Twilio Inc (TWLO) 2017 Q2 法說會逐字稿

Good afternoon, and welcome to Twilio's Q2 2017 Earnings Conference Call.

My name is Kelly, and I'll be your operator for today's call.

(Operator Instructions) I will now turn the call over to Greg Kleiner, Vice President of Investor Relations and Treasurer.

Mr. Kleiner, you may begin.

Thank you.

Good afternoon, everyone, and welcome to Twilio's Second Quarter 2017 Earnings Conference Call.

Joining me today are Jeff Lawson, Twilio's Co-Founder and CEO; and Lee Kirkpatrick, Twilio's CFO.

The primary purpose of today's call is to provide you with information regarding our 2017 second quarter performance in addition to our financial outlook for our 2017 third quarter and full year.

Some of our discussion and responses to your questions may contain forward-looking statements, including, but not limited to, statements regarding our future performance, including our financial outlook, impacts and expected results from changes in our relationship with our largest customer; our market opportunity and market trends; the growth of our customer base; customer adoption of our products; our momentum; the benefits from our business model, timing and focus of expenses; our delivery of new products and new product features; and our ability to execute on our vision.

These statements are subject to risks, uncertainties and assumptions.

Should any of these risks or uncertainties materialize or should any of our assumptions, as outlined in our earnings release and the documents referred to in that release, prove to be incorrect, actual company results could differ materially from these forward-looking statements.

A discussion of the risks and uncertainties related to our business is contained in our Form 10-Q filed with the SEC on May 10, 2017, and our remarks during today's discussion should be considered to incorporate this information by reference.

Forward-looking statements represent our beliefs and assumptions only as of the date of such statements were made.

We undertake no obligation to update any forward-looking statements made during this call to reflect events or circumstances after today or to reflect new information or the incurrence of unanticipated events except as required by law.

Also during this call, we may present both GAAP and non-GAAP financial measures.

Reconciliations to the most directly comparable GAAP financial measures are available on our earnings release, which we issued a short time ago.

We encourage you to read our earnings release as it contains important information about GAAP and non-GAAP results as well as the reasons why we present guidance for non-GAAP financial measures of loss from operations and net loss per share but not the comparable GAAP measures.

The earnings release is available on the Investor Relations page of our website and is part of a Form 8-K furnished to the SEC.

Finally, at times in our prepared comments or in responses to your questions, we may offer incremental metrics to provide a greater insight into the dynamics of our business or our quarterly or annual results.

Please be advised that additional detail may be onetime in nature, and we may or may not provide an update in the future on these metrics.

I encourage you to visit our Investor Relations website at investors.twilio.com to access our earnings release, periodic SEC reports, webcast replay of today's call or to learn more about Twilio.

I'll now turn the call over to Jeff.

Thanks, Greg, and welcome, everybody.

One of the founding principles of Twilio was that creating better communications experiences was a mere universal business need.

Today, virtually every type of company is changing the way they engage with their customers, and we saw that firsthand this quarter.

Companies at a wide variety of traditional verticals, including fast food, automotive, supermarket, media, education and video games, chose to adopt Twilio to improve their customer engagement in Q2.

Our relentless focus on customer success, along with the product and go-to-market efforts we've been putting into place to build upon our developer-first approach, continues to drive our growth.

We feel great about the momentum of the business as we head into the second half of the year.

The business performed well in the second quarter as we saw further growth with both new and existing customers.

Our base revenue grew by 55% year-over-year in the second quarter.

Without Uber, base revenue grew even faster, 65% compared to Q2 of last year.

We also planted important seeds for future growth by adding more than 2,700 active customer accounts in the quarter.

We recently held our SIGNAL Conference, where we showcase our customers and outline our vision for the future of this market.

We had a number of fantastic speakers, including customers like Salesforce, Netflix, Lyft and Weight Watchers, who participated in our general sessions, and many more in breakouts over the course of 2 days.

SIGNAL is a celebration of our customers' success and of what's possible by leveraging the power and functionality of our platform.

With roughly half of our headcount in R&D, we invest a tremendous amount of money to innovate on behalf of our customers, and these investments provide tangible benefits for our customers that our team loves to put on display at SIGNAL.

We often talk about the concept of agility with resiliency within Twilio.

It's the idea that you can deliver both rapid innovation and high quality without sacrificing one for the other.

Our engineering teams are on track to ship approximately 30,000 production deployments of Twilio software this year with a new feature or product coming every 3.5 days on average.

And all of these improvements come while maintaining an industry-leading five nines performance across not only API availability but more importantly, our API success rate.

This devotion to providing value to our customers continues to drive our business and expand our leadership position.

We announced at SIGNAL that we now have roughly 1.6 million developers registered on our platform, up from 1 million at SIGNAL just a year ago.

This continued developer momentum is parlaying into further traction with a wide variety of companies, including the traditional enterprise.

For example, at SIGNAL this year, we created an executive briefing center for the first time to discuss the strategic role of communications and to introduce companies to the Engagement Cloud, the new layer of our software stack.

We had a packed schedule resulting in some great discussions with both new and existing customers, including some global brands engaging with Twilio for the first time.

SIGNAL is also a forum for us to chart the future of our growing platform of products.

We had a total of 42 product announcements around SIGNAL this year, including Twilio Functions, our serverless computing product; new analytics products; a voice recognition product we built in conjunction with Google; and support for several new third-party communications channels like Alexa, Slack, Twitter and e-mail for companies who want to communicate with their customers in a growing list of new mediums.

We also introduced a machine learning product called Understand that will allow our customers to capture the meaning behind their customer interactions and drive actions based on that meeting.

Importantly, this natural language understanding product will work across platforms like voice, SMS, chat and other messaging channels, plus digital assistants such as Amazon Alexa, allowing customers to add natural language interactions to whatever channel they want to use to engage with their customers.

But the biggest highlight of SIGNAL was the launch of the Engagement Cloud, the next layer of our software stack.

Part of our core strategy is to provide a broad platform with lower-level building blocks and watch how our customers use them.

As we looked at all the things that our customers build on Twilio, nearly all of it involves some form of customer engagement.

The Engagement Cloud is a set of higher-level APIs that encompass best practices and accelerate our customers' ability to adopt Twilio.

You've heard us talk about use case APIs in the past.

Well, the Engagement Cloud formalizes this concept.

In broad terms, the Engagement Cloud is designed to help our customers build better ways to engage with their customers.

When we looked at all the ways in which companies communicate with their customers, they typically fall into 3 categories: system, department and individual.

Let me explain.

First, system communications.

That's where a company is providing an automated alert or notification such as "Your dentist appointment is tomorrow," or "Here's your PIN code to log in."

Next is department communications.

That's where a consumer's looking to talk to someone within a department at a company such as the sales team or the support team.

And then last is individual communications.

This is where a consumer is trying to talk to a specific person representing a company, not just anybody, like their driver or their banker.

And the software interaction within the Engagement Cloud works differently at a higher layer.

With the lower-level building blocks of our programmable communications cloud, a developer's instructing our platform how to do something like send an SMS to a particular number.

We call these imperative APIs.

Products at the Engagement Cloud layer operate around the concept of what the developer is trying to accomplish like notify the end customer that their package has shipped.

But the how part of that equation is now abstracted, and our software figures out the best way to send that message.

We call these declarative APIs, and they do more heavy lifting for our customers, allowing our customers to get to production faster.

So I mentioned there are 3 categories of engagement that companies have with their customers.

Within the system category, we have Authy and Notify as part of the Engagement Cloud.

Within the department category, we have TaskRouter.

And at SIGNAL, we announced the newest addition to the Engagement Cloud family, Proxy, to address the individual category of engagements.

See, over time, we've seen a number of companies from major retailers to dating websites, to ride-sharing companies want to enable their employees or contractors to interact with customers from their own mobile devices but to do so securely.

And they've built these so-called proxied communications with Twilio.

And now Proxy provides that functionality out of the box.

Building out the Engagement Cloud is a substantial software undertaking, once again demonstrating the capabilities of our R&D team to deliver further innovation on our platform and value for our customers, agility with resiliency.

With the Engagement Cloud, we make it easier for our customers to get to production faster and focus on the particulars of their company, not on reinventing the wheel.

The Engagement Cloud also expands the market of potential buyers and allows us to up level the conversation with our customers to the strategic value of how they engage with their customers.

The Engagement Cloud layer of products provides us with yet another point of differentiation in the market.

Overall, I'm incredibly proud of the experience that our team delivered for our customers and prospects in attendance at SIGNAL.

We're taking the SIGNAL experience to London again next month, so stay tuned for more announcements.

In terms of new business in Q2, we saw strength once again across new and existing customers as businesses of all shapes and sizes are adopting our platform.

Some highlights include lululemon, reddit, Aircall and Go-Jek.

On the last call, I mentioned the addition of George Hu as our new COO, and he's certainly hit the ground running as the new leader of our go-to-market effort.

He's added a level of alignment and focus that will serve us well as we look to continue to expand our market presence.

We will continue to invest to support our go-to-market efforts given the results that we're seeing.

Another new win from the last quarter was with Weight Watchers.

Today, Weight Watchers is an entire lifestyle program, helping members to eat better, move more and shift their mindset to live healthier lives.

Over the last 3 years, Weight Watchers has undergone a significant technology transformation to build more agile, flexible, cost-effective and modern systems to support its members and grow its business.

In the contact center, customer engagement that used to live in a single brick-and-mortar call center is now augmented by part-time leaders and coaches in distributed locations.

In addition to these distributed agents, they also need to expand their footprint from one channel, voice, to many channels.

So Weight Watchers has embarked on a multiphase call center modernization project to enable them to manage their entire workforce efficiently for each interaction, ensuring that a Weight Watchers expert is available 24/7 whenever and however a member needs support.

The flexibility provided by our building blocks approach along with the highly scalable global nature of our platform were keys to this new relationship.

You may have also seen in the news recently that Pinterest is using Authy to offer 2-factor authentication to their more than 150 million monthly active users around the world.

This effort is aimed at minimizing fraud and account takeovers along with protecting the growing social commerce taking place on the site.

The multichannel capabilities and the completeness of the Authy solution were key to their decision to move forward with us.

In addition, one of Europe's largest entertainment companies has added our messaging products to their existing call center footprint to improve not only agent satisfaction but to further raise their NPS core with customers as well.

They now allow customers to answer basic questions over SMS with a bot and then escalate into a full 2-way conversation with an agent when needed.

The value of our software and the potential Twilio gives them to expand from SMS into other channels and use cases were key to this new relationship.

So before I pass the call over to Lee, I did want to note one other recent item, the addition of Jeff Epstein to our Board of Directors.

Jeff brings a wealth of operational and financial experience, operating in the software industry at global scale.

Many of you may know Jeff from his time as the CFO of Oracle, but he has a long history of financial leadership across DoubleClick, King World Productions and Nielsen Media, and we're thrilled to add him to our board.

Overall, Q2 was another successful quarter across many fronts.

I'm always energized by the SIGNAL experience and what that means for our customer relationships, both existing and prospective.

The combined effort of Twilions around the globe have put us in the unique position to further our mission of fueling the future of communications.

We are at day 1 of a massive opportunity.

And now I'm going to turn it over to Lee to discuss our financial results.

Thank you, Jeff, and good afternoon, everyone.

Overall, the core tenets of our growth engine performed well again in the second quarter.

We saw further expansion within our existing customers and welcomed a healthy number of new customers to the fold.

In addition, we continue to invest for growth, particularly in product innovation and scaling our go-to-market efforts as we look to extend our leadership position.

I'm going to take a different approach in my prepared remarks today.

You've all presumably read the press release at this point, so I'll focus my commentary on additional color and data points.

Base revenue growth was quite strong once again, coming in at 55% year-over-year.

Our growth rate was dampened, however, by the decline in revenue from Uber that we previewed on the last earnings call.

Uber accounted for 9% of total revenue in the second quarter compared to 12% in Q1 2017 and 13% in Q2 of 2016.

In dollar terms, revenue from Uber was roughly flat in Q2 this year versus Q2 last year.

Removing Uber, base revenue grew by 65% year-over-year in Q2.

This had a similar impact on our dollar-based net expansion rate, which was 131% in the second quarter or 137% without Uber.

Uber's impact will continue to dampen our base revenue growth and expansion rate until we lap these larger figures in mid-2018.

Our top 10 customer accounts were 21% of total revenue in the quarter.

WhatsApp was the second largest contributor to total revenue, coming in at 5% in the quarter.

The next largest customer was around 3%, and the list scales down gradually after that.

We had 6 variable customer accounts in the second quarter compared to 7 in Q1 of this year and 9 in Q2 of last year.

As you look further down the P&L, note that our gross margins were down a bit sequentially as we outlined on our last call.

There are a number of inputs into gross margin, and one of those items that had positively impacted gross margin in the last 2 quarters, the mix of international traffic, provided less uplift than in the prior 2 quarters.

We expect this trend to continue in the third quarter.

Please recall that we're currently operating our business to optimize for reach and scale to drive revenue growth rather than maximizing for gross margin.

Gross margin may fluctuate in the near term as we pursue this deliberate strategy to further extend our market leadership.

You will also note in our recon table, a $12 million release related to our sales tax accruals.

We settled with some additional jurisdiction and made some progress a little faster than expected in some other jurisdictions.

This resulted in both a release of prior accruals and the accrual within the second quarter being nearly $2 million below our prior assumptions.

We still have significant work to do here, and we'll continue to accrue the remaining jurisdictions going forward until this process is completed.

We ended Q2 with 887 employees.

Before turning it over to your questions, I did want to spend a moment discussing Uber's impact to our guidance.

The situation has played out largely as we expected and outlined on our last call.

We did lose some of the lower-value bulk messaging use cases and readjusted pricing in other areas based on their rapidly growing volumes.

Most of these changes were implemented in the latter part of Q2 and are therefore partially reflected in this quarter's results.

Embedded in our third quarter guidance is a full quarter of impact of these changes.

This will translate into a larger sequential decline in revenue from Uber in Q3 than what occurred in the second quarter.

We currently forecast a more modest sequential decline in the fourth quarter.

This revenue forecast is consistent with our prior expectations guidance.

We expect that Uber will remain a significant customer for us going forward as we partner with them to support their needs globally for both existing and new initiatives.

Overall, we continue to be pleased with the performance of the business as evidenced by our revenue growth, expansion rate and customer additions.

We are very excited about the opportunity in front of us and the progress we are making in our mission to fuel the future of communications.

I'll now turn the floor over to your questions.

(Operator Instructions) Our first question comes from Bhavan Suri from William Blair.

I guess, just touching on maybe some (inaudible) instance for everybody here for a second.

As you move upmarket and you look at the traditional enterprise and you sell into that, how receptive is that group right now to a platform?

Or are they still sort of picking the individual offerings, SMS or just voice?

Or are they starting to approach this as a [plap] especially given the Engagement Cloud and Notify?

I just want to get some color on how those guys are adopting the solution.

Thanks, Bhavan.

This is Jeff.

I'll answer the question.

I think the thing to think about as we enter the enterprise, the developer-first approach is still leading the charge, right?

Developers inside the organization are finding Twilio or are starting by building out prototypes, and then we follow up with customers and help them make sure that the early traction that we're getting inside with the developer-first approach then turns into the final solution that gets implemented for the customer.

And what's interesting is that we've been able to up level conversations with customers, with the Engagement Cloud and really give customers a road map of how they can continually use Twilio to implement more and new advanced things and really open up a broader picture of what is their engagement strategy with their customers and what are the many ways in which Twilio can help do that.

And those are taking place at various levels of the organization.

The Engagement Cloud products are obviously new, so it's still playing out.

But we do feel that the Engagement Cloud products make it easier for customers to get the applications they build to production.

They can get there faster.

They don't have to reinvent the wheel because we're bringing best practices to the table for them, and that just helps them get their job done faster, which customers appreciate.

And, and that's it.

That color's helpful.

I guess, I'm going to -- just a quick follow-up here.

Obviously, Uber is something you commented on and gave a little more granular color on the call.

I guess, when you look at the use cases and you're in conversations with the company, obviously, on a regular basis, when you look at them and you said some of the bulk sort of messaging you had to be maybe read it out and sort of pass it on, what are the critical use cases for Uber where you feel that, that's something that's so important between quality of service and everything else that those remain very, very steady for you to give you sort of that confidence of a lower sequential growth rate in Q4 and then more importantly, sort of an uptick, potentially growth next year?

But just a little more color on that.

Yes, absolutely.

I mean, there's a number of use cases at play in a company like Uber, and the way they look at those use cases and what they value differ based on how the use case works and as well as where in the world it is.

For example, if you're doing a blast notification, bulk notification to many thousands of drivers and then get out on the road or something like that, there's a different quality of service required than the SMS telling you that your ride is arriving, right?

You expect that to be timely.

You want it to get there, and then it impacts the customer experience if it doesn't get there on time.

Whereas when you're blasting out many thousands of messages, if there's some that don't get there, it's just a different quality of service that you'd expect from that sort of bulk type use case.

And overall, we think there's just a lot of use cases and a lot of opportunities inside of Uber and a lot of opportunities both on existing and new use cases.

And we are working very effectively with the team there.

Your next question comes from Richard Davis from Canaccord.

You kind of touched on this a little bit.

But kind of to what extent should we expect to see Twilio kind of used as a platform to make calls between extensions and even kind of via some of these connected endpoint devices like Home and Echo and things like that?

Yes, absolutely, Richard.

So we announced at SIGNAL just a couple of months ago our new channels product, which actually is our method of interconnecting with a wider variety of new channels.

We actually onboard these new channels very efficiently as new ones are coming online and as they actually change because there's constant changes happening.

And so we've become very good at onboarding and maintaining these channels.

And this opens up new opportunities for customers to use a variety of our products in a cross-channel way.

So Notify is a great example of this.

Notify uses multiple channels for notifications so that a developer can harness the value of where their end users want to be notified and can actually experiment with new channels as they come online very quickly because it's built into the platform.

And Alexa's another good, interesting use case, both for notifications but also something that we are looking to power with our Understand products so that a customer who builds a natural language understanding model using our Understand product will be able to power communications over, say, an IVR, if you call a company and say, "Tell us what you want to do," but also be able to have the same styles of interactions across Alexa and other -- some more platforms there.

But to essentially build those models and build that application once and power it across many different platforms.

So we're very excited about the opportunity that these channels bring.

Your next question comes from Mark Murphy from JPMorgan.

Jeff, it's been an interesting couple of months with 1 of your competitors suffering from, I believe, a 22-hour outage and another 1 of your competitors seemingly closing its doors to new customers and new projects.

And meanwhile, your new customer additions continue to strengthen.

So I'm curious, from your perspective, has the competitive environment become a bit less intense?

Or has it changed in any way in recent months?

And just also, how frequently are you winning new customers who have, in fact, suffered from poor quality with a competitor and therefore, are turning to Twilio?

Yes.

Thanks, Mark.

As a company, we're very focused on our customers, not as much on the competition.

And as such, that's where we get our focus on agility with resiliency, the idea that we deliver innovation that benefits our customers, but we do so while -- the rapid pace of innovation while maintaining a high level of quality and resiliency in our product.

And that manifests as five nines API availability and a five nines success rate and things like that.

And I think that, that ability to deliver innovation with resiliency is a core strength of Twilio's.

And to answer the second part of your question, yes, we do see customers who come to us from other platforms and due to the struggle that a vendor might have because of either lack of innovation or the availability of their solutions.

And so customers do come to us and value the things that we've invested in that make our platform extremely reliable but also an engine of innovation for them.

Okay.

As a follow-up, Jeff, what stage is your typical customer at in terms of adopting SMS text versus push or in-app notifications?

And if you looked among those who have introduced push or in-app notifications, is it more commonly working out favorably or unfavorably for Twilio?

Yes.

It varies a bit.

Notification -- or Notify, our product that supports push notifications, is a relatively new product for us, so it just recently went into a beta earlier this year.

And so it is obviously a newer product for us, and so the adoption is fresher than SMS, which has been around for many years.

But as far as how it works out, we -- our goal is to focus on our customers' needs.

And we feel that if we focus on helping our customers notify their customers in the appropriate way, our job is to offer choices to our customers and give them maximum flexibility.

And the way this results is, look, if the end customer wants a push and our customers should offer push, that's the solution that we should offer.

And I'll point out that this has the capability of actually bringing net new push and SMS workloads onto Twilio as customers are looking for a solution that can encompass both their push desires as well as SMS desires.

And SMS is a very large mature market where that business may not even be with Twilio today, so it's actually an acquisition strategy for us as well.

Okay.

One final one if I may.

I'm curious, Jeff, just how rapidly do you think you can -- how rapidly do you think your mix can shift toward enterprise businesses?

You mentioned quite a few of them this quarter that are non-tech-first types of companies.

And is it possible to see a future where the enterprise revenue stream would surpass that of the tech-first companies?

Well, enterprise, relatively speaking, is a smaller part of our mix today but growing faster than the overall, and so it's a segment we're very excited about.

We think the enterprise opportunity is very large.

And we think about TAM, a lot of that is with traditional enterprises and products that they bought traditionally.

And so we see this as a very large opportunity for both segments, but the enterprise is an exciting one that is growing extremely fast for us.

Your next question comes from Heather Bellini from Goldman Sachs.

Jeff, just a couple of questions.

I was wondering if you could share with us what products are driving the most customer engagement right now, if you could share with us the top 1 or 2. I know you've gone into some detail across the board but if you can zero in on that.

And how do we think about the pace of expansion of multiple products into your installed base?

And then I just had a question about ARPU as a follow-up.

Absolutely.

As far as which products drive the most engagement, SMS and voice are our most mature products, the most mature channels.

So those are the ones that dominate because of the maturity inside of Twilio.

We've offered those products for many years as opposed to some of our newer products.

As far as the ramp-up goes, when we introduce a product, you do have the time it takes for developers who get up and running on that product to build what they see as their requirements and understand what they can do, match it to their customers' needs, build it and then deploy it.

And so the ramp to revenue is one that takes the developer to do their job, building the solution as well as the end customer adopting the thing they built.

And so you do have that ramp built in, and that's why newer products sometimes do take a little longer to turn into revenue.

But they're still very good investments that we see playing out over a long time.

But it does take some time for a developer product to get in the market and to the extent where it will be highly visible on our revenue line.

Yes.

And I guess, that's what I was trying to [get at].

Some of these newer products that you're mentioning, is it 12 months or a 24-month process before we start to see that bubble up and become more apparent in the revenue stream?

Is there some time frame that we should think about?

Well, I mean, I think you're not far off there.

The thing to think about some of these products are more standalone, so you have to kind of build starting -- the new use case starting from scratch, let's say.

Other of our products, for example, our Marketplace, are attachments to your existing usage of Twilio.

So you already have a use case, and you can add on a product to get more value on the thing you've already built.

Marketplace, Insights is another example of that.

And then other products actually, the adoption of them pulls in revenue from our other products.

So for example, TaskRouter, when you're building a call center, TaskRouter is the core of that initiative, but it will, generally speaking, pull in voice revenue along with it.

So even though we've got new customer traction going on with a new product, the spend on voice may equal or even exceed that of the spend on the new product, and so it is hard to separate these 2 things.

Although what I can tell you for sure is that TaskRouter, the fact that it was there, helps us acquire that customer and helps them get onboard with that use case, even though the revenue may be split between the new product and the old product.

Okay, great.

And then just one last follow-up.

Is there anything you could share with us about how we might want to think about the pace of ARPU growth as we look out?

Heather, this is Lee.

We have had strong ARPU growth over -- actually, over the last few years.

In Q2 of '17, ARPU with -- for everything -- for base revenue was $8,300.

That's up 9% year-over-year.

If you strip out Uber, that's actually up 17% year-over-year.

We've seen that pace grow, but it's not something we're going to continue to guide to specifically.

Your next question comes from Pat Walravens from JMP Securities.

I guess, my first question would be what has George been focused on in his first 5 months there?

Absolutely, Pat.

So George is leading the go-to-market efforts.

So sales and marketing has been his primary focus, and we feel very confident in the direction and execution that he's bringing.

And as I mentioned earlier, he is bringing this great amount of alignment between our sales and marketing organizations, both for our existing products as well as new products we bring into market and really helping us to align those efforts, specifically in things like hiring process, structure, focus, things like that.

Great.

And then just big picture, Jeff, what points -- what sort of key points would you make to investors who worry that this is a commodity?

Well, what it I would say is that there's a lot of ways to approach this market, right?

And the way we approach this market is by delivering software value to our customers, right?

So we hear the engagement challenges that our customers are trying to solve with their customer base, and we're investing in the software that makes it possible for them to build those solutions on top of Twilio.

And that is a software endeavor.

And so you really get all the challenges that you get building a great software company of how to deliver agility, constant stream of innovation but also agility being able to -- also resiliency and doing so with five nines availability and five nines success rate and all that kind of stuff.

And that's how we've approached the market of -- as a software endeavor.

But we balance a big market.

There's room for many different approaches about how people bring it to market, but how we've approached it is building the software that our customers need in order to engage with their customers effectively at scale and globally and use a platform that they know they can trust.

Our next question comes from Ittai Kidron from Oppenheimer.

Couple questions.

First, for you, Lee.

Regarding Uber, you gave very specific commentary on Uber's comment -- contribution for next quarter and the fourth at least directionally, percent of revenue.

How much of that is you being conservative versus, at this point, you have a true and complete picture and understanding of the nature of your interaction with Uber?

Yes, sure, Ittai.

So [assumably] as we look at the case we have a better handle this quarter, there's all sorts of puts and takes with Uber.

This is a complex organization.

It's unique, and it's growing fast.

It's taking a lot of risks.

We're trying to set the guidance accordingly, and the numbers in the fourth quarter is our best estimate of where we see things netting out.

Okay.

And then, Jeff, I mean, the engagement vision is very compelling.

I guess, I'm trying to understand, as you try to move forward with this, how much of this do you think is a push from you versus a pull from the customers for this?

That's number one.

And number two, what do you think would be the time line by which revenues from all the different elements, whether it be Authy, Notify, TaskRouter and the others, and I'm sure there'll be other things coming along, before they account for, say, 10% of revenue?

How do we think about the progression and contribution of this vision to your financials?

Yes, absolutely.

So our developer-first approach has continued working, where the developers inside the organization are pulling us into their organization by being able to get started, building a prototype and getting essentially momentum going inside of a company for a solution that is built on top of Twilio.

And we do see that continuing to grow and that approach working very well, so we're continuing to invest in that developer-first go-to-market approach.

But what's interesting is that often spurs the discussion inside of a company of, hey, how are we using Twilio, what else can we be using Twilio for, and it actually creates a great basis for a conversation at a variety of levels inside the organization about a broader picture of engagement strategy, right?

And that's the thing that executives or C level cares most about, and so the Engagement Cloud and the set of products in there are a nice handle to be able to have that conversation with a customer, especially an enterprise-type customer, about where to go next and how the adoption of Twilio might be strategic to their road map going forward for many different use cases.

And the other thing I'll remind you on the second part of your question about revenue, which is Engagement Cloud is just announced.

It's relatively small now but it's growing quickly.

But I'll remind you, as I mentioned with TaskRouter, adoption of a lot of these products actually pulled through other revenue as well.

So if you adopt TaskRouter, it might pull through voice.

If you adopt Notify, it might pull through SMS.

And so I don't know that it's exactly a -- this product versus that kind of product because they can actually drive growth between each other.

And the traditional products are growing very quickly as well.

So I wouldn't necessarily speculate at what dates the lines will cross other than we see opportunities across both.

Our next question comes from Mike Latimore from Northland Capital Markets.

This is actually Nick Altmann on for Mike.

You guys mentioned a couple quarters ago the number of customers that are doing over 10,000 in ARR a year.

Can you just give us an update there?

Nick, this is Lee.

That number is growing very nicely, but that's not something we're going to give on a continual basis.

That might be something we mention going forward, perhaps in our Analyst Day.

Got it.

Okay.

And then just a second if I may.

Can you help us understand where the bulk of the growth is coming from?

Is it increased usage?

Is it customers adding new APIs, new use cases?

Really, any color there would be much appreciated.

Sure, I mean, the growth being driven still from our most mature products, Programmable Voice, Programmable Messaging.

And as you saw from the quarter, we had significant customer additions.

It takes a while for those customers to come up, but we benefit from the older [corp] or the new corp coming onboard and driving revenue.

So it's hard for us really to break out in specifics, but it's both strong from -- strong growth from new customers and expansion within existing customers.

Our next question comes from Brian White from Drexel.

Yes.

Jeff, if you could talk a little bit about what was the biggest surprise in the quarter, it looks like you beat your midpoint of your guidance on revenue by 11%.

So it was a huge beat.

So what surprise?

And also, you've got so many new product and a lot of new, exciting things in the pipeline here, if you can just give us some color on the average customer at Twilio using 2 products or 3 products.

Just give us some feel what the average customer uses in terms of number of Twilio products.

Yes, sure, Brian.

I mean, I think it's hard to pinpoint one surprise like that, and I think we've seen strength across a number of different areas of our execution.

Really excited about the products we announced, great new customer adds this quarter.

And so I think it's just a testament to the team here at Twilio about executing on a wide variety of fronts and seeing strength there.

And I think there was a second part of your question.

Yes.

Just if we looked at your average customer today and if you can compare it maybe to a year ago, how many Twilio products are they using?

Is the average customer using just a couple products?

Or is it more than that?

Just to give us a feel here because you've had a lot that's come out over the past 2 to 3 years and obviously, a lot just in May alone.

So if you gave us that type of color, I think it'd be very helpful.

This is Lee jumping in.

That's not a number that we've disclosed to date.

The number of products being used by our customers is growing and reflected in the strong ARPU growth, up to $8,300 for the quarter.

But that's something that we have not talked about publicly.

Okay.

Is there any way to just give us maybe color on what happened internationally in the quarter?

I'm sorry, I didn't catch the question.

Color on...

How was growth overseas, international markets?

Yes, sure.

I'll -- this is Lee.

I'll start.

So revenue from customers based outside the U.S. was 22% of total revenues, so that's a nice uptick from the prior quarter.

And it grew nicely and growing well -- doing well according to plan.

(Operator Instructions) Our next question comes from Jonathan Kees from Summit Redstone.

Just want to ask about 2 particular topics.

One is pricing.

You talked about you had to realign pricing with Uber and one with your bulk messaging.

Just curious, have you had to realign your pricing with your other customers?

I realize they're less -- they're not as big as Uber.

And then how would you characterize pricing in general?

I also had a second question on top of that.

This is Lee.

So we continually look at pricing, so that's one of the beauties of our business model.

As customers scale and grow, they know they can get better pricing.

So as customers scale and move up, we do offer them better pricing, and that's a continual process.

Regarding the overall view of pricing, there's -- we haven't seen any major changes to the overall pricing environment.

Okay, great.

And then secondly, I want to ask about, is the guidance you gave, you talked about -- it was based on the feedback from Uber, the puts and takes there and your estimate based on the feedback you've gotten from Uber.

It sounds like visibility has improved within there.

Would you -- how would you characterize visibility across-the-board?

This is Lee.

I'll start on that.

So we have a great relationship with Uber.

We work very closely with them on their existing products.

We're working closely with them in terms of new initiatives.

And so in terms of working with them, we do in that close relationship.

As I mentioned before, it's a unique account.

It's complex.

They move very fast.

And so we feel we have a better handle on what's going on, but again, it's a unique company.

And as I mentioned, the numbers reflect in the forecast our best estimates how things are going, I think they've tracked out -- as we look at it, certainly for Q2, things are moving as expected.

Then, that'll be across-the-board with other customers, too, right, in terms of the feedback, the -- a visibility with them?

Sorry, I missed that.

And you would characterize visibility with your other customers as the same or has improved?

Yes.

I think in terms of the overall business of it, one of the features of the usage-based model is we have to have high visibility and predictability with thousands -- hundreds of transactions happening per second across thousands of customers.

And we continue to see that high visibility.

It's a platform.

It's very valued.

It's embedded, and it's integral to our customers' usage.

So that has been a big driver of our historical visibility and predictability.

And our next question comes from Will Power from Baird.

I guess, a couple of questions.

First, just on the third quarter base revenue guidance, it implies a slight sequential downtick.

I recognize part of that's the Uber pressure.

But I wonder if there's any further color you could provide there, or if there's just a measure of conservatism built in there.

And then secondly, Lee, you referenced some comments on gross margins moving forward.

I wonder if you could just reiterate those and how we think about that longer term as well.

Yes, sure.

So as we look into the quarter, sure, we like to be conservative in our guidance.

The number is just guidance.

And as a company, we aspire to beat that.

But you're correct.

We are absorbing the impact of Uber, which is a reason for that.

And then your second -- could you repeat your second question about gross margin?

Yes.

I think in the prepared remarks, you made some comments on gross margins, I think, for Q3, and I wondered if you could reiterate those and just how we think about gross margins over the next couple of years.

Yes, sure.

So gross margins, as I did talk about in my prepared remarks, we manage that line item like a hawk.

It's very important to us, but we're not maximizing for gross margin in the short term.

We're maximizing for customer reach and scale and building the business long term.

In any short-term time period, there's puts and takes in terms of the gross margin.

So we may make decisions regarding quality that result actually in a higher cost for us, so we feel that's the right thing to do, or if we entered into geographies where we don't -- we're not at scale yet, that could have a negative impact on gross margin.

Many other factors there.

So over the last couple quarters, the gross margin was a bit elevated due to some geographic mix, but we see that benefit has lessened in this quarter, and we see that benefit lessening [off] in terms of the next quarter.

In the long term, we feel really good about our long-term model of gross margin 60% plus, and that's going to be driven by continued operating efficiencies and primarily driven by the product mix, increasing the engagement [capital] products, IP-only products.

So it's become a larger and larger percentage of our revenue portfolio.

And there are no further questions at this time.

Thank you for participating in today's conference.

You may now disconnect.