use marketing as a tool not just to drive incremental revenue or incremental activity, but actually to improve the overall health of your network.
Austin Hankwitz (14:37):
Makes a ton of sense. Bill. And speaking of customers, I really want to move the conversation now to your recent earnings report. Congrats again on the great results here. But the first thing I want to dig into is sort of how you all approach net new customers versus upselling existing customers. I think your CFO mentioned in your earnings call, it's sort of like a 50 50 split from a focus perspective, but with you all now just announcing your first eight figure customer customers plural this quarter, congratulations. I would imagine upselling against existing customers is increasingly more important. Can you break that down for us?
Bill Magnuson (15:13):
Yeah, well, so just to keep the record straight, one eight figure customer, we've got many in seven figures and we certainly expect to grow more of them to that milestone in the future. But I kind of think about this over a longer term time horizon. And definitely depending on what's going on in the macro, these things ebb and flow relative to each other. And so I'll go back to 2020 as an example where during the summer of Covid there were obviously a lot of businesses where demand fell off a cliff, but there were others where it's surge forward very quickly as our behaviors changed rapidly. But almost every company was in some form of crisis mode dealing with change. And so there was almost no one in the summer of 2020 that was actually looking at bringing on new software vendors, but there were a lot of people that were dealing with an influx of brand new demand.
(16:03)
And so that was a summer where actually upsell in the categories that we're experiencing rapid demand growth was relatively easy and new business was extremely hard. And so we looked at it and then we said, alright, let's not try to fight this. It's an external shock on the system that has led to this imbalance and let's just have a higher proportion of upsell for the next few quarters. But this is often referred to as hunting and farming, right? You farm the install base versus hunting for new logos, you know that you are eventually going to exhaust the farmland unless you grow the estate over time. And so I think it's always okay to lean into upsell strengths for a number of quarters if that's what the environment is dealing to you, but after that's been going on for a couple quarters, you need to make sure that you're not going to kind of exhaust the available farmland or the estate and make sure that you're growing that.
(16:58)
And so what we actually did and how we manage that through 2020 and into 2021 was that we refocused our marketing efforts and some of our sales incentives as we got toward the end of the year in order to incentivize more new business coming in versus the upsell. And so across a two year period there you saw roughly a 50 50, but you do see variance from quarter to quarter. Now a similar thing is happening right now because of what's going on in the macro, what's going on with interest rates and venture capital. And one of the things that we highlighted is that we are still seeing, especially on the small business side, more business failure. There's a lot of venture backed startups out there that are running out of runway or are at existential points when they look at how they're deploying cash. And that's something that we continue to see heightened logo churn on the SMB side, but enterprises who tend to have more mature planning cycles, they have more momentum or stability to be able to plan on longer time horizons.
(18:00)
They are actually already shifting back toward buying multi committing to multi-year contracts to being able to continue to expand. And so right now what we're seeing is that that upsell motion within the enterprise is relatively easier while new business in various places is much harder. And so I think that the way that I think about it is that an imbalance there over the short term, as long as we think there's a good external reason for it, it's certainly something that we will kind of ride through. But over the long term you want to make sure that there's a good balance between the two so that you're always growing your existing customers, but you're never making sure that you're going to starve your future supply of new customers that are going to eventually be the ones that grow years from now. It
Katie Perry (18:47):
Was interesting to hear you all articulate the differences of the macro impact on SMBs versus enterprise and when you were talking about enterprise and sort of the consolidations of tech that benefit you all, I was curious, when people are consolidating on the enterprise level, what tools are they moving away from to consolidate within Braze generally? What does that look like?
Bill Magnuson (19:11):
Yeah, so I spoke about the breadth of channels that we have available, and so at a high level what it usually is is that over the past 10 years or so, they usually have assembled a collection of software tools, each of which lives in its own channel. And so maybe they got something to be able to do messaging on their website. They bought another product to be able to do push notifications to the browser. They got another product to do mobile push notifications, iOS and Android, they had a different tool for email. The story that often plays out, especially in the enterprise is that they bought a legacy marketing cloud, a Salesforce, Adobe or Oracle product, one of their supposedly cross channel marketing clouds. All of those actually have a history where they were started with an acquisition of an email platform. So Salesforce bought ExactTarget, Oracle bought responses, Adobe bought neolane.
(20:04)
They each subsequently turned that into a Marketing Cloud product. And then what they did is they became multi-channel over time by acquiring other channel specific providers. But of course the consequence of that is that those tools are not deeply integrated. And so when you use them, you're using a Salesforce marketing cloud for email and then you want to go expand into an in-product channel or you want to be able to send push notifications with it, you're effectively moving into a different product. At that point it was acquired but never really fully integrated. Definitely the foundations are different, and what that results in is both complexity for you as the customer. There's complexity in the implementation, there's usually then also additional latency that's going on. The original email providers were built to be batch and blast. They were built for a generation of consumer behavior where you were reading most email on your desktop at work when you were ready to read it, right?
(20:59)
It's a completely different circumstance than when we're on the run all the time and the consumer is really interacting with brands on their terms, and then you add in these other silos of channels, you never really integrate them well and you end up with this mess of complexity and all these little gotchas and all this latency. And so the reality that ends up happening, and this is where when we move in our vendor consolidation, this might surprise some people, but it's usually a legacy marketing cloud that was procured to be able to be cross channel, but when the rubber hit the road, it wasn't actually well-suited to be able to do those other channels. So then they went and bought a cheap add-on some other single channel provider and before they knew it, they had five of them because modern work and customer engagement does require communicating across a lot of different channels simultaneously and being able to coordinate that.
(21:52)
And over the years they had a mess during the longest uninterrupted bull market that we've ever had, the one that ended in late 2021, early 2022, a lot of companies were just growing at such a breakneck pace that they never had an opportunity to slow down and take a breath and reevaluate or whether or not they were in a good place. They were making decisions for the time in order to keep moving. And so I think what we've seen in particular over the last 18 months is that there's a lot of companies who spent years building this fragmented landscape and now they find themselves delivering to consumers this incoherent the left hand's, not talking to the right, they get something on mobile, it doesn't know what they did on the website. They get something from the loyalty program that has no clue what's going on over here.
(22:41)
We've all experienced these examples and a lot of that is due to that fragmented technology. And so that embrace is able to come in and just take full responsibility for all of those channels and consolidate all of those out. And of course AI is now a really important part of that picture and that transition story because if you ask yourself, Hey, how am I making quality decisions about exactly where to communicate with each individual consumer, and it's not just like a one shot, like, okay, I have one message I want to send, which channel should I send it to? It's like I have an entire communication strategy which is going to unfold over the course of a customer's first 30 days. How do I have that automatically adapt itself depending on which behaviors that customer shows me about what their preferences are and how they're engaging. And by the way, some of these message channels have wildly different prices. Push notifications are effectively free to send. Email has a bit of a marginal cost. SMS and then WhatsApp are 10 or a hundred x that depending on what country we're in. Why
Katie Perry (23:40):
Is certain WhatsApp, can you explain why that's more than other platforms?
Bill Magnuson (23:48):
Well, so the simple answer is because Meta's really good at making profit, and so there's value in those additional channels because they perform better. And there's actually a correlated relationship there, which is that the more expensive it is to deliver into a channel, the less noisy it is and so therefore the more likely you are as a consumer to pay attention to it. And so when we look at these channels, they have various marginal costs. It's usually worth it because you are much more likely to have the consumer respond the more that you pay to get into an uncluttered inbox, right? So your WhatsApp inbox has very little marketing in it. Your SMS has until a few years ago it had almost none, and now it is starting to get more of it, but you're pretty likely to read it. You probably ignore most of the push notifications that you get.
(24:38)
But for a given customer in the early days, if you have a customer who when you send them their onboarding push notifications in the first seven days and they're very reliable at responding to those and connecting with them, there's really no reason for you to ever be sending those people SMS or WhatsApp messages both because there's no reason to be violating the sanctity of those inboxes for those people, but also because it's going to cost you more money. And so when you look at that problem of how do I orchestrate across these different dimensions of what's going to perform, how do I have the best ROI, how do I make sure that I'm respecting my customer's expectations around where I'm showing up with my message? And then you ask yourself, how do you do that? If you have a half a dozen different technologies that are responsible for sending these messages, right?
(25:25)
There's no overseer and to the extent that maybe you try to then layer another piece of software on top in order to be the overseer, it's just going to be too slow to be able to respond in the moment and too disjoint. And so I think that's where Braze ISS original architecture where we vertically integrated the whole stack with an event driven stream processor to ensure that we were going to be able to respond in the moment, and then we built a strong abstraction layer over the different message channels so that when a marketer uses Braze all of the concepts that they learn around how do I target people, how do I orchestrate strategies, how do I personalize content in order to maximize relevance? How do I run experiments across these? All of those concepts are equally available in every single channel. And so when you learn how to use braise, you're not just learning how to send email to people or how to use an SMS platform, you're learning how to do customer engagement and the channel ends up being more of an implementation detail to be optimized.
(26:22)
The distinction being that saying when you've got a hammer, everything looks like a nail. When you start out with a software platform that's already siloed, you say, oh, I have this push notification tool which push notifications am I going to send to people? Or I had this email service provider, what emails am I going to send to people? It's the wrong way to approach the strategy. You have goals as a company. Your customer has goals as a consumer and as a user of your products and services, and you need to do the best job possible to be able to marry those together. And then which channels you ultimately end up communicating with that customer in should be an implementation detail to be experimented with and optimized, and you're just not able to do that when you've got a cluster of a half a dozen tools. And so I think more and more marketers are waking up to that reality and that's why you're seeing an acceleration in that vendor consolidation trend.
Austin Hankwitz (27:12):
I think you explained that perfectly and not just more and more marketers waking up to that reality, but you all are certainly proving to the market that you are this dominant overseer of all these sort of different product tools that are not exactly talking to one another. You guys reported 29% increase in customers spending over $500,000 a year with you on your platform. So considering that you guys are inching toward profitability and aiming for that in Q4, non-gap profitability, is this sort of becoming the dominant overseer? Is that the strategy that's going to get you there and how do you balance that growth with hiring people trying to be profitable but also expanding the sales team? I'm sure you guys are looking at every single light item from the expenses side of the equation right now as you inch toward that profitability.
Bill Magnuson (28:01):
Yeah, absolutely. We made the decision and we shared that with the market around this time last year, that we would end our prior fiscal year, the one that we just reported at better than negative 6% non-GAAP or better than negative 7% non-gap operating margin. We improved the target as we shared our Q3 earnings late last year, and we ended up exceeding that as we just reported earlier this week. And we also shared the goal that we would, as you referenced, reach non-gap operating income profitability by Q4 of this year as well as free cashflow positive by the end of this year on a non-GAAP basis. And those are commitments that we made to the street. We feel we're really excited about them. It's obviously an important milestone for any business to eventually be profitable and make money. We do have, however, $500 million sitting on our balance sheet from our IPO one of the benefits of IPOing at the top of the market, and we actually IPOed at something like the third or fourth highest day of the NASDAQ of all time in late November of 2021.
(29:04)
And of course, it's important to remember that an IPO is a fundraise in addition to everything else it is. And so we brought a lot of cash into the business that we're able to use to invest. And so as you alluded to, definitely making big investments in r and d, we're growing our sales team for the first time in over a year as we're responding to the demand environment that we're seeing. We also went through a rebrand just a little bit ago, and so we've been spending from a marketing perspective in order to help with broader awareness and help with top of the funnel activity as well as get our new brand out there, which is really exciting. And so all of those things are certainly being balanced. We're confident, excited about hitting that profitability milestone in Q4 of this year. And of course, these are all, they're all important to keep in balance.
(29:50)
We are making investments today for long-term growth capability. We are doing that because we believe that there's tremendous future potential and opportunity. You may have kind of surmised this as I talked through the examples earlier, but Braze is an incredibly diversified business. We get over 40% of our revenue from outside of the United States. We work across a whole bunch of different verticals. We have a product that is capable, used by small one person marketing teams all the way up to the world's most complex global multinational conglomerates. And that all adds up to tremendous future opportunity for us. I think we're still very early in the total addressable market when you look at each of those sub slices. And one of the things that really excites me about Rays in our growth path is that our ability to continue to grow in each of those verticals is not hypothetical.
(30:43)
We're not like, oh yeah, we're really good in e-commerce or we're really good with small businesses, or we're really good at the enterprise and we're going to someday expand into those other places. We're already there. We already have customer proof points across all of those different verticals all the way around the world across all of these customer sizes. And we have customers as well that use Braze in a lot of different ways. There's use cases that are purely marketing and promotional things. You're used to sending a newsletter, promotional flyers, things like that. There's also product delivery where we're modifying the behavior of a product while a customer is onboarding through it. There's transactional and operational. We talked about delivery earlier, telling you that your order is on the way and sending you the follow-up survey afterward as well as being able to be preemptive about it.
(31:28)
We talked about the Draftking and FanDuel example where you're live and in the moment we have a whole bunch of news and sports customers as well, as well as customers in the FinTech area that are sending live updates about things like earnings or scores or breaking news or what have you. Really just runs the gamut across use cases. And so I think that Braze has really done the hard work of proving that we have a role to play in a lot of different opportunity spaces, and that's why you're seeing such tremendous investment in the business. We have line of sight to being able to continue to scale our existing customer engagement platform, and of course we're expanding the surface area of it every day as well. And
Katie Perry (32:07):
As someone who comes from the marketing world, another thing for people to understand is the stickiness of certain MarTech platforms. And at my prior role, it was actually becoming core to the job to be able to know how to use braise. And I think that's a big piece of it too. If you look on skills on a resume, being a product that lines up with a requirement for a lot of companies is also important. I want to go back to 2011, what a time to be alive. So you all came about the same year as Airbnb, Spotify and Square, quite the draft class there, but being with the company for 13 years, I'm curious how does being a founder influence the way somebody runs a public company in your opinion?
Bill Magnuson (32:54):
Yeah, I think that, and actually I'll add to that, it was an interesting part of my story throughout those 13 years, which is that I was actually our CTO in the beginning. And so I started out mid 2011 running the technology organization and then it was in 2017 that I took over the CEO role. And so over the lifetime I spent about half the time those in the early years as a technology leader and then took over the CEO role. I think that being a founder in the public markets, I haven't ever been a non founder in the CEO seat in the public markets, so it's hard to provide a real comparison, but absolutely I think the tenure and the experience provides you with an amount of context that can allow you to really have high conviction in making investments. I think that there's a lot of companies that end up operating according to the whims of the public markets and where the winds are blowing from a kind of growth versus profitability and all these other things.
(33:55)
And I think that by being in the founder role, you not only have the context and the information and the credibility to be able to make those calls, but you also have the conviction to do it. And so to be able to look at like, Hey, what's the flavor of the month right now? We could have decided that we were going to push toward profitability by Q4 of this year that just ended, right? There are a lot of companies that did that. We were really happy with the improvement that we had in our operating income margin. We improved by over 1100 basis points, which I just shared in earnings earlier this week, and that's obviously a tremendous improvement in the course of a given year, but we decided that that was enough and that we wanted to actually chart that course over two years because there were great investment opportunities to be had during this environment.
(34:43)
And so I think that's a great example where it was, maybe it was, I wouldn't necessarily call it unpopular because obviously we've grown the stock price quite a bit over the course of the last year, but it was definitely something that was uncommon. There was definitely many other companies were rushing toward profitability trying to get there as soon as they could in order to prove the markets they were going to get there. We decided to be patient about it because we've got that strength in the balance sheet combined with the tremendous future opportunity. When you put those things together, I just felt like we'd be shortchanging our future if we tried to rush toward that too quickly. And I think that's a decision that is certainly easier to make when you sit in the founder seat.
Austin Hankwitz (35:25):
That makes a ton of sense. And you mentioned being the CTO and sort of here in the last call it 12 to 18 months wanting to invest into new things, artificial intelligence being one of 'em. You guys were early on the AI trend, I think it was May of 2022 when you all introduced it to your platform. How did you realize that AI was going to be something big in 23 and beyond 24? Now everyone's talking about it and one, how'd you discover it? And two, what was the decision, a sort of light bulb moment where you're like, okay, this is going to be something that is now going to be pivotal for our company?
Bill Magnuson (35:59):
Well, yeah. So first I'll correct the taxonomy here, which is that I think AI is being used as shorthand for Gen AI now and specifically being used as shorthand for use of LLMs or other sorts of transformers. We've actually had artificial intelligence and machine learning in the platform for the better part of the last decade because if you take a step back and think about all of the optimization decisions that I've been alluding to you benefit from, and I try to always break this down and be really pragmatic about how we describe things, and I put a lot of these in the class of automated decision making. So as you're moving through the customer journey, it's increasingly non-linear, all these different touch points. You have a lot of data flowing in both from the actions the customer is taking as well as from how the context in the outside world is changing.
(36:48)
And we talked through some examples about that earlier around restaurant inventories or weather or news or the games or odds or what have you. Depending on what vertical you're in, you need to be able to bring those things together. And so there's a lot of opportunity for advanced statistical methods, various machine learning methods inclusive of using neural nets and other sorts of regression capabilities, and just generally looking at every single decision that's being made, whether it's timing, channel selection, content in order to improve relevance. There's just a lot of opportunities for artificial intelligence, machine learning and various statistical methods to be used in order to improve outcomes. And so we've had teams dedicated to that for a long time for many years and have had that in the product as well. And so part of the answer to how were we so early on the gen AI front is that we already had teams that were working on these things.
(37:40)
And when Transformers came out before the chatbot interface, which I think through chat two PT is how most people became aware of it, there was already exciting work going on. The early versions of chat GPT were not as impressive, the ones that were running on earlier versions of GPT, but we immediately recognized that that ability to provide, even if it was just a creative spark, was really valuable. And I think that I hear this example used a lot, which is you're with your colleagues, you're trying to decide where to go to lunch, and everyone's like, oh, I don't really know where to go to lunch, et cetera. And someone's like, all right, well, we're going to go to the corner Deli. And then someone's like, oh, wait, no, now I have an idea. And as soon as something shows up, that is an example that people can respond to immediately their brain starts working in a more creative way.
(38:33)
And you really can get brainstorming going with anything. But the reality for a lot of marketers is actually they do their jobs alone. What we saw was that we had these great experimentation capabilities which were like, Hey, if you have a half a dozen different approaches to try to engage with lapsing users, people that haven't used the application in the last week or two weeks or what have you, and you want to draw them back in, you might do that in a variety of different ways, but it's hard to sit down as a single person and be able to come up with six completely different ways that might resonate with someone. And so we've always been on the lookout for how do we inspire new ideas, new creativity, et cetera. So even the earliest days of the Gen AI capabilities were really interesting to us, how do we make marketers more productive?
(39:18)
Because Braze is a very sophisticated platform when it comes to experimentation and optimization, and so the easier it is for inputs to come in, the better job we can do using other machine learning and AI in order to optimize outcomes for the customer. And so that's why you saw very early integrations of both Dolly for Image generation as well as GPT as soon as it was available via API. We had it integrated into the platform. I just earlier this week announced tone control. We also have, we've had some various cultural and context aware content checkers as well. One of the interesting things about being an individual or even if you're a team of people working on these marketing problems is that you probably even as a team grew up in one or a handful of different cultures or languages or what have you, but when you're communicating with a global audience, there's a lot of things that you need to be careful of first.
(40:12)
And also there's a lot of opportunities to optimize for additional relevance and resonance that you might not even realize. And so one of the really cool things about an LLM being trained on the entire globe of text content including another languages, is that you can be a marketer sitting in New York who grew up in the United States and you can say, Hey, this is a marketing angle that I am working on, but I want to really make it resonate for a Brazilian audience. How do you recommend I do that? That's a really great example where the LLM can reach into all of the patterning that it has around Brazilian culture and take the message that you have and try to transform it. That then creates an opportunity for you to be able to do personalization or localization in a way that you wouldn't before. And when you then put that in something really powerful like braise, we're able to then optimize that.
(41:04)
And a really cool thing we just launched earlier this week as well into general availability, our intelligent path feature, you don't even need to necessarily know who is in Brazil or who associates with Brazilian culture on that anymore. If you merely create the option where you have a path that resonates for this particular culture or this customer persona, we will automatically find other customers in your customer base that have properties that are similar to people for whom that has resonated more actively in the past and send those people down the right paths. But you still need to obviously create those paths. And so we always look at that interplay between the content production and inspiring people and creativity and all of that, and then also being able to optimize it from there.
Katie Perry (41:48):
I got to ping from a political candidate that was unnamed pro fish candidate, which nothing to argue with there. I feel like that's a platform we can all embrace. I know you're focused on brands, but are you working with any political campaigns? If not, can you because their engagement strategy seems to be a little stuck in the dark ages. Talk to us about political implications within Braze.
Bill Magnuson (42:15):
Actually, we don't work with political campaigns. We as a subscription business really focused on long-term relationships. Political campaigns unfortunately for them have a bit of a reputation for not paying their bills if they lose the election and just kind of disappearing into the night. We've really focused our customer acquisition on categories where people are trying to build longstanding businesses. I totally agree with you though I don't understand how they're not doing any personalization at all, how they're not frequency capping anything. They also all increasingly end up in everyone's spam inboxes, and so these messages that they're spraying out to everyone are not even effective. So it's definitely a problem that should get solved. We are open to working with political campaigns as long as they're open to doing multi-year commitments with us, but it's not something that we've really been pursuing, even though I totally agree that there could be some improvements there.
(43:13)
I actually have kind of a funny story from one that did reach out to us, and we are very strict around deliverability because we've always tried to make sure we're on the right side of history from a privacy perspective, and we want to be really respectful to the customer, really respect that they've given us the right and the opportunity to be able to communicate with them when we want to, and that's a responsibility that people should take seriously. And so we have a bunch of, there's strict rules and then there's rules of thumb or what have you. And the goals of those are to keep you in the inbox. If you're sending email to people or now increasingly SMS, which also has a spam and promotions tabs, your messaging is substantially less effective if you're sending it to huge groups of people that don't want it, they're not opening it or marketing a spam.
(44:06)
And then for the people that maybe did want it, now you're in their promotions tab or you're in their spam folder and you might as well have not sent those messages. And those are things that are actually in the email world in particular, there's a lot of great organizations that help with deliverability and reputation monitoring, and actually they operate in a way that is purposefully a black box because they are an adversarial activity. They've been fighting spam and fraud and abuse and organized crime for years and years and years and so much Google doesn't publish all of their SEO algorithms because they could be more easily gamed if they did. So these organizations don't strictly publish their rules or exactly how they operate. And similarly, all of the ESPs like Gmail and Yahoo Mail and what have you, they all have their own rules as well that they follow.
(44:58)
And so when we have a acceptable use policy with rules around what kind of email you can send, what kind of consent you need, et cetera, a lot of times those are actually, these are just things that you should do because otherwise you end up in the SW inbox or you end up blocked anyway. And we had a political organization that reached out to us and they were arguing with us that our acceptable use policy was violating their first amendment rights because as a political organization, they should be able to say whatever they want to people and it doesn't matter what's in our acceptable use policy. And I kind of laughed at it because I was like, guys, spam House and Gmail are not the Supreme Court. They don't really care what your legal argument is about what you're allowed to say or not or what freedom of speech is.
(45:40)
They're just going to put you in the spam folder. And so by all means have this opinion about what you're allowed to say that's based on the law, but when you operate in this world, you need to realize that actually they have been engaged and in this adversarial activity against organized crime and fraud and spam and abuse for literally decades. And they're not going to listen to your legal opinion. They're just going to put you in the spam in box. So if you're going to leave, by all means go and work with someone else. That's okay if you want to do that, but we still think you should take hard advice about what kind of content you should be sending.
Katie Perry (46:16):
Yeah, I guess if you're going to yell at me in my government name, that is your first amendment, but it's probably going to end up in a junk folder. Super interesting. Bill, thank you so much for coming on. It was awesome getting to know Braze a bit better, going through earnings, hearing a little bit about your leadership style. Again, we really appreciate you coming on today.
Bill Magnuson (46:37):
Absolutely. Thank you for having me, and I hope you guys have a great weekend.
Katie Perry (46:40):
You too. You
Austin Hankwitz (46:41):
As well, bill. Thanks Katie. What a great conversation with Bill from Braze. I feel like I learned a lot. This guy had all the answers. What was your favorite takeaway from the conversation?
Katie Perry (46:52):
I really enjoyed his articulation of the science of marketing and what goes on behind the scenes. It's a lot of, if this, then that flow charts that lead to when you get a ping from an app you use or a reminder or an email. And I thought it was really interesting how we talked about how the data is both internal, so data they know about the customer as well as external. So if you're using DraftKings, they're going to pull real-time data about what's going on in the next game, but they're also going to know who you are as a user and combine those. So I thought that was really fascinating. I think when a lot of us think of marketing, we think of the billboards, the TV ads, and there's a lot kind of behind the scenes that's actually quite interesting that goes into that.
Austin Hankwitz (47:33):
Yeah, it's kind of scary to think about all the data that these companies have on us, especially from a marketing perspective. I mean, you mentioned the Uber Eats, right? When it says, are you hungry? But maybe it really knows that this is my eating pattern. It's not just a random time that they're getting sent this information. So I don't know. It's spooky to say the least. I think my perspective on the conversation was more about, obviously this is a founder led company. This company started in 2011, 13 years into the future. Now he's still running the company. And what's so important about that is his resilience as it relates toward profitability. They mentioned how a lot of these publicly traded companies, especially in 20 22, 20 23, were like, okay, we got to be profitable. Interest rates are rising. Macros not looking that great. We need to start actually printing money for our shareholders versus just trying to grow, grow. I'm not saying Braze is focused on growing, but it's pretty interesting that he was like, well, there are better tools, there's better things that we should be investing our money into versus just trying to get profitable in the short term. So I thought that was a pretty interesting long-term sort of founder led perspective on his company there, sort of really focused on what's not just going to move the needle today from a stock price perspective, but more importantly over the long term, call it five, ten, fifteen years into the future.
Katie Perry (48:52):
And also, I mean, what about the political campaign insight? Very interesting how he talks about people blaming First Amendment sort of violations when it's really just like the message sucks and he gets into that. And that was super interesting as well. So we covered a lot of ground, great conversation
Austin Hankwitz (49:12):
And hopefully, I mean, I am praying that you're not getting any more of those fish text messages. Everyone, thanks so much for hanging out with us on this episode of After earnings. If you have a publicly traded company that you want us to interview, you want to know more about the company, the financials, the CEO, the c-suite executives running that company, leave a comment, email us, do whatever you can to get in touch because we are on the pr. We have a bunch of really cool names coming up over the coming weeks, but we want to make this show around you, all the listeners, so be sure to let us know your thoughts and we can't wait to dig in.
Katie Perry (49:48):
And with that, I'm Katie Perry.
Austin Hankwitz (49:50):
And I'm Austin Hankwitz,
Katie Perry (49:51):
And this is the After Earning show brought to you by Stakeholder Labs and Morning Brew. Be
Austin Hankwitz (49:55):
Sure to subscribe and shed this episode with a friend and we'll catch you next time.
