Twilio Inc (TWLO) 2016 Q2 法說會逐字稿

Good afternoon. My name is Stephanie, and I will be your conference operator today. At this time, I would like to welcome everyone to the Twilio Q2 2016 earnings conference call.

(Operator Instructions)

Thank you. I would now like to turn the call over to Mr. Greg Kleiner, Head of Investor Relations for Twilio.

You may begin.

Thank you.

Good afternoon, everyone, and welcome to Twilio's second-quarter 2016 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 2016 second-quarter performance, in addition to our financial outlook for 2016 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, our market opportunity and market trends, customer adoption of our products, our momentum, 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.

Discussions of the risks and uncertainties related to our business is contained in our final prospectus filed with the Securities and Exchange Commission on June 23, 2016. 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 such statements are 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 occurrence of unanticipated events except as required by law.

Our commentary today will contain certain non-GAAP financial measures, including discussions of gross profit, operating expenses, loss from operations, net loss per share, and weighted average shares outstanding. Non-GAAP financial measures may exclude certain amounts which are included in our GAAP financial measures. Accordingly, the following items are available in our earnings release, which we issued a short time ago. The most directly comparable GAAP financial measures, the reconciliations between these GAAP financial measures and non-GAAP financial measures, information regarding the reasons why we are presenting non-GAAP financial measures, as well as the reasons why we present guidance for non-GAAP measures of loss from operations and net loss per share, but not the comparable GAAP measures. The earnings release is available in the Investor Relations page of our website and as part of Form 8-K furnished to the SEC.

Finally, at times in our prepared remarks or in response to your questions, we may offer incremental metrics to provide greater insight into the dynamics of our business or our quarterly results. Please be advised that this additional detail may be one-time 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 www.investors.twilio.com to access our earnings release, periodic SEC reports, a webcast replay of today's call, or to learn more about Twilio.

I will now turn the call over to Jeff.

Thank you, Greg.

Good afternoon to everyone on the line today. Thank you for joining us today for our first earnings call as a public company. We are excited to share both our results from the second quarter, as well as our vision for how software will drive a massive shift in how we communicate and how Twilio plans to power that future.

Twilio is a Company founded by developers for developers, a concept that most said couldn't be done when we started the Company back in 2008. Many of you listening today may be new to the Twilio story. So let me spend a few moments describing our business and what makes Twilio unique before walking through some of our Q2 highlights.

As a backdrop, we have entered into a new era of software. In the first era of software, 25-30 years ago, companies used software for back-office applications. And each project involved multi-million dollar decisions and multi-year time frames to implement, and thus required that the CIO drive the decisions.

Then 10-15 years ago came the next era of software with software as a service. At that point, line-of-business managers could buy the software systems required to run their parts of the business without needing IT to deploy it. And this was a major innovation.

But the fastest-growing enterprise software company in history is not on-prem, and it is not SaaS; it is not even an application. It is Amazon Web Services.

This signals a new era of software, the era of cloud platforms. Instead of selling applications, you are selling building blocks. Instead of selling seats or licenses, you have a usage-based pricing model. And instead of selling directly to CIOs or line-of-business owners, you are selling through developers.

As every company starts to invest in software and software developers as a key point of differentiation in their businesses, their developers turn to these cloud platforms to build software faster and more cost effectively than ever before. Whether it's Amazon Web Services for computing and storage, whether it's Google Maps for mapping, PayPal or Stripe for payments, these are the new building blocks of every app, consumer and enterprise, that we use every day.

But what was missing from this picture before we started Twilio was communications. We created a new category; and today, Twilio is the leading cloud communications platform. Our mission is to fuel the future of communications. We don't provide a single software application like most software companies. Rather, we are a horizontal platform with building blocks that help developers build communications into other software applications.

Just as Amazon Web Services provides access to compute and storage through their platform, we allow developers to embed real-time communications into their software applications. Our revenue model is primarily usage-based. So whether this is Uber allowing its customers to communicate with drivers, Nordstrom enabling salespeople to text with customers via their mobile CRM app, or Zendesk Voice providing a call center solution to its customers, we participate in all of this. Every message sent, every minute of voice traffic, every video session, every authentication performed in these apps built by our active customers drives revenue to Twilio.

To understand Twilio, you can think of us as three things. And these three things work together to provide value to our customers and differentiation in the market. The first of these three things is our programmable communications cloud. It is the software platform we provide to our customers. These are the building blocks exposed as APIs, or application programming interfaces, that developers can incorporate into their application.

Our products today span programmable voice, messaging, video, and authentication. They allow developers to build applications that support voice calling, sending text-based communications, embedding video, and incorporating two-factor authentication into their apps.

But it is not just the communication, but also what happens in the communication that these building blocks enable. For example, higher-level functionality like conferencing, call recording, text-to-speech, picture messaging, intelligent message encoding, English localization, even integrations with Facebook messenger, among many other things that add value to our customers.

Given the power and flexibility of our APIs, developers can build a wide variety of use cases. Some of the more common use cases for Twilio include anonymous communications, alerts and notifications, contextual support, call tracking, mobile marketing, user security, and more. But the options are really only bounded by the creativity of developers.

The second of our three things is our super network. This is our software running in 22 data centers in 7 regions around the world that is interconnected with the carriers of the world, allowing us to connect with nearly any end user on the planet. It uses data and software to optimize the performance of our network for both quality and cost structure for the voice calls and text messages that traverse our network.

We are constantly adding more interconnections to the super network. And with more volume of traffic, interconnections and, importantly, data, this forms a flywheel of scale and intelligence that reinforces our leading market position. Is our business grows, the super network becomes more difficult to replicate over time.

Finally, our business model for innovators. Put simply, we acquire developers like consumers but enable them to spend like enterprises. Our goal here is ubiquity. We want Twilio to be in the tool belt of every software developer in the world because we believe communications is so core to business that eventually these developers will have a use case at work that requires communications. And when they do, we want them to turn to Twilio to solve it.

As a Company founded by developers for developers, we have been actively engaging the community for the last eight years. Today we have more than 1 million developer accounts registered on our platform. And we are adding more than 30,000 more per month.

We start with a low friction on-boarding process, where a developer needs nothing more than an email address to get started. From there, our usage-based revenue model is primarily pay as you go. So as our customers with from a prototype to a beta to a general release and more end users interact with the software they've built on Twilio, we generate more revenue.

When they succeed, we succeed. As they grow, we deepen our relationship through sales and customer support to help them along this path and realize additional use cases as time goes on. The result of this model is twofold, efficient customer acquisition and best-in-class revenue expansion. We will discuss each of these in more detail shortly.

The combination of these three things is what drives our leadership position and provides our competitive differentiation in the marketplace. No other company has these three things operating at our scale.

With that as background, let me discuss how we did in the second quarter executing against this vision.

The business produced another quarter of excellent financial results. Base revenue grew by 84% in the quarter to $56.4 million versus the same period in the prior year. Total revenue in Q2 grew by 70% year over year to $64.5 million. In addition, we showed significant leverage on the operating line. As Lee will outline in a moment, we saw continued strong expansion rate in our customer base.

We also signed a number of important new deals and expansions in the quarter, including with enterprise customers. We had some great wins at FICO, BMW, Home Depot, Nike, and the New York Times, as examples. In addition, I'd like to spend a moment talking about our wins at ING Bank, Salesforce.com, and Amazon Web Services.

First, ING Bank, where their team decided to replace 17 different legacy call center systems deployed in over 40 countries with one new solution their developers are building on Twilio. This win in particular shows how the effectiveness of our business model for innovators translates into the enterprise.

The development team within ING pulled us into the opportunity after discovering how our communications building blocks could be leveraged to build a new type of omnichannel contact center to service the many countries and business units of ING. By enabling ING's development team to build prototypes of their solution on Twilio, they were able to evangelize our platform within the broader ING organization. As a result, with the support of their development team and the Netherland group CIO as our executive sponsor, we were able to win this global deployment in one of the largest banks in Europe against many of the traditional call center vendors.

Also in Q2, we announced an expanded relationship with Amazon where we will be helping to deliver SMS messages as part of their simple notification service. This builds on our existing relationship with AWS as an investor, customer and supplier to Twilio. Salesforce is a great example of our work with a category of customers we call solution partners -- independent software vendors, or ISVs, that embed Twilio into their software and bring a complete solution to their customers. With the latest release of Sales Cloud and Lightning Voice, Salesforce customers can now call their customers with one click directly from their browser and without the need for any external phone systems.

This is a tremendous productivity enhancer for salespeople, who no longer have to log calls manually or leave Salesforce to effectively interact with their customers. Whether it is customers building directly on Twilio, like ING, or buying solutions brought to market by our solution partners, like Salesforce or Zendesk, we believe we can address the full communications market. And we are excited by the progress on both of these fronts in Q2.

We continue to make significant progress on the product front as well. We were pleased to join Facebook on stage at their F8 user conference in April as we announced our integration with Facebook Messenger. Businesses today need to be active wherever their customers are. And we can now power interactions with Facebook Messenger from within our platform.

In addition, we recently held our annual user event, SIGNAL, in San Francisco, where we celebrated our customers, launched products, educated developers, and helped further our vision for the future of communications. We announced three important new innovations at SIGNAL this year.

The first was the launch of the Twilio marketplace with a new product we call Add-ons. With this marketplace partners, such as IBM Watson, Wolfram Alpha, White Pages Pro, and more can pre-integrate their services into Twilio and allow developers and businesses to light up that functionality in their applications with one click.

For example, a developer can add sentiment analysis to their messaging application in minutes using Watson or automatically deflect spam calls from their call centers using Nomorobo. When our customers adopt these pre-integrated Add-ons, we share in the revenue with our partners.

The second announcement was Twilio programmable wireless, essentially the first fully programmable wireless carrier. Developers can now program every part of how a carrier operates with voice, messaging, and data in order to build applications that operate natively with the phone, all powered by just inserting a Twilio-powered SIM into the device. We make it easy for developers to start building IoT solutions with low upfront costs and little friction, and then scale up their solution when they start shipping devices. This is the Twilio model at work again in a new market, and one that we are excited about, IoT.

We also anticipate developers building BYOD solutions for the enterprise, especially in heavily-regulated industries like banking or healthcare, where a developer can automatically log every phone call and text message programmatically. In short, we don't believe the world needs another wireless carrier. But it just may be that the world needs 1,000 micro carriers for a wide variety of use cases. This is why we built Twilio programmable wireless.

The final product announcement from SIGNAL was Notify, an API for orchestrating notifications across SMS, push, and messaging applications, such as Facebook Messenger, as I mentioned a moment ago. The communications landscape is growing increasingly complex for businesses to manage. But with complexity comes opportunity. And Notify is designed to help businesses navigate this rapidly changing landscape by effectively managing customer communications at scale across a wide and growing variety of channels.

A few weeks ago, we launched Twilio Sync, a new API to help keep Web and mobile apps in sync across devices and between users. State synchronization is a core aspect of a growing set of modern Internet applications, like enabling coediting of an online document or updating the location of a driver on a map in real time. And we are excited to get this functionality in the hands of developers with our new Sync API.

Our customer successes and continued innovations have led IVC to name Twilio as the sole leader in the marketplace report for worldwide cloud communications platforms.

On the management front, we were pleased to welcome Erin Reilly to our executive team as the Executive Director of Twilio.org. We have taken the 1/1/1 pledge and last year committed 1% of our pre-IPO equity to funding Twilio.org. Erin brings a wealth of experience in social responsibility as a veteran of leading organizations like Google, Yahoo, and Nike. She will help drive our continued focus on empowering nonprofit organizations to use communications to solve societal problems, which is a core part of the Twilio.org mission.

We also expanded our Board by adding Erika Rottenberg. Erika brings with her more than 15 years of leadership experience with the technology industry. Formerly the General Counsel at LinkedIn, SumTotal, and Creative Labs, she has been instrumental in guiding leading technology companies as they scale their organizations globally. We look forward to her contributions as Twilio continues to grow at a rapid pace.

Before I turn the call over to Lee, I wanted to thank our customers, our team, and our shareholders, old and new, for their contributions that have gotten us to where we are. I would especially like to call out our employees. We call them Twiliots. As this has been an unusually busy and special quarter for the Company, one that saw SIGNAL, our best user conference yet; a tremendous amount of growth; and, of course, our IPO. Our employees handled these events with poise, grace, and above all else, a focus on customer success.

I'm proud that Q2 was another quarter of strong growth and accomplishments, but this is Day 1. As we look forward we believe the opportunity has never been greater. We are just beginning the journey of migrating our communications from a legacy and hardware to a future in software.

And it's the software developers of the world who will unlock the potential of software to make our communications more meaningful and relevant. We are excited to partner with them on that future, and we can't wait to see what they build.

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

Think you, Jeff.

Welcome to the new investors on the call.

I will begin with a brief overview of our financial model and key metrics, then cover our financial highlights from the past quarter. As Jeff mentioned earlier, we created the cloud communications category, providing a platform that allows companies to easily embed and scale communications into their software applications. We have built a business that combines rapid revenue growth at scale with an efficient go-to-market model.

Our platform revenue model is primarily usage based. So every message sent, every minute of phone traffic, or every authentication that is processed by our cloud communications platform, generates revenue for Twilio. This activity occurs hundreds of times per second across our broad customer base.

We break our revenue into two categories, base revenue and variable revenue. Base revenue constitutes the majority of our revenue. Base revenue has historically been predictable and is the basis on which we have always planned and operated our business and, accordingly, will be the focus of our discussion today. We also plan to provide specific guidance on base revenue to help investors evaluate and track our business.

The second category of our revenue is variable revenue. Variable revenue comes from a small subset of our customers called variable customers, where we believe the revenue is more likely to fluctuate significantly between periods. We define a variable customer as a customer whose revenue has exceeded 1% of our total revenue in a quarter, but has not entered into a 12-month minimum revenue commitment agreement.

Payable revenue provides benefit to our business through added scale and gross profit dollars, despite its limited predictability. In the second quarter, we had nine variable customer accounts, accounting for approximately 13% of our total revenue, with the majority of this variable revenue coming from one company, What'sApp.

Another key metric we track is our customer account. An active customer count is an individual account that has generated at least $5 of revenue in the last month of any period. We also have a large number of customer accounts either utilizing free trials or operating below the $5 monthly threshold that we believe will be our next big customers as they start building and deploying their applications.

A third key metric we track is our dollar base net expansion rate, which compares revenue generated by our base customer accounts in one quarter to the same quarter in the following year, including churn. Base customer accounts are active customer accounts other than our variable customer counts that I discussed a moment ago. We are very proud of this metric as it tangibly expresses the continued value we are providing to customers and the benefit of our business model for innovators.

This metric has been consistently above 140% in recent periods, best in class amongst our peers in the software space. This growth has been driven by our ability to maintain and grow our relationships with our existing customers.

Let me transition to some highlights from the second quarter.

We are quite pleased with our results for the second quarter, as it demonstrates the power of our platform model and the continued success we are driving for our customers. Base revenue for the second quarter of 2016 was $56.4 million, up 84% year over year from the second quarter of 2015. We saw continued strength across our product line and from customers of all sizes.

Total revenue for the second quarter of 2016 was $64.5 million, up 70% year over year from Q2 of 2015. Variable revenue decreased sequentially from Q1 and, as noted earlier, may fluctuate from period to period. What'sApp is the majority of this category; and their usage, like other variable customers, is unpredictable. That is why I created this revenue category and the reason we plan our business around and focus on base revenue.

In the quarter, our top 10 customers were 31% of total revenue, consistent with recent quarters. As of June 30, 2016, active customer accounts were 30,780, up from 21,226 as of June 30, 2015. These figures include nine variable customer accounts in the second quarter of 2016 and 2015.

Our dollar-based net expansion rate was very strong once again in the second quarter, coming in at 164%. Our active customer count growth and dollar base net expansion rate both demonstrate the power of Twilio's low-friction platform business model.

Before moving on to profit and loss items, I would like to point out that I will be discussing non-GAAP results going forward. Our GAAP financial results, along with the full reconciliation between GAAP and non-GAAP results, can be found in our earnings release.

Non-GAAP gross margins in the second quarter of 2016 were 56%, flat compared the second quarter of 2015. We are a platform provider in the early days of a huge market opportunity. As a result, we are currently operating our business to optimize for reach and scale 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.

Non-GAAP operating expenses in the second quarter of 2016 in total were $42.1 million or 65% of total revenue. This compares to $28.5 million in total for the second quarter of 2015 or 75% of total revenue.

One line item I would like to focus on for a moment is sales and marketing expense. Our business model for innovators enables efficient revenue acquisition. For the second quarter, our non-GAAP sales and marketing expense was 26% of total revenue, much lower than many other high-growth software companies. Note that this figure includes expense from our annual user conference, SIGNAL, that occurred in the second quarter.

Non-GAAP operating loss was $5.7 million in the second quarter of 2016, compared to $7.4 million in the second quarter of 2015. Our operating margin improved by 1,000 basis points year over year from negative 19% to negative 9%. Going forward, we plan to make continued progress toward breakeven and beyond, but it may not be in a linear fashion.

Our non-GAAP loss per share in the second quarter was $0.08 per share, based upon a non-GAAP weighted average share count of 72.8 million shares, which assumes the conversion of preferred stock as of the beginning of that period. This compares to a non-GAAP loss per share of $0.11 per share in the second quarter of 2015, based upon a non-GAAP weighted share count of 60.3 million shares, which also assumes the conversion of preferred stock at the beginning of that quarter.

We ended the quarter with $261 million in cash and cash equivalents, compared to $103 million at the end of the previous quarter. Note that the net proceeds from our IPO at the end of June were approximately $156 million. As we look into the second half of the year, we expect our free cash flow to include the net impact of roughly $7 million related to the build out of our new headquarters in San Francisco.

Let me turn to guidance. We are providing our initial guidance as follows.

For the third quarter ending September 30, 2016, base revenue in the range of $58.5 million to $59.5 million; total revenue in the range of $63 million to $65 million; non-GAAP loss from operations of $7.5 million to $8.5 million; non-GAAP net loss per share of $0.09 to $0.10, based on 84 million weighted shares outstanding.

For the full year ending December 31, 2016, base revenue in the range of $227.5 million to $229.5 million; total revenue in the range of $253 million to $257 million; non-GAAP loss from operations of $21.5 million to $23.5 million; non-GAAP net loss per share of $0.28 to $0.30, based on 78 million non-GAAP weighted shares outstanding.

While we are not providing detailed guidance beyond 2016 at this point, I did want to add a few higher-level thoughts.

The second half of 2016 will be focused on continued investments. We are looking to expand our lead in this large and growing category. We are investing across multiple fronts a plan to take advantage of the increased developer and customer awareness provided by our IPO. As we look to 2017, we anticipate reaching non-GAAP operating breakeven in the fourth quarter of 2017.

In summary, we are very happy with the continuing momentum exhibited by our business. The value provided by our platform is clearly resonating in the market. And we believe we are still in the early innings of what is a massive opportunity to reshape one of IT's largest spending categories.

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

(Operator Instructions)

Your first question comes from the line of Heather Bellini with Goldman Sachs. Your line is open.

Great. Thank you very much and congratulations on the offering, gentlemen. I was just wondering if you could share with us, obviously everyone knows that the customer concentration that you have. But I was wondering if you could go a little bit deeper and think of customer 10 and below and maybe talk to us about the spending trends and the penetration trends you are seeing amongst that base of customers as they continue to expand their use of your technology?

Hi, Heather, this is Lee.

In terms of customer trends, it's been very consistent across a broad base. We have our 30,000-plus active customers. Our top-10 customers are 31% of revenue and then we trail all the way off through the long tail to evenly spend all the way down to our active customers, which are over $5, so we have seen consistent trends in growth across all revenue categories.

Are those customers growing faster than the top 10? I guess that is what I'm trying to get at.

I would say across the base I don't think any bucket or category of revenue growth rate is faster than any larger customer at a macro level.

Okay. Great. Thank you.

Your next question comes from the line of Mark Murphy with JPMorgan. Your line is open.

Thank you very much, and congratulations on the strong results.

Jeff, I had a couple of questions for you. First of all, how much upside potential do you see over time in targeting enterprises more along the line of what you have announced with ING and also with others? I'm wondering what is your confidence level in terms of being able to replicate what you've done for technology firms in trying to do that on the same scale across many other industries

Absolutely, thank you, Mark.

We're optimistic about the ability to continue to expand into the enterprise. We see some of the largest enterprises on the planet adopting the same software strategies that technology-first companies have been adopting for years. That's because every company is needing to begin to figure out how it uses software to effectively compete in the market.

We see strength from the developers is that as they began working for those larger companies and as those companies start employing software developers of the world, that those developers are going to bring in the tools that they know how to use to use those jobs, and we see an increasing number of enterprise companies in our pipeline. As those companies begin to leverage the power of software and invest in developers to build differentiated software offerings, those developers will bring Twiliio, and we see that as the opportunity.

Great. Also, Lee, I heard you comment on the gross margins. They did come in a little higher than we would have expected, and I am curious if you can just comment on what you are experiencing in terms of pricing discussions with carriers as your volume grows so rapidly, and I would think that trust in the brand is moving further ahead of the field. Are you benefiting at all just in terms of the pricing that we see manifest in your cost structure over time?

Yes, Mark, I'll answer that in two parts. The first part, I wouldn't read anything into a fluctuation on a quarterly basis. Our gross margin can fluctuate based on geographic mix, customer mix, rollout of infrastructure and all, and as a refresher, we are not running the business to maximize for gross margin in near-term, we are running the business for reach and scale. The broader answer, as we continue to scale the business, we do see benefits from our super networks so we continually see improvements in our cost structure as we grow and scale the business.

Okay. One last one, Jeff. I am wondering just how you are looking at the usage of in-app notifications or push notifications and if you could comment on how you are notified a product is playing into that trend. For instance, are you seeing that as a net positive? If the customers are deciding if they can't solely rely on in-app or push notifications and therefore they have to orchestrate across multiple channels, is it something that is driving the interest in the notify product?

Yes, I think that is roughly accurate, Mark. There are obviously many channels available for companies to communicate with their customers. A lot of it actually ends up coming down, because customer preference about how they want to be communicated with. And while push has been around for a while, probably 2010 when it first came out, what you find is a different customers will be in different places.

One customer you may be able to talk to via SMS, because essentially what you have is their phone number. Another customer may have decided to download your mobile app, and when you ask them enable notifications, they say yes, well now you can actually talk to them via push. Then again, maybe the following week they get a new phone and they forget to reinstall your app, so now you can't talk to them over push and so you will fall back to SMS.

Those are the kinds of real-world scenarios that companies see as they try to communicate with their customers over multiple channels. And that is why we built the Notify product, because it takes care of the very complex business logic needed for a company to keep track of how best to communicate with each one of its customers based on the preferences of that customer.

Thank you very much.

Your next question comes from the line of Pat Walravens with JMP Securities. Your line is open.

Terrific. Thank you and congratulations.

I was hoping we could dig into the dollar base net expansion rate a little bit. Because it is really best in class in terms of our coverage universe. So maybe if you could comment on how it is that you generate such high, effectively, same-store sales. Also, Lee, what should we expect the trend to be like in that number just because you have some tougher comps coming up.

Thank you.

Thank you, Pat. This is Jeff. I will answer a little bit about how the mechanic works there and I'll let Lee answer some of the quantitative aspects. The way we look at our usage-based model as part of our business model for innovators provides us multiple expansion vectors inside of every account. As a developer starts building and using our product, they will often spend very little, because as they're prototyping it is a very small amount they need. And we think that is a positive thing because the more efficient you can get started building, that is going to encourage more innovation.

Then we take some prototype and roll it out maybe at the beta to a subset of customers, the usage is going to grow and therefore our revenue is going to grow. Then when they roll it out to the entire customer base, again, the usage will grow and our revenue will grow. As an innovation migrates through the product development lifestyle, Twilio's revenue grows. That is the first vector driving our dollar base net expansion rate.

The second vector is when that developer does that again and she builds the next use case and then that use case gets traction as it goes from a beta to a general (inaudible), the same thing happens again, drives more usage, Twilio gets more revenue.

The third vector of expansion is just as our customers are building their own customer bases and they have more people to communicate with as end-users, that will also drive more usage and increase our revenue. The three vectors of growth are working inside of our customer base to drive this dollar base net expansion rate that we see.

With that, Lee, I'll let you answer the second half of that question.

We're not going to be guiding specific to expansion rate, but I think it is a good point talking about some comps and what happened over the prior quarters. The past two quarters we had a very high revenue growth rate, in the 90%s, and very high expansion rates. We benefited from some easy compare over the prior year.

If you actually look at our revenue growth rate was in the 90%s the past two quarters and in the prior quarter, the prior three, four quarters in the 70% range, a similar trend in terms of expansion rate. We did have the benefit of comparables over the last quarters. Again, I'm not going to guide going forward, but the inputs to the business are strong, remain strong. Those vectors that Jeff talked about will hold. So we do expect to have strong expansion rates going forward.

Great, thank you very much.

Next question comes from the line of Bhavan Suri with William Blair. Your line is open.

Hey. Thanks for taking my question and congratulations. Apologize for the background noise.

My first question, maybe to Jeff, is on the partner channel, how you are thinking about investing to drive the partnerships? Obviously the platform is great, but are you seeing some of the partners start to build applications and some of these customized communication applications to their customers and how you guys are approaching going to market in that space? Thanks.

Thank you so we believe that partners, we call them solution partners, are an important segment of our business because it's solution partners. These customers of ours like Zendesk or Salesforce or Service Now who build solutions on top of Twilio and then sell that solution into the customer base.

That category of customers really allows us to address the fullness of the market. Whether a company wants to build a solution or whether they want to buy one, Twilio can power that customer regardless. We have historically seen success in the segment of the market. I think our ability to build APIs that understand the needs of a solution partner, building multi-tenet applications on top of Twilio is an important part of the investments that we've made and we will continue to make those investments.

We have a focused effort by our sales team to really engage with these types of new solution partners, and to further enable our existing solution partners, as well to build more advanced or enter new markets with Twilio. Good example of that is Zendesk launching last quarter Zendesk SMS, which extends Zendesk voice into a new channel communication that is also powered by Twilio.

Got it. One quick follow-up for me. You spoke a bunch about IOT here. Just a little bit of an of date on the relationship with T-Mobile, the programmable SIM opportunity and how you guys think about that and how does that layer into this business? Maybe not in, let's say, 2016 or 2017, but maybe two or three years out, how should we think about what that could be from a revenue opportunity perspective for Twilio?

Thanks.

Absolutely, and for the benefit of those listening in, this is our Twilio programmable wireless product. This is that SIM cards that are powered by Twilio that allow a developer to program every aspect of the communication that runs over that SIM card, whether it's voice, SMS, data, as well as the provisioning lifecycle. And we're excited about the opportunity to power both IOT devices with this product line, as well as BYOD use cases inside the enterprise.

I will say that the relationship with T-Mobile, who is our launch partner in this, has been fantastic and we're excited for where this product is going to go. It is the earliest days of the product. We announced it at SIGNAL, our developer conference, in May. And that was announced in a developer preview. As that product gets rolled from developer preview into broader a release and eventually into a [Geeaid] release, we will be excited to see what use cases developers build and how us get deployed. But right now is the earliest days of that product and in the most literal sense we can't wait to see what developers build.

Thanks. I guess it is clear you and Lee don't want to put up a 5% of revenue in 2018 kind of number. I appreciate it. Thanks for taking my questions and congratulations again.

Thank you.

Your next question comes from line of Richard Davis with Canaccord. Your line is open.

Great. Two kind of [goose] questions. One, you have talked in the past about investing in scalability and stuff like that. Could you maybe just flesh that out a little bit? Then this is more hypothetical, but you're working with Facebook on the messenger, there's all these chat bots and things like that. It's early days there. Do you see that business as a general statement taking off and things like that? It just reminds me of some of the artificial intelligence back 15 years ago. But I was just curious if you have point of view on that. Thank you very much.

Great. Thank you very much, Richard.

As far as the scalability goes, we view resiliency as one of the key aspects of the service that we offer. And we have a mantra internally that we also talked about at our conference for how we think about R&D at Twilio and that is to achieve agility with resiliency. We're always investing a substantial amount of energy into not just building new products. As you've seen, we launched a number of new products in this quarter, but also increasing the resiliency and the scale and the quality of the products that we've already go in market.

That's why I think you see as a platform we do invest substantially in our product. We believe that is the right thing to do. We have invested 23% of revenue in engineering and R&D in Q2 and we believe that an outsized investment in R&D both from the perspective of innovation, but also from a prospective of resiliency is the right thing to do as a platform business.

I believe the second part of your question was about Facebook?

Yes, and just chat bots and all that whole -- there's a lot of talk about that. Do you think it is going to be a big thing for the industry in broad terms or not? I'm just curious because you're right there at Ground Zero.

Yes, a fantastic question.

We do believe that as with a lot of things there is a hype cycle around artificial intelligence and bots. Exactly where we are in that hype cycle is probably not entirely clear, but there is a lot of optimism around what can be done with artificial intelligence and bots.

We are excited for what some customers of ours are doing with bots and we think that an investment there makes sense and we believe that there is a lot of interest. We believe it is early days of discovering what the use cases are, although it's clear that there will be use cases as artificial intelligence gains more and more ability to do some tasks for us.

An interesting example of bots -- or actually I will also point out that sometimes I think bots mean artificial intelligence, but other times bots simply mean using messaging as an efficient interface to communicate with customers. And we see interesting things like the New York Times, who is currently powering is Olympic coverage with SMS to provide flash updates to its subscribers over Twilio SMS with what is happening down in Rio.

We think that is a neat innovation. We expect to see more of that as an innovation occur as many industries discover the power of messaging to drive new forms of engagement with their customers. As far as whether it is exactly artificial intelligence that drives all of that, artificial intelligence will play a role, undoubtedly, but it also may also be slightly over-hyped at the current, would be my guess.

Okay. Thank you so much.

As there are no further questions at this time, we thank you for your participation. You may now disconnect.