Braze, Inc. (BRZE) Earnings Call Transcript & Summary

Thanks for joining Citi's tech conference. Day 3, Happy Friday to everybody. My name is Tyler Radke, Co-Head of U.S. Software Equity Research here at Citi. To kick the day off for software, we have New York's local marketing automation company, Braze. We have the Founder, CEO, Bill Magnuson; and CFO, Isabelle Winkles. I know you were busy last night with earnings. So I appreciate you coming bright and early to our conference.

Thanks...

So maybe for folks here in the room and on the webcast who haven't heard of Braze, just give us a quick overview of who the company is and what you do?

Yes, sure. So we were founded in 2011 right here in New York City. So we just celebrated our 13-year anniversary over the summer. And what we do is customer engagement, primarily for consumer brands. And what that means is that -- like what that means literally is that we ingest a tremendous amount of data, literally trillions of data points on behalf of our -- a bit over 2,000 customers every year. and use that in order to drive also trillions of messages to people. We send messages across push notifications, e-mail, SMS, WhatsApp. We deliver messages inside of product experiences, both familiar ones, things like surveys and pop-ups and notifications as well as doing things like running inboxes, inside of products. We ingest or we include a huge amount of intelligence into all that, obviously, helping with things like who are you talking to, orchestrating and prioritizing different strategies that brand might have, being able to personalize and optimize relevance within the content, whether that relevance means showing up at the right time, saying the right thing, connecting with someone through that context. Over the years, we continue to expand to new channels, new platforms being effectively anywhere that the customer is on their journey. And I think that a really big part of what distinguishes modern customer engagement from what we would have talked about before is either e-mail marketing or marketing automation or what have you are a few things. One is that the modern customer experience is always on. And so the importance of the ability to activate and understand data in real time is of the utmost especially for brands that are in highly competitive global digital industries. And we'll talk a little bit more about the global nature of this problem as we continue through this chat. The other side is that I think that the importance of customer engagement really being a companion to the user journey, is a really important thing to realize kind of understanding the context of the customer as they move through your utilize your product or your service and being able to -- be there to provide the right nudges to be silent when it's appropriate to be silent, right? This isn't just about kind of broadcasting spam messages and to a cluttered in box anymore. You need to be able to be in the right places in the right moments or else. It's not just like less effective, you actually you get shut off, you lose those customers. You don't have the right to be able to communicate with them anymore. And that really can expect into another important theme, which is the importance of first-party data, right? So many brands now are investing and building up first-party data sets and investing in evolutions of their own products and services that allow for them to have a first-party connection with the customer. And this isn't just the obvious examples like Disney launching Disney+. It also includes things like looking at the automotive industry, where you would drive a car from an automobile manufacturer every single day of your life and largely only interact with them by throwing away their warranty cards when they send them to you in your mailbox, right? Now through a connected car and being able to unlock your vehicle with the mobile app, all of a sudden, they actually have an ability to understand your driving habits and communicate with you every single day. And that fundamentally changes what they can do from a customer engagement standpoint, and it allows them to actually inform the business model in interesting and innovative ways. We're seeing examples of that across all different categories as brands are investing and getting closer to the consumer, understanding them better and then leveraging that understanding as an asset to their business in order to drive additional revenue, higher levels of attachment to those customers and more efficient ways of communicating with them.

Yes. It's a great overview. And maybe before we get into the details of sort of the business momentum on the results. Just sort of a high-level question, as you think about where we are on that journey in terms of brands having that first-party relationship with customers. I guess how far do you think we are in that journey? And I guess, in some ways, is an example of driving a car off of the dealer's lot, that might be the only time you interact with them. But at the same time, consumers are inundated with notifications from kind of chatty or spammy app. So how do you sort of see this moving forward, where in that cycle? And how do you kind of find that...

Yes, I think we're still very early. When we kind of look at first-party data evolution for a brand, there's kind of 3 stages. The first one has been -- that most brands are in [indiscernible] and you read a lot about the death of the cookie and -- I guess, the death of the cookie continues to get postponed. But the simple fact of the matter when you look at GDPR and a lot of the changes in the kind of privacy landscape is that a lot of brands that were previously relying on kind of renting or buying third-party data sets in order to do their targeting are now finding that those strategies are ineffective, incomplete or illegal in some cases. And that is obviously an impetus for many of them to start to build their own first-party data sets. And that's kind of a -- it's a loss avoidance pressure that exists. They were doing something before that things becoming less effective now. They want to be able to keep doing it to build that third-party data set in order to keep on running their performance -- in many cases, their performance marketing strategies in a similar way. Then there's the next phase, which is actually rethinking the way that your products and services delivered to your customer in order to be able to create more comprehensive first party data sets and also get permission to be able to communicate with the customer through some sort of first-party media. And this includes things like having connected fitness products, other sort of connected devices. It's building products and services, subscription programs, loyalty programs right? There's examples of this across a lot of different categories. The automobile one is interesting because I think it's not the dealer, but actually, the interesting thing there is the manufacturer, right? Like you use their vehicle, their products every single day, but you fill it up from an oil company, you get service from a service station, you buy the insurance from a financial services company, right? There's all these other brands that you interact with and never with the actual manufacturer. And so they're getting an opportunity to come back to that. You look at the entire CPG category, right? And for decades, they thought about their customer as a CPG company as being like a Walmart or a Target or a Big Box retailer, right, not the end consumer. Now they actually have an opportunity to interact with the end consumer through direct-to-consumer programs. And indeed, like many large enterprises in the CPG space are being forced to do that because there's a lot of start-ups that are doing it very effectively, right, within the D2C space. You look out across places like learning or personal health and personal finance, right, to be able to build companions to people's journey so that when they want to develop healthy habits, whether it's to help them be healthier, have better financial wellness, have a more enriching life like -- those are all places where customer engagement really drives the customer toward better outcomes because we're able to bring these products and services into their lives in a fulsome way. And so I think when we -- when you look at those 3 stages, there's kind of the loss avoidance around the performance marketing and the loss of the third-party data sets. Most brands are there now because they've kind of been forced to be. In the second phase of how do I rethink my business in order to have an opportunity to build up a first-party relationship. And I mentioned Disney+ before. It's a really great example where Disney had this incredible multifaceted relationship with the customer, but they didn't really have much direct interaction until Disney+ really entered the picture, right? They were -- you were seeing their movies in a movie theater chain. You were getting the Disney channel through a cable bundle. You were buying the merchandise from a Big Box retailer, right? There were all these interactions with their characters and their stories, but you didn't actually have a connection to Disney the company other than in the theme parks but that usually ended as soon as you walked out those days, right? And so they're actually rethinking their products and services in order to create a multifaceted direct connection, and that gives them better data to understand the customer, they can interconnect those parts of their business Empire in really interesting and valuable ways, both for them and for the customer. And then there's a third phase of then how do you actually rethink your business strategy? Like what do you get to do once you actually have that connection. And you go back to the automobile manufacturer like now all of a sudden, you have the opportunity to get in front of a customer as their lease is renewing and you don't need to just hope they drive by your billboard or want a commercial on a football game, right? You can actually communicate with them directly. And you know how they're driving, it have changed over the course of their lease length. And you can also deliver to them personalized and tailored insurance offers. So you're the one that's aggregating that demand. We see that in the QSR space as well, which is that as individual quick-service restaurant and fast food brands have been able to build out their own mobile ordering and enhanced takeout offering and the mobile wallets and et cetera. They've been able to flip the script on many of the delivery companies. So instead of the delivery companies aggregating the demand and charging dearly for it, every time someone orders McDonald's through a delivery app or what have you, -- now McDonald's can actually negotiate with the delivery companies as a commodity supplier effectively, right? It totally flips the script. It allows McDonald's to own the first-party data on the customer to communicate with them afterwards and it improves their margin as a business. And so when we kind of see how that transformation then it kind of serves the business model to both give them more innovation and give them control over the demand, it needs to a more healthy, more robust business. But I think we're still very early in terms of that progression through those 3 stages, permeating the entire business ecosystem.

Yes. That's a really fascinating vision. I did want to sort of ask how you think about GenAI tying into all that. Because on one hand, obviously, we're talking about lots of data. The first-party data is going to be used perhaps to be training some of these models. Like do you view GenAI sort of an accelerant or tailwind to kind of this vision that you laid out. Help us understand what you're seeing just from a demand perspective.

Yes, I think it absolutely is, and there's a few stages to it. First, upfront, obviously, as you may have surmised, a lot of the work that's done with Braze is fairly sophisticated when it comes to customer engagement, right? There's a lot of data activation. There's a lot of aspects of the kind of marketing that require technical expertise, whether that's like working through things like scripting or building -- building messaging. It's not just a little bit of text, right? If you want to build engaging, dynamic, multimedia messaging, a lot of times that has development and content production requirements, associated with it, being able to test and understand like either new insights coming out of data or be able to do advanced analysis of that data. There's a lot of places where our most productive customers are bringing together skill sets inclusive of marketing, product and engineering and data science. But it's also the case that not all marketing teams have access to those resources. And so one of the great things that GenAI has been doing upfront is that it's allowing an individual person to be able to use software like Brave as if they are working with a whole team right? So you've got a developer to help you with your lightweight scripting and generate your SQL and help you dig through reports and surface insights better. You've got a copywriter to be able to help you -- and maybe the problem is even harder because you're trying to communicate with people in 10 different cultures and socioeconomic backgrounds and different languages as you're communicating across a vast global -- vast global customer base, being able to do things like check for cultural appropriateness and make sure that you don't get canceled when you send out a miss -- kind of miss in for message or what have you. Those are all things that hold back individual marketers from being able to be more productive, be more agile, test and experiment more where GenAI is already coming in to be an assistant and to really be part of a team for them. And so that's all kind of in the composition side, right, where the marketer is using the dashboard, building out their campaigns, building out what we call canvases, which are being a companion to the user journey and just letting them be more productive and operate with more confidence and really in that way, then run more strategies. On the other side, I mentioned it at the front, the volumes we're talking about here are in the trillion, right? So obviously, automated decision-making has been an important part of what Braze has done the whole time. And we do -- we kind of make those decisions automatically in terms of who we're talking to, what strategies we're prioritizing and what we're seeing, how we optimize relevance, et cetera, using a variety of different techniques like basic conditionals, obviously, you start with the if and else case statement, right? Then you move to more statistical methods to be able to do prediction and other forms of deep learning and machine learning within that. From there, being able to use things like transformers. So we've actually seen an interesting progression from a content recommendation standpoint where you -- first, you kind of start out with being able to just respond to relevance, what's new, what's hot, what's trending. These are basic statistical approaches. Then moving to more of a deep learning approach where you're able to do kind of driving in the same way that recommendations come out of that you see in like a Spotify Discover playlist or the way that you see like item and recommendations when you're shopping. More recently, we actually reimplemented our item recommendation capability using Transformers, which, of course, for T in GPT. And those approaches actually not only are outperforming buyer approaches using deep learning, but they're doing so more generically. And so it's sitting on top of not just what item am I going to recommend to someone, but actually maybe I'm a fitness application. And like what are the exercises that I could recommend to someone in the morning? What are the different trails that are in your area that I would recommend you go hiking or biking on this weekend, being able to kind of dig through different activities that are like restaurant recommendations or travel itineraries as well as, obviously, all the other use cases around retail and food and being able to recommend things from catalogs, music, movies, et cetera. The Transformer approach has actually proven to both be more effective and be more generic, which has obviously been really good for us. And so when were -- when we kind of look across that [ fab own ] space of both, how do you make the market more productive and how do you make more decisions automatically, more quickly with higher levels of effectiveness. The introduction of GenAI kind of at the frontier of that space has helped on both of those fronts pretty tremendously.

Yes. So it almost seems like you're an enabler of sort of GenAI, but also sort of a downstream beneficiary if you see more volumes and more...

Yes, absolutely. An enabler in the automated decision-making world, and those are places where we're just delivering those new -- we're delivering those new capabilities to our customers that are using our automation and that automation is getting better, and it's getting more comprehensive. It's able to make decisions at a higher level as we continue to advance it. And then, yes, on the other side, making the marketer more productive, obviously, leads to more usage, right? And it leads to better campaign deployment, they can move more quickly. And we've also seen that the more ability that we can deliver to the -- to our customers, the more creativity that they're able to work with as well, the better outcomes that they get.

Great. Great. Maybe translating all the great momentum and great trends into what you're seeing in the most recent quarter and maybe we bring in Isabelle here, too. Give us sort of the quick synopsis on how this recent quarter played out. Obviously, the context of a lot of fun office software companies, it's choppy. We have seen several companies talk about worsening trends in Q2 and elevated churn and everything. I think Braze seems to be outperforming some of the smaller companies have called those challenges. So give us a sense on sort of what you're seeing out there in the market.

Yes. So I think, look, the market, the macro continues to remain challenging. I don't think it's gotten any worse, but it certainly hasn't gotten any better. We've talked about this over the last couple of quarters. We've seen different parts of our business start to kind of perform better than expected or better than in prior quarters. Issue is we're not really seeing anything sustained for more than sort of 1 or 2 quarters sequentially. And so that continues. We -- what we'll say is we are executing at plan, so there's no negative surprises out there. You heard us say earlier in the year that from our guidance philosophy perspective, we're going to be a little bit closer to the pin this year, and you're seeing that continue to play out. We are seeing -- over the last several quarters, we've done a lot to invest in the productivity of our sales force. That hasn't quite totally gotten to the other side, but we are seeing enhancements and improvements there. And so we -- and we're encouraged by that. And we're starting to put boots on the ground in increasing number of parts of the world, which is really exciting to see that from 2 perspectives. One is continuing to sow the seeds of growth opportunities and continuing to expand our global footprint and also capitalizing on some of these new strategic locations that are helping us on our path to profitability. So I think both from a revenue trajectory perspective, cost optimization perspective, we're very pleased with our results for Q2.

Great. Great. And on some of those sales initiatives you talked about productivity enhancements. Maybe just walk us through what you're doing and kind of what you need to see to be able to say we're on the other side.

Yes. So I think a few things. Obviously, training has been an important part of it. I think a lot of companies went through a lot of rapid growth of their sales teams through the kind of 0 interest rate period and just kind of going back, taking stock of that, making sure that we operate as a global team and that we get strength out of our global footprint. We definitely see a lot that competitors will kind of grow up in a certain part of the world and then they'll show up in other places or we'll see on the other side, buyer sentiment and kind of buyer behaviors. We tend to see those travel around the world as you go from geos or industries that are -- that tend to be more forward leaning into places that tend to be more risk-averse and you're just making sure that as a global organization, we're sharing the learnings that we're achieving in different parts of the world, and we're using those to strengthen other global teams. And that comes out of better alignment, more transparency, more coordination, more knowledge sharing, et cetera. And so I think that's been really good, and we've seen great progress in terms of our competitive win rates over the last few quarters, in particular in the different parts of the world that we've been focused on, the competitors that we've been focused on within those. Similarly, we've gone through and done some meaningful changes to the way that we sell our products. So we've spoken a little bit over the last couple of quarters on the shift to a credit model. This has been really important from a sales or productivity standpoint. Previously -- and just kind of as a -- to quickly bring people up to speed on that. If you had bought SMS from Braze a year ago or SMS, MMS, WhatsApp, you literally had that contract an annual volume on a per-country basis for each country and each channel. And so you might have 30 different line items in order to just buy what we would think of is SMS. We were kind of pushed in that direction due to a variety of accounting and other sorts of considerations. It certainly didn't feel ideal right out of the gate. But that's something where we've gone back. We've developed what we needed to in order to now build a more flexible credit model. Customers are able to buy more a fungible kind of basket of credits to be able to send those premium messages now, it's easier for them to buy. It's easier for our sales team to sell. And we think also will lead to lower amounts of unused entitlements and therefore, partial churn sources over time. It also better prepares us as we launch new channels. And so when we look ahead to the general availability of line which will be happening a little bit later this fall, as well as expansion into new channels like landing pages and some of the other things that are on the horizon that we're excited to share at Forge in just a few weeks. This is really setting the stage for our go-to-market machine to be able to deploy new channels faster and customers to be able to utilize them more flexibly and efficiently. And so that's been an important part of it. I think also optimizing the onboarding and usability side. And so you saw the announcement of our free trial motion from just a couple of days ago. That has actually been the culmination of over a year of focused efforts around product like growth motions. And so we started out originally within the installed base of looking at some of the expansions, so expanding into things like our product catalogs or our predictive churn capabilities or things like the AI item recommendations that I mentioned earlier. These are all upsells that an existing customer would be -- would want to be able to use in many cases, beforehand in order to kick the tires on them. And we didn't really have a motion to be able to do that before. It was a more classic enterprise software sale, where you had to sign up the data blind before you got access to anything. What we've been doing is really making sure that more of our product can really cut the life for customers, and they can -- they're able to trial it before actually signing the contract. And that's something that I think has been really important as there's been more stakeholders involved in each of these deal cycles. We talk about the things that make the macro hard. One of those major things is that there's a lot more kind of multi-team scrutiny on every buying decision that's happening in software, and that includes technical buyers and technical buyers want to be able to use products. And so even in the, I think, people's initial reaction to the free trial announcement was that it was a lot about the low end and kind of smaller deal sizes. But actually, what we've been seeing over the last year is that in the enterprise, in particular, the need for something like a Sandbox experience has like really increased in importance. And so we've been providing those on a one-off basis and a lot of times that requires kind of careful attention from the sales team or the solutions consultants as they're going through that buying cycle. What we've done over the last year through that product like growth motion is that we built these automatic onboarding campaigns, we've templatized things campaigns and campuses. We've really worked through that early education journey and been able to automate all of that. And so we've done that in service to being able to kind of have these lower effort opportunities for, in particular, technical teams to be able to go in and try out Brave as they're in their purchasing journey. And so certainly we think that will be a source of increased kind of demands and -- in the commercial side of the business over the long term, but the immediate impact is really helping us out more in these enterprise deal cycles.

Yes. So one of the points you made earlier just around the credit model. I thought seem very logical in terms of reducing friction in these deals. Where are we at in terms of the rollout? And is that something you expect the majority of your customers to move to over time?

So we made it available right at the end of Q1. In Q2, it was the vast majority of purchases of any of those panels that I mentioned went through the flexible credit model, and we're effectively establishing it as a default moving forward.

Yes. Have you noticed any changes in customer behavior that hey; now that it's more fungible, I have no problem buying more because I can use the SMS credit for WhatsApp or...

Yes. Yes. I think there's less -- there's kind of less consternation around the commitments because people know that there's a lot of flexibility in how they're going to be able to use it. The other thing that we're more excited about is that it opens the door to automatic experimentation with new channels as they come out. Before you may have like bought Braze and you bought SMS and all these channels and then WhatsApp comes out. And you conceptually think of those as similar channels, right? They're both like kind of premium phone number-based messaging, and it just depends on which country or market you're in or what permission you have with a given customer. But before we would have required that you sign a whole new order form and make a whole new commitment and maybe you already used all of your budget allocation for the SMS before and you would shift that over to Whatsapp, but the way our contract structure work, we wouldn't have allowed you to do that. And so that's obviously customer unfriendly in those cases, and we're out of the woods on those things now. And so I think that it's going to benefit in a lot of different ways. And that's why we've moved so fast to establish those the default since we launched it.

Great. As you think about the second half of the year, last night, you raised the full year guidance, I think, by $0.5 million more than you beat in the quarter. So effectively kind of passing through the beat maybe slightly more. Talk to us how you're seeing the pipeline shape up. Obviously, you've got elections this year, maybe a little bit more macro uncertainty versus this time a year ago. What assumptions, what type of conservatism are you applying to that pipeline?

Yes. So no material change in the level of conservatism. So I just want to be clear that the sort of being closer to the pin to consider that to be persistent through the back half of this year. We are comfortable with sort of where the pipeline is shaping up. But the sales cycles are continuing to be elongated and challenging and lots of stakeholders and you're seeing CFOs get involved in the conversations. So it's not as frictionless. And we're still, like I said, there are pockets where we're seeing some kind of goodness in strength and then other places where there's ongoing weakness and the places of strength are not persistent. And so I would say the back half of the year, I wouldn't look for any kind of acceleration on anything. I know that what's in the guide implies some ongoing deceleration. You can see ongoing weakness in things like dollar-based net retention. The churn continues to remain at elevated levels, not levels that are surprising or outside of the realm of what we budgeted for, even if I rewind the clock back to like our January, February time frame when we were setting the budget up for the year. But it's higher than we would like it to be on a consistent go-forward basis. And so we are very comfortable that we can kind of live within the confines of the current environment, but we expect or we would hope for a better environment in future periods.

Right. And as you think about the contributions from new logos, we talked a minute ago about the new trials the Braze for start-up plan that you've talked about on the earnings call. Where do you sort of aspire to get that new logo contribution back to in terms of growth? And is this race for startups? Is that sort of helping drive that?

Yes. So I think in order to kind of get the full picture here, you need go back a little bit more than a year ago when we actually -- we did a reduction in our sales capacity, and it was primarily focused on the commercial side of the business. And a big part of that was due to just the relative acquisition efficiency across these different categories. We're feeling good about where the acquisition efficiency was from a kind of customer lifetime value, retention and acquisition cost standpoint, within the enterprise category, but we knew that SMB was going to be a much tougher place moving into the future. We had seen that the venture dollars had dried up and there was obviously a lot of business failure that was pending and actually if you've seen numbers from places like [indiscernible] that have been published, that small business failure has only accelerated this year, right? They're actually highest number of kind of shutdown start-ups happened in Q1 and then Q2 of this year. And so that kind of allocation of sales capacity has certainly led to a pressure on the net new customers because we count customers at the ultimate parent level. So a lot of the expansion into the enterprise. We mentioned these 3, 8-figure deals that we've achieved over the last 2 quarters. Those don't add anything to the customer count, right? Even though, obviously, those are great wins as we continue to expand. And on the same side, we've seen that pressure on the SMB side. And then what we've done in the meantime is that we've really gone back and said, okay, before we reaccelerated investment into the commercial side, the SMB and the scale sectors. We want to make sure that we can really kind of structurally improve the efficiency of onboarding and bringing on those customers. And so that's where a lot of the work from a product like growth motion. You've heard me talk about the important investments we've made from a usability standpoint in each of the last 4 quarters as we continue to work on the on ramp in the Braze, the education side of it, making sure that as a customer is ready to adopt Braze that they can step into this more sophisticated approach around customer engagement, with less friction, right, and less work like manual work on our side. And so I think what you see in the launch of the free trial and the Braze's startup's program is also just a testament to the confidence that we have that we've really structurally improved that kind of onboarding and the usability journey. We've got a really robust partner ecosystem now where we can bring people into partner-led onboarding so that Braze doesn't need to have integration and onboarding teams kind of in a stand ready fashion as the as new customers kind of ebb and flow in and high volume as they would have before. We now have a really robust partner motion with a lot of digital agencies around the world that are ready and on standby to be able to help onboard and integrate all of those customers as they come in. And so I think that it's really a culmination of a long-term investment strategy to be able to say, how do we fundamentally improve the acquisition efficiency in the commercial side of the business before we go and pour any fuel onto that fire. And so we're excited to see what this does over the next couple of quarters, and we think that it could usher in a new -- kind of a new generation of how we approach the commercial side of the market, and we're really excited about it.

Okay. Great. Earlier, you talked about some improvements in win rates. And I wanted to double-click on that for a moment. I think you -- we've been hearing more and more displacements of legacy marketing clouds, including Salesforce. At the same time, one of the big themes we've heard at this conference, we had some big companies like Microsoft, ServiceNow yesterday, just this topic of vendor consolidation. So on one hand, you are larger than many of your smaller competitors in the marketing space, but there are certainly companies bigger than you. So how do you sort of see that vendor consolidation theme? And where specifically are you seeing those uptick in win rates?

Yes, it's a good insight because when we talk about consolidation, there's obviously a lot of different places you can kind of draw the abstraction layer, right? And while we are certainly not a point solution within the messaging space we also don't -- we don't have 12 clouds that we're trying to kind of bundle together and sell to people either. And so a few things there. One is that definitely, when we look at the problem of customer engagement, and being able to communicate across touch point -- across all the different touch points that are relevant to a consumer, be able to collect data from all the relevant data sources and action on it real time and be able to do that around the world, right, because the relevant set of channels for a given consumer changes from country to country or region to region, generally due to just local preferences or various kind of components of history in terms of how technology developed within those areas. And so in order to really comprehensively deliver on those, you need both the breadth and the depth of Braze's kind of channel set as well as the sophistication of the orchestration above it and the vertical integration that we can deliver there. And then, of course, more recently, we announced the launch of the Braze data platform, which is really helping bridge some of the gaps that would have been there before between the data as it's generated or the insights as they're being produced by your data science or machine learning teams and then getting them into a place where you take action on them, right, which is obviously what Braze's bread and butter is. And so that consolidation trend, what it looks like most often is that we'll go into an organization who has different vendors for e-mail and SMS and being able to deliver surveys and sending push notifications and what have you, and they might even have like different ones depending on what country or what brand they're in. We actually work with a telco in the U.K. where we've replaced 25 different point solutions across their different brands and geos and channels that they were working with. And so there's a tremendous amount of great examples of that. And when you look at all 3 of those 8-figure contract expansions that we've spoken about over the last couple of quarters, in one case, that was the continued expansion across additional franchises, brands and geos. And in another case, it was expansion across additional channels, expanding into e-mail in a place where we'd previously primarily been in product and mobile or through push notifications and the third one, expanding into new titles within a bigger family brands. And so those are all great examples where that consolidation play is at work across several different dimensions. We do still see occasionally and especially in places like the United States where companies like Salesforce or Adobe have really strong entrenched relationships with the GSIs, in particular that a company might go through a massive digital transformation effort and they're just going to kind of go wall-to-wall, right? And those are areas where we've got a lot of experience kind of breaking into wall-to-wall deployments and displacing the likes of the Salesforce Marketing Cloud or Adobe Campaign within that I think that when we look at those examples actually Salesforce obviously super focused on the Data Cloud right now. The Marketing Cloud, not getting a lot of innovation, a lot of love. It gets bundled in. But I think when we look at that long-term picture, it's like, yes, there certainly are situations where that bundling or that kind of wall-to-wall deployment can block us out. But I don't while that's annoying, I don't worry about that in the long term because it's not driving additional innovation. And I think this problem space is getting harder and harder every year. And so I think that the gap between what they are able to deliver for customer engagement and what the market is demanding is only widening over time. And it's really Braze's vision that's leading that market. And so I think that's the -- those are the broad themes that we see at play and make the mistake, there's absolutely still incumbent advantage that comes from some of these really big like cloud deployments or what have you. But within the customer engagement space, I feel really good about the comprehensiveness of Braze's solution, our ability to really deliver an optimal -- total cost of ownership and time to value story within that and then, of course, differentiated capabilities as well. And we only benefit from just how intensely competitive, especially a lot of these global digital markets continue to be because your strategies from 5 years ago are just like they're not good enough anymore, right? And what that means is that more and more companies, it might take them time to kind of understand the competitive pressure, especially if you're in a highly regulated industry, like you're in financial services or you're in big airlines or what have you. And you see that in our customer base where we work with a lot of fintech challenger banks. We work with a lot of like payments companies, et cetera, as well as like credit unions and regional banks like all over the world, but we don't yet have a meaningful footprint in the too big to fail banks, right? And they tend to be able to operate more conservatively. Similarly, we work with private jet charter companies, and we work with travel booking, vacation booking and we work with a bunch of budget airlines. They tend to be more nimble and more agile and they move more quickly because they're in more competitive spaces -- like airlines have got more regulation, stronger commit advantage. And so it's taken us longer to be able to kind of land those as customers. But when we look at that kind of the longitudinal picture there, obviously, we're building up the right proof points with the companies that are really pushing the edge of innovation that are highly competitive in those spaces. And that gives us a lot of confidence that we not only have a lot of room to run in some of these really gigantic categories, right, where we still have a lot of the enterprise to penetrate, but we've got all the right proof points that we're going to be ready to do that. So...

Yes. I'd be remiss if we didn't talk about margins a little bit, Isabelle. I think this quarter was -- there were a couple of positive surprises on the margins. Both gross margins picked up nicely sequentially in year-over-year and operating margins were a lot better. All the exciting stuff that Bill has been talking about, I think, may sometimes get the complexity of managing the costs also may sometimes get overlooked when you're talking about tokens and just this really complex orchestration layer. Give us a sense for some of the efficiencies that are underway. How are we able to deliver kind of the upside performance you saw this last quarter?

Yes. So on the gross margin front, if we rewind the thought to Q1, you'll remember, we sort of announced that we had hit a bit of a low. And what we had indicated is, look, we're -- there was a concentrated volume messaging situation that sort of occurred in Q1 that we knew we were going to work ourselves out of. Plus, we were working across our engineering organization to continue to find performance efficiencies for the platform. The messaging component worked itself out. So -- and what I've said is we were going to sort of monotonically increase from there. What I would say is we've been wildly successful in the efficiencies that we found through the engineering organization, both on our hosting and kind of other components of the technology stack, both from kind of a contract renegotiation standpoint as well as just our own adherence to internal policies around data retention and the performance of the platform. So that's been great to see. I would not from here, assume there is further material monotonic improvement over the course of the coming quarters. Plus, I remind everybody that Q4 experiences the seasonal low because of the activity. I think as we look down the P&L, the operating efficiencies that we continue to get on operating income Q1 naturally has a seasonal a bit of a bump, same as Q3. For Q1, it's related to our kickoff meetings and some sales-related enablement that happens right at the beginning of the year. Q2 obviously doesn't have that. So that's part of the reason why the cost steps down. Back half of the year, we have the impact of comp increases that go into effect in August. So you'll see that impact as well as our Forge, our annual global customer conference and some other global events that are happening throughout the year so -- or throughout the quarter. So what's the reason why we'll kind of step back, but -- and as Bill and I have both mentioned, some of the investments that we've been making as we've continued to increase headcount have been in strategic locations with a differentiated cost structure. So both on kind of technology investment side as well as the human side, we are finding ways to scale more efficiently across the organization. And we're going to continue to do that while continuing to invest in places like R&D so that we can kind of keep the engine and the flywheel going from a product differentiation perspective.

Great. Right. Well, I feel like I only got through about 1/3 of the questions because this is such an insightful discussion. But we're are out of time. Bill and Isabelle, thank you very much for being here the day after you reported earnings. And thanks, everyone, for coming to the presentation.

Yes, thanks for having us.

Thank you.