Twilio Inc. (TWLO) Earnings Call Transcript & Summary

Thanks, everyone, for sticking with us. Day 2 of the Wells Fargo TMT Summit. I'm Michael Turrin, the software analyst. With us, we have Khozema Shipchandler from Twilio, COO of the company. Good, packed environment to take it all and even though there's a lot going on. So that's a good sign in terms of engagement. There's been a lot of metrics, a lot of disclosure, a lot of things have happened with Twilio over the past month or so. So maybe we can start with Q3, the puts and takes of what you reported and then we can get into some metrics. We'll start with the top line and work through the rest.

Yes, that sounds good. First of all, thanks for having me. It's good to be here.

Thanks for joining.

So I mean, the Q3 metrics, the growth profile continues to hold in relatively strong. I think you're not alone in calling attention to some of the macro considerations that you're seeing. But maybe just help us unpack or talk through what you're seeing in Q3, what informs guidance for Q4 and just how to think about just the moving pieces of the business model.

Yes. I mean I'd say maybe just to start off with on the top line, I mean, we're just seeing a very different macro environment that's unfolded over really just a relatively short period of time than what we've seen previously. And I think you see it in what some other companies are talking about. I think our business is particularly sensitive to what ends up happening with end consumers. And so we've been really paying very close attention to like what retail is doing, what happened over the most recent holiday period, obviously. And as those things have played out, I would say, in an increasingly worse way, that had some impact on our business as well. I think when we looked at it maybe 120 days ago or so or even just prior to that, I think we were starting to see some possible headwinds. We called out at the time, in particular, like crypto was like kind of an obvious one that was really starting to come down. I think that's like effectively concluded now, and we'll see how that kind of shakes out. But we also saw some stuff around social, for example. And I think we're just starting to see all of those things manifest. I mean we still see healthy growth levels, certainly not the growth levels that we want or that we've been previously forecasting. But I think, given the macro environment, we still feel good about the results. I do realize we're getting a lot of important feedback from investors, so I don't want to skirt that topic either, and we are listening. And obviously, there's limited to what we can do on the top line, but we're obviously watching the bottom line pretty carefully, too.

Yes, so I mean you put some parameters in place, you made some hard decisions, you've taken some controls. There was a headcount reduction. There's a margin target for next year. How much of that is just informed by the signals that you're seeing in the demand environment, the investor feedback? And how do you make sure that you're not attacking too far in one direction if the world turns in a different direction?

Yes, I don't -- the macro is actually not informing that stuff too much. I think the reality is that, at our size and scale, kind of the #1 conversation that we've been having with investors for some time is that, well, at a nearly $4 billion run rate, like it's time to be a profitable company. We obviously agree with that. And I think we've been saying for a while now that we intend to be a profitable company in 2023. And I think what we're working through is, obviously, given some of the feedback that we're receiving is like what additional parameters can we potentially put around, what that profitability looks like in 2023, how much is there, et cetera. But I don't really think the macro is impacting that too much. Obviously, you've got other macro factors like with interest rates and what have you. But I think, at our size and scale, it's really important for us to be a profitable company. I think it's really important for us to be able to grow organically, not rely on acquisitions that we've done over the last several years. And I'd take it one step there and say that it's also important for us to be able to see a path and to be able to illustrate a path to investors around GAAP profitability as well. I mean we -- and obviously, the big lever in the middle is stock-based compensation, if you strip out intangibles at least. And that's a topic that we're pretty focused on as well. There's some math that makes that one hard to bring down like instantaneously, but we are focused on it. And I think, by having a lower headcount profile, that will help.

There's a few different directions I could take this, but, on the gross margin side, it's always been, I think, the case of the messaging business has just grown so fast. It's remained a disproportionate impact on the mix. How much of that is viewed as just that's the input that you work with and then the controls are on the OpEx side? Or are there things you can do on the messaging side to stabilize, drive improvement in the margin profile? And maybe talk through some of the stats that you gave on messaging. Because there's a lot of detail at the Investor Day around the economics of U.S. versus international. So what the takeaways for investors were intended to be just with that additional disclosure?

Yes. I think the bottom line is -- and I'll unpack it in a few different ways for you, but I think the bottom line is, is that we feel that messaging is a good business to be in. It does have structurally difficult gross margins relative to the rest of our portfolio. Those gross margins are obviously lower than when you comp against kind of a traditional SaaS software company. There's one reason for that, which is carrier fees. And so what we tried to do as best as we could was basically say, "Look, we've had carrier fees go in a number of different directions over the last several years. You've had layers of them actually get added over the last couple of years. And in spite of any of that, in spite of the fact that we've been mixing more internationally, which has even lower structural gross margins, this is a good business to be in." And so what we tried to do, in particular, was to break out what are the unit economics of messages as they end up terminating internationally, if they end up terminating in the U.S. markets. What do they look like? Are they stable? Because I think one of the issues that we've been hearing a lot from investors was, "Look, are you basically buying bad business? Are you chasing growth and not making any money on that growth?" Or is there incredible price pressure? And neither of those things is really true, right? I mean the pricing, I think, is vetted by the fact that the unit economics are very strong. And I think, on the profitability side, I mean, we are very, very convicted about delivering profitability next year. And given how large a piece of the business that, that is, it's going to be the big driver of that.

I mean we've talked about marketing for a long time. And obviously, there are headwinds around marketing spend currently as well. But you have evidence you have 27 customers that are spending $10 million with Twilio currently. I'm just curious if you could just spend more time on what informs the importance of vesting. You have a gigantic customer base that you built on the back of messaging. A lot more companies have gotten comfortable with that as a very valuable way to reach end consumers. So just spending a bit more time on, I think, the importance in what you're realizing there and then we can go into what it builds on top of.

Yes, I'd say it's kind of two or threefold. I think the first is sort of obvious, I guess, in the sense that we have nearly 300,000 customers, right? Every single one of those customers at some time, and maybe it's early with many of them, but at some time there's an opportunity for us to convert one of those customers that's currently using messaging and, in many cases, using e-mail into a flex customer, into a segment customer. And if you look at the customer journey that we've often been on with some of these logos that end up being in excess of $10 million, like that's where they start, right? They start using one of our simpler use cases. They happen to really like it as they've been using it, and they continue growing with us and they grow into some of these larger areas. And so if it creates a relationship with a customer in the first place, and it's an opportunity for us to gain share, as long as we're doing it profitably and see a path one day to growing them to something else, I think that's kind of an overall positive. And I think we like that about the messaging business. I think the second dynamic is, is that when you think about delivering a marketing experience, it has to be done ultimately through some communications vehicle, right? So any of the kind of pure-play marketers, if you will, they all use communications eventually, right, to be able to get these messages out to consumers. And so for us, it's like, "Well, we own this amazing comms business. We're the market leader in that business. We have the best data platform in the world. The technology is excellent. And so doesn't it make just logical sense for us to kind of pair these things together and deliver a unique experience back to consumer using rich data paired with great communications?" And then it's not really kind of directly part of the whole thing, but I think what also is increasingly true is, even with our contact center product, what we're starting to see is that -- I mean most people don't want a contact center experience that has no communication in it, right? Like a lot of it happens through voice, a lot of it increasingly is moving over to text. But what's even interesting about that is, is that, increasingly, a lot of that is going the way of data to, and I think there's an opportunity for us to add more value to customers that use the contact center product.

One of the questions we often get -- I'm sure you often get is just around, is selling the software applications piece very different than the core selling of the messaging piece was because there was such a strong inbound developer presence? And the way that, that was adopted was viral. And I think there's always been an optimistic case that developers have more pull and will have more decision-making and marketing decisions in other areas. In terms of the go-to-market motion and ability to supplement that and how important that is on the software selling side, what's your perspective as the portfolio expands? And is that something -- are you able to do both and do both and also work towards cost controls?

Yes. I think there's 2 different questions there, sort of. I think the short answer is yes, they're different, okay? And I think the easiest way to think about that is in terms of who the buyer is, right? So a lot of, as you pointed out, messaging and even e-mail is developer-led. In our e-mail business, when we bought the business, since we've grown the business, we've never really had that much sales attached to it. It kind of sells itself. I mean it's a great console. It's a great product. It's easy to use, and you're up and running. And SMS is more or less kind of the same thing. Now can you make it larger with salespeople? Absolutely, right? And so I think dialing in the right number is really important. But by and large, your customers, they start out being developers, they may grow into enterprise buyers over time. But generally speaking, the developer kind of drives that sale. I think when you look at both flex and segment, you have actually 2 different buyers there as well, right? Like your segment buyer, I think it tends to be your CMO, and she's often the decision-maker. I mean you're talking about leveraging deep personal data about consumers using that in a way to create a rich consumer experience, like she's typically the person that's going to be making that buying decision. It's a different sales cycle. It takes a different persona. It takes a different person to be able to reach the CMO. So I think we have found that, that is quite different. And we sort of came clean on some of the issues that we've had in the sales motion with segment recently by combining the sales forces. And so that's one of the reasons that we kind of broke that apart again. And then I think Flex is yet again different. I think it tends to be the head of operations. It tends to be head of customer service. It could be the CIO. We don't really find that that's the case typically, but it's possible, certainly a stakeholder in the process. And I think the common thread, if there is one across these, is at the enterprise level. At the enterprise level, like these are big dollar decisions, and so they tend to get made by senior people regardless. And so I think where we've seen synergy is when it's an enterprise-oriented sale, you can sell any one of those kind of to the same people because you're dealing with the same people. When it's below that, I think we're not finding the synergy, and so it does make sense to kind of conceive of those sales organizations a little bit differently. Now in terms of cost, which I think is the other part of your question, I think the trick is for us how do we dial the sales organizations that call on these different areas and use their skills to call on the different personas, but then have a common shared service underneath that, if you will, right? So common sales ops, common systems, common tools, common marketing, what have you.

And then -- so does that mean -- is there a way to construct an industry overlay to get Segment and Flex to talk to certain enterprise buyers? Or what does that mean in terms of the composition? I'm just curious what the decisions have looked like that have led to...

I don't think you need much of an industry overlay. I think the area that we found success and that kind of took off is probably more in health care. I think it is true that you have like big financial services buyers, obviously. You have big retail buyers. I don't think we have to organize the sales force that way necessarily. I think health care is a little bit different because you've got different aspects of privacy and HIPAA, obviously. So you need a little bit different persona there. But I think for the rest of it, it's kind of the same kind of individual that we end up hiring. It doesn't have to be that specialized.

Let's come up for a second and talk about just impressions of Segment since the acquisition outside of the go-to-market and just some of the things that are happening there. You mentioned it's a best-in-class data platform. That's blessed by a lot of industry experts that we talk to as well. So maybe more on the product, the learnings since you've worked to integrate the solution within Twilio's portfolio.

Yes, I guess, the way I'd characterize it is that the product is even better than we thought it would be. I think that they actually had a decent sales motion that we screwed up, frankly. I think we could have done a much better job in the integration of that, and we didn't. We thought by pulling it back into the generalist sales force that, that would help, and that turned out not to be the case. I think the resonance with big marketing companies like -- meaning big retailers, big financial services industries or institutions. I think the resonance of a marketing product that combines the CDP with communications is really, really powerful. I think we underestimated the amount of engagement -- no pun intended -- that we would have with Engage, so that's been really interesting. And I think the team is even better than we thought they would be when we acquired the company.

Is it -- do you have a view, I'm sure you do, on are we so early in CDP that evangelism is more difficult because everyone is taking longer to evaluate new decisions? Or is it so focused and personalized on where marketers ultimately want to go that there's something that you can lean into that can help drive Segment towards greater growth in the 30-plus percent you're targeting on the software side?

Yes. I think, by and large, everybody knows that they want to better use data. I think, to be honest, like the greatest friction in the sales cycle is someone has a built-themselves solution, a DIY solution, and they like it. And so we just kind of wait it out, and we'll show up again later. And I think the technology is superior. That's really, really helpful, obviously. It pairs really well with the communications. That's super helpful as we talked about earlier. And I think it just takes time. And I think we'll grow the reps for that part of the business in a responsible way. I think we see great growth opportunities. Obviously, it's a choppy environment right now. But I think in spite of that we'll play through. And I think we'll be able to reaccelerate that business, I think it will be fine.

Can you talk more about just Engage and what else you're doing on the marketing side? I think Segment and Flex we've touched on a little bit, but the remainder of the software portfolio, just things you're excited about, more of the 4 leading things kind of getting out of the meat of what's happening now and the metrics, more on just the forward-looking vision.

Yes, I mean, look, I'd say, in general, we are very excited about Engage. No question. I'll come back to that in one second. I think that -- we're also just very excited to grow the products that we have. I think what -- over the last couple of years, we pioneered a lot of interesting new features. We've pioneered a lot of interesting new products. And I think we have a great and very disruptive contact center product today that every time we win we end up displacing one of the other guys, right? So the growth characteristics there we really like. We like the margin characteristics there. I think we just have to grow it at a faster rate so that it becomes a bigger proportion of the business. I think the same is kind of true with Segment. Again, like I think we know what works there. We have to kind of go back to that in a weird way. But we know it works, and we just need to grow that part of the business faster. So in terms of like where we're all focused as a management team, like we spend a tremendous amount of time talking and thinking about, okay, we have this great data asset. We have this great disruptive contact center product. How do we continue accelerating the growth of those 2 things? Yes, we want to innovate on a bunch of other things, too, but the focus is really on those 2 things. Now to answer your question on Engage. I mean I think Engage is an important new product from the perspective that it's the first one really that combines all of the products within the Twilio umbrella, right? Through a singular API, you can deliver this really unique experience that previously was only available in the physical world and that in a kind of a post-COVID environment where some of our habits have changed, but many have not, it's really, really important to be able to utilize data in a special way for a consumer to be able to deliver a really interesting experience, which creates loyalty. And I think that's what Engage is now -- finally, the product that's able to deliver that.

I want to go back to something that you said on Flex. You mentioned that you're often displacing legacy vendors in the contact center space. Is it more often the case that Flex comes in as a supplementary layer given you can flex up, flex down contact center agents within certain seasonal profiles? Or is it a full-scale rip and replace currently in the way that you're generally approaching customers?

Yes, it's a great question. It's all of the above, honestly. I think, initially, it was much more augmentation. And so you'd go to an airline, let's say, and they would actually have multiple contact center organizations, right? You'd have customer service. You'd have -- some of that, that would face consumers, some of that, that would face other parts of the airline industry. And in many of those cases, what we would initially -- just as an example, what we would see is they'd say, "Look, we like your technology a lot. We're going to try it here. And then if it works, we'll leverage it in some other part of the organization." Or the other thing that we would see a lot of is, "Look, we're going to keep our current contact center product, but we're going to kind of run features of your side-by-side with it to augment what we do. And then if we like it, we'll grow from there." And I think, in both cases, we're fine either way because they both allow us to grow faster. I think, increasingly, what we're starting to see is a lot more rip and replace. Now the one thing that we're cautious about with customers, and we were -- we caution them to approach it this way is that if you're running 20,000-plus agents, it's probably not a great idea to flip it on day 1, right? So we'll graduate there. We'll go 500 seats. We'll go 5,000 seats, then we'll go 10,000, 20,000. So it's meant to be a rip and replace, but it's a gradual progression, if that makes sense.

You gave an interesting stat on Flex as well on the -- just the [ Kim O'Reilly ] has characterized that there was a slide that mentioned that this is Flex revenue, but there's also a pull-through where you're capturing additional revenue on the Twilio platform from those Flex seats. Can you walk through just the additional economics that you can realize beyond Flex?

Yes. Basically, there's an attach that goes along with the pure software sale. I think we actually -- we've debated this at times internally. Do we provide kind of a pure-play software number? How do we think about the attach, et cetera? So we've just decided to provide both to make it, hopefully, clearer. But the way that we kind of conceptualize the business is those messaging sales or those voice sales or whatever they are, they wouldn't have happened had there not been a contact center play in the first place. It's probably how most kind of contact center companies evaluate themselves or present themselves to the public, and so we provided that with it. And typically, the way that it works is, is that for every dollar that you'll end up having on Flex, you'll probably spend another $0.70 on either voice or messaging or e-mail.

One of the other questions we've gotten from investors since the Investor Day session was just on the changes that you're making and the margin improvements and a question on if they might not imply that it's actually potentially more expansion than what the current targets assume. And I don't expect you to change your stance on what the current targets assume, but can you just talk through some of the cost-related impacts that you mentioned? There was real estate, moderation of new hiring. There are a few things that you laid out. And then just thinking about the sequencing of when margin progression from some of those things should start to show through in the model.

Yes, I mean there's a lot in that question. And I want to be careful, obviously, about kind of restating anything either. But I think -- first of all, I mean, we feel quite good about like the medium-term targets, okay? And then, obviously, we want to be on the higher end of those, okay? We provided ranges, but it would, of course, be our aim to the extent possible to do better than that. I mean that's always the goal of every company, obviously. And I think the short term is just really choppy. And we're seeing, I think, a macro environment that we just couldn't have imagined. We certainly didn't think about when we put out kind of our sort of a couple-of-year ago 30% annual growth targets on an organic basis. And that's giving us a little bit of pause as we think about what transpires in terms of operating margin in 2023. All that said, I think that what we've done on headcount, what we've done in real estate, some of the other kind of other structural actions that we've taken internally, they all yield a really, really good setup in spite of what the macro environment is going to be in 2023. And we expect all of those things -- it's not all going to happen on January 1, obviously. But through the balance of the year, all of that stuff is going to bleed into the bottom line and I think generate good profitability. And I think it creates a good base off of which we'll get even more profitable in following periods. I think that the knock-on effect to some of those things, too, is that some of -- because we have lower headcount, because some of the moves that we've made also yield transfers from high cost to low cost as well as, in some cases, a shift to managed services, which don't involve headcount. I think all of that is an upside for stock-based compensation as well, which we've got to burn through the math, and I think we actually have to probably do a better job of just explaining like what our math problem is in the short term. But I think over the medium term, like we do see an -- a significant improvement that will transpire in stock-based compensation as well. And it just has to happen at some level because you have lower headcount. The final thing I'd say about it all is that, I mean, obviously, we've gotten a lot of feedback from investors. And I mean I'm -- it's not maybe about what transpired with the stock either, obviously. And I think we are listening, and that feedback has been helpful. And I think it's causing us to kind of think about like how might we present the company differently and what things might we think about doing differently, but that feedback has been really constructive, and I think they're done in the right spirit.

Just time for a couple more. I think focusing on just the path into 2023. One question we've been asking CFOs throughout the event is more just less on the Twilio investor-facing side and more on just the internal CFO side. One of the things we're hearing is just CFOs are more involved than signing up with sales processes and things. And so, just from your perspective, are there things that you're more focused on from a planning perspective heading into the coming year than was the case?

I wouldn't say there's been like a really marked change or material change versus prior periods. I mean, I've always been pretty involved in most of those things. So not really, not really.

And then the last question is for the investor-facing side, the things you're focused on that define success for the next year. We've talked about some of the baselines that you're setting, but we come back here and have this conversation next year. The metrics, I know you're contemplating, looking at a few different things there, but what the mile markers that are going to inform your perspective and how you define success going forward?

Yes. I think for us, probably the most important thing is that we've got to not just become a profitable company. But for that profit to give an indication to folks that we can be an incredibly profitable company over a long and sustained period of time. And then I think when the macro subsides that there are good signals, especially in the software side of our business where we put so much emphasis that there are good signs that we can reaccelerate growth there and that there's a really valuable franchise that's starting to build there. And I think, for us, that fundamentally means Segment and Flex. And I think for us on the communications side, especially messaging, it means we basically have to throw off a tremendous amount of profit so that we can keep growing in the other parts of the business in the way that we think we can.

It's a great point to close on, Khozema. Thanks for making the time. Appreciate you coming to Vegas.

Yes. Thanks for having me.