Twilio (TWLO) Q2 2021 Update

Twilio released their Q2 2021 earnings report on July 29th. The results were mixed. Twilio outperformed significantly on the top-line, but gross margin and operating margin were lower than investors had hoped. TWLO stock dropped 5% the day after earnings and has sunk almost 13% from the pre-earnings price to date. Looking forward to Q3, Twilio’s total and organic growth should accelerate slightly, but additional headcount and acquisition absorption will keep operating margins under pressure.

At a high level, Twilio is continuing their rapid expansion trajectory, marked by organic growth over 50% and significant uptick in international activity. These are being driven by the broad enterprise migration of offline customer experiences into new digital channels. This digital transformation often takes the form of building custom applications, for which Twilio’s programmable communications APIs provide key building blocks. I expect these secular tailwinds to continue and provide ongoing demand to drive Twilio’s future expansion.

Usage growth is originating both from organic product development and strategic acquisitions. Segment in particular generates a new dimension of insight for Twilio’s communications capabilities, allowing the combined company to close the loop on customer preferences and optimize marketing performance in a way that CPaaS competitors cannot match. Twilio is quickly evolving beyond just providing communications plumbing to enabling data-infused applications that drive enriched customer engagement for enterprises. These will be further enhanced by the return of trust to voice and SMS engagement through initiatives like STIR/SHAKEN and the Zipwhip acquisition.

I’ll list the most significant take-aways from the quarter below, with more details provided in the sections that follow. For additional background, readers can review my extensive previous coverage of Twilio results as well as the original investment thesis back in late 2019 when the stock was trading for $98. These set the foundation for my optimistic view long term and supplement this Q2 update.

• High Revenue growth is continuing at scale and keeps exceeding analyst models. With an annual run rate now over $2.5B, Twilio is still putting up 50% annual growth organically. Analysts have long modeled a drop to 30% growth, yet we still haven’t seen the inflection, even as revenue numbers become very large. Investors have to wonder when growth will slow down.
• While growth is strong, profitability improvements keep pushing out. Gross margins have dropped into the low-50% range. Operating margin is back to roughly break-even. Gross margins are being pressured by international messaging growth. At some point, higher margin software products should make up a larger percentage of the mix. In the meantime, gross profit continues to be plowed back into business investment.
• Some of this additional operational cost may be short term. Twilio grew total headcount by 93% year/year. Besides investment in go-to-market, this growth was bolstered by several acquisitions completed recently. Assuming redundancies will be worked out, we can expect some moderation in headcount expansion and cost savings going forward.
• On the flip side, the additional headcount, particularly in go-to-market functions, should drive further sales growth. S&M headcount grew by 103% y/y, with international headcount up 144%. International revenue was a large beneficiary, growing by 99% year/year. International will provide an outsized growth opportunity for Twilio going forward. This also aligns with recent commentary from other software providers about opportunities in international markets.
• The Segment integration is progressing well. Customer demand caused the team to accelerate co-selling initiation to July 1st, which is 6 months sooner than planned. I view this as a very bullish signal.  Segment brings a whole new dimension of product efficacy to the Twilio CPaaS offering, by enabling end-to-end tracking of consumer behavior following a communication. This introduces a significant competitive advantage over other players and will justify a pricing premium. Further, products built on top of Twilio Segment will be higher margin software offerings.

Financial Summary

Top Line Growth

Twilio continued to deliver high revenue growth in Q2. Even accounting for contributions from acquisitions, organic revenue growth was above 50%. The fact that this high growth is persisting at scale is worth consideration. Analysts have long been modeling a drop to revenue growth in the 30% range. Yet, each quarter, Twilio delivers organic growth near or above 50% annually. This has been the case going back to Q1 2020, after Twilio passed $1B in revenue for 2019. As of Q2 2021, Twilio’s organic revenue run rate is approaching $2.5B, with annualized organic growth even accelerating over the prior quarter.

As I will discuss later, the addressable market for Twilio’s services is large and the company has multiple avenues for product expansion. While it is hard to imagine a $1B+ company sustaining organic revenue growth above 50%, Twilio is doing just that. We have to wonder when the law of large numbers will kick in, if at all. In some ways, this could provide the underlying investment thesis for Twilio. That terminal velocity is sustainable for a longer period, justifying the continued investment back into the business generating relatively low operating leverage.

Getting back to the Q2 results, total revenue was $668.9M, which was up 66.9% annually. This beat the analyst estimate for $599.1M and the company’s prior guidance for revenue of $591M – $601M or growth of 48.7% at the midpoint. The beat over prior guidance was about 18% of annualized revenue, which is a huge increase. In Q1, the relative beat was about 16% of annualized revenue, reinforcing the trend that Twilio leadership is being conservative with forward estimates.

Looking at sequential growth, the story is more interesting. The initial Q2 revenue estimate of $596M was just $6M over the Q1 actual, causing some analysts to conclude Twilio would experience significant deceleration. The sequential revenue increase turned out to be about $79M over Q1 revenue, representing a sequential growth rate of 13.4%. This also drove acceleration in the annual growth rate to 67% in Q2 versus 62% in Q1.

Segment contributed $46.6M of revenue in Q2, slightly above the $44.6M reported for Q1. The smallish sequential growth of 4.5% is concerning, but I think we have to keep in mind that sequential growth from Q4 to Q1 was unusually large for Segment. Using an extrapolated estimate of about $35M of Segment revenue for all of Q4, the sequential growth rate from Q4 to Q1 was 28%. There may be some seasonality here and I’d like to see more data before drawing a conclusion about Segment’s growth.

Backing out the Segment contribution from the Q2 results yields organic revenue of $622.3M, up 55.3% annually. This compares to $545.4M of organic revenue in Q1, which grew 49.5% annually and highlights the acceleration going into Q2. Also notable is that sequential revenue growth on an organic basis was 14.1%. There is some benefit from higher A2P fees to carriers in these numbers, but not large enough to change the acceleration dynamic.

Another area of interest is international growth. In Q2, U.S. revenue was $452.9M and made up 68% of the total. International revenue was $216.0M or 32% of the total. Comparing to Q2 2020, U.S. revenue was $292.1M for 73% of total and $108.7M or 27% or total for international. The difference between U.S. versus international revenue is based on the location of the business account (physical address or IP).

Readers will notice that on an absolute basis, the revenue generated from international businesses nearly doubled year/year. And this is on a fairly sizable base, exceeding an $800M annual run rate. That growth is really important to appreciate, because if it continues at that rate, it will have an outsized impact on overall revenue growth going forward. Revenue growth in the U.S. isn’t stagnant either. Twilio is still maintaining a 55% annual growth rate in its home market.

Growth in international was expected, as Twilio has been investing heavily to drive it. They have been opening new sales offices outside the U.S. and highlighted enterprise customer wins outside the U.S. on a similar level to domestic wins in Q2’s prepared remarks. Acquisitions have been helping with geographic expansion as well. ValueFirst certainly established a strong presence in India and SouthEast Asia. There have been other small tuck-in acquisitions in the past, like Quiubas Mobile in Mexico.

This opportunity in international mirrors that for DocuSign and MongoDB. On their recent earnings calls, both companies highlighted growth in international markets as businesses globally are taking on digital transformation projects and building new custom applications to improve customer experience. These are driving demand for electronic signature and application data storage. The same argument applies to programmatic messaging and Twilio should similarly ride this secular tailwind.

Looking forward, Twilio is projecting this high revenue growth to continue. For Q3, they estimate total revenue of $670M – $680M, representing growth of 50.7% at the midpoint. This compared to the analyst estimate for $639.4M or growth of 42.7%. Importantly, the midpoint of the growth estimate for Q3 is about 2% higher than the original Q2 estimate’s growth rate. This implies that if Twilio maintains its consistent magnitude of beat in Q3, that the annual revenue growth rate will accelerate slightly over Q2. Further, on a sequential basis, if we apply the same $73M beat to Q3’s estimates as we saw in Q2, the sequential growth rate would be at least 12%. Over a four quarter period, a sequential revenue growth rate of 12% correlates to about 57% of annual growth.

These high sequential growth rates for Q2 and Q3 are important to consider, as they include the contribution from Segment. In other words, they provide a glimpse into what the total annual revenue growth rate should be after we lapse the Segment acquisition in Q4 2021. Unless Twilio sequential growth slows down substantially, the implication is that annual revenue growth in Q1 2022 will be in the 55-60% range. This is above the organic growth in Q1 2021 of around 50%.

Twilio doesn’t provide annual revenue guidance at this point. With Q2’s performance and Q3’s projection, I have modeled the following revenue estimates for the remainder of 2021. Note that Q4 is generally Twilio’s strongest quarter. The sequential growth rate from Q4 2020 on an organic basis was higher than my estimate (14.4% accounting for political traffic), but this makes for round numbers and allows some conservatism.

Q1: $590M (Actual)
Q2: $669M (Actual, 13.4% seq)
Q3: $750M (Est, 12.1% seq)
Q4: $850M (Est, 13.3% seq)

Total: $2.859B (up 62.4% y/y)

Analysts currently have $2.680B projected for 2021, representing growth of 52.1%. This is up from their estimates before Q2 earnings for $2.547B or 44.5%. However, it is still lower than my projection and assumes little upside to Q3 estimates and little sequential growth to Q4. Both of which are not consistent historically. I think this explains why Twilio’s full year forward P/S ratio is closer to the current value than I would expect. Because Twilio doesn’t provide a full year estimate for analysts to baseline, their estimates are low.

Profitability Measures

While growth metrics are very favorable, profitability measures for Twilio are a work in progress. The issues are explainable, but the fact that they must be explored creates a negative perception for investors, as they compare Twilio to software infrastructure peers. As I discussed in my Q1 update, I would like to see gross margins improve and operating margins start to increase. These would give confidence to investors that Twilio is realizing operating leverage as revenues scale. This would mirror the pattern that other software infrastructure providers like Datadog, Crowdstrike, Snowflake, Elastic, as others exhibit. That pattern is a gradual increase and stabilization of gross margins, followed by regular progression of operating margins above break-even and incrementally upwards towards 10%.

Twilio leadership insists that their long term targets for gross margin and operating margin are intact. Gross margin is expected to lift back into a range of 55-60% and will reach 60-65% in the long term model. For operating margin, the expectation is that it will eventually be at or above 20%. As Twilio revenue pushes over multiple billions, this operating margin target would provide a sizable amount of income.

The big question is when this inflection will occur. For comparable software infrastructure companies, we typically see this as they surpass $500M in annual revenue and certainly as they approach the $1B point. For Twilio, we can draw one of two conclusions – either the point of inflection will be a much bigger revenue number or it will never happen. Perhaps Twilio’s much larger TAM than peers ($110B by 2023) justifies the delay. A bullish view of Twilio assumes that rapid growth continues at this scale and the point of inflection may be around $4-5B in revenue. That is the position I hold for now.

In terms of specific metrics, Non-GAAP gross margin for Q2 dropped to 53.9%. This is down from 55.5% in Q1 and 56% a year ago. The explanation for the drop continues to be a larger revenue mix from lower margin products, particularly international messaging. For investors not familiar with the Twilio story, gross margin varies by product. Pure software products enjoy the high margins that one would expect, exceeding 80%. On the other end of the spectrum, SMS usage generates low margins, as Twilio has to pass some of the revenue to the network carriers. For international, these fees are higher than in the U.S., pushing the expected margin below 50%.

Because growth in the newer software products, like Video, Flex, Frontline, Conversations, Segment etc., is higher than over revenue sources, these should eventually pull gross margins back up. However, the base of SMS revenue is much larger and international expansion is a focus, applying a counter force to the gross margin mix. On top of this, messaging margins are eroding further in the U.S. as some carriers have introduced new A2P fees. These are a pass-through, but have been contributing another part of the decrease year/year.

We saw some evidence of product mix improvement in Q2’s report related to numbers for usage-based fees. Usage-based revenue is primarily earned from customers using Channel API products, like programmable messaging, voice and video. In Q2 2021, the percent of revenue from usage-based fees represented 72% of revenue, down from 76% a year ago. The remaining 28% of revenue in Q2 came from flat monthly fee products, like Email, Marketing Campaigns, Flex and Segment. While not a perfect correlation to gross margins, the flat fee products tend to represent those with high gross margins. Video and voice are favorable for usage fee gross margins, implying the growth in text messaging was substantial.

If we look strictly at gross profit growth, the high growth in revenue is marginalized somewhat. For Q2, Twilio generated $360.5M in Non-GAAP gross profit. This was up 60.9% year/year from $224.0M in Q2 2020 and 10.1% sequentially. In Q1, Twilio generated $327.4M of Non-GAAP gross profit, versus $207.7M in Q1 2020. This represented an increase of 57.6% year/year and 6.9% sequentially. Therefore, while gross profit growth was lower than overall revenue growth in Q2, the growth rate increased both annually and sequentially from Q1.

Moving to operating margin, Twilio leadership is investing almost all gross profit back into the business. This is evidenced by the practically deliberate effort to keep operating margin just above break-even. In Q2, Twilio generated $4.2M of Non-GAAP operating income for an operating margin of 0.6%. A year ago, this value was $9.5M with an operating margin of 2.3%. In Q1, they generated $17.3M of Non-GAAP operating income for an operating margin of 2.9%.

The Q2 results did beat Twilio’s estimate from Q1 for an operating loss of $27M – $22M. It’s nice to see that they can flip a projected loss into a slight gain, providing evidence that leadership has control over where operating income lands. However, the fact that operating income was less than the prior quarter is a little disconcerting. I think that acquisitions are coming into play here. As Twilio acquired Segment and then ValueFirst, they have also absorbed those companies’ operating expenses. This is likely bloating the expense lines until redundancies can be worked out. At the same time, the substantial increases in headcount, particularly S&M, should drive more growth in the future.

Looking to Q3, Twilio projected an operating loss of $30M – $25M, which is slightly lower than the loss that had been estimated for Q2 originally. This implies that operating margin may just break even, if Twilio delivers the same sized beat on actual operating income that we saw in Q2. Regarding the inflated operating expenses driven by accelerated hiring, the CFO commented on the Q2 earnings call. “As of the end of Q2, we have largely caught up on the hiring related to these investments, which will generate losses in the short term while allowing us to grow at elevated levels, and sets us up well to scale in the long term.” My interpretation is that he acknowledges that losses have been inflated in last few quarters, but that isn’t the long term view.

Looking at how gross profit is being distributed, we can see the relative allocations to each expense line. In terms of functional areas, what is most interesting is Twilio’s rapid increase in R&D and S&M spending. On a Non-GAAP basis, R&D spend increased by 52% year/year, while S&M spend increased by 76%.

• R&D = 18% (versus 19% in Q2 2020 and 18% in Q1)
• S&M = 25% (versus 24% in Q2 2020 and 24% in Q1)
• G&A = 10% (versus 10% in Q2 2020 and 10% in Q1)

Increases in headcount year/year are more dramatic. In their 10-Q, Twilio disclosed the changes in headcount by department. Headcount in R&D increased by 67% year/year at end of Q2. More significant was the headcount increase in S&M, which more than doubled with 102% growth. Twilio cites the reason “to expand our sales efforts in the United States and abroad.” That is an outsized increase in employees dedicated to generating more sales, including quota-carrying reps.

These increases, of course, included employees from acquisitions. I think we can assume there will be some attrition and efficiencies gained where there is duplication. The G&A line item increased significantly year/year in Q2, up 70% in spend and 103% in headcount. These functions, like HR, Legal, Facilities, Finance, etc., usually do offer the most opportunity for optimization following an acquisition, as there is often more overlap in these roles. I would expect to see some cost savings in this area as Twilio moves through absorbing these acquisitions.

Customer Activity

Twilio’s growth is driven by a combination of new customer usage and expansion of existing customer spend. Twilio has an efficient go-to-market motion that primarily targets developers. When looking for a solution to quickly add SMS, video, email or voice capabilities to their application, a developer can easily sign up for Twilio and integrate that communications function through a few lines of code.

An active customer is defined by Twilio as an individual account for which they have recognized at least $5 of revenue in the last month of the period (usually a quarter). Twilio uses this $5 threshold to filter out trial usage and light testing of the product.

At the end of Q2, Twilio had 240k active customers. This represented growth of about 20% from 200k at the end Q2 2020. Sequentially, the increase was just 5k over Q1 or 2.1%. Twilio ended Q1 with 235k customers, up from 190k the year prior, and added 14k in the quarter. This translates into 24% annualized growth and 6.3% sequentially. In Q4, they had added 13k. Q2 represents one of the slowest sequential growth rates for Twilio in total customers. It isn’t necessarily a seasonal effect. Between Q1-Q2 2020, Twilio added 10k customers, but there may have been a bit of COVID burst at that time.

Some of the increases in customer counts for the past few quarters were contributions from acquisitions. Specifically, Segment closed in Nov 2020 (Q4 adds) and had disclosed 5k customers. ValueFirst was acquired in March 2021 (Q1 adds) and had 2,500 customers at the time of acquisition. Segment likely shared a lot of overlap with Twilio. We know this because both CEOs commented about how the two companies were brought together in response to customer inquiries about using the two services together.

Active customer growth is definitely a metric to watch for Q3 and the remainder of the year. I think a few factors should keep customer counts growing:

• International Growth. Twilio is underpenetrated internationally. In Q2, companies based outside the U.S. contributed 32% of revenue, up from 27% a year ago. I noted earlier that international revenue doubled year/year, as a result of Twilio’s large investment in go-to-market resources outside the U.S. This high international growth should translate into higher customer additions going forward outside the U.S.
• Acquisitions. ZipWhip closed on July 14th. Their customer count should be included in the Q3 results. Zipwhip has an extensive customer set of over 30,000 businesses, including some marquee brands like Comcast, Barre3, LawnDoctor, Harry & David, Stanley Steemer, Massage Envy and the Detroit Tigers. Given that their business is primarily U.S. based, there will be a fair amount of overlap with Twilio, but even if 1/3 were new customers, that contribution would double the customer additions from Q2.
• Partner Network. Twilio launched their Build Partner program in 2018, but it really began to gain traction in 2020. Build program member companies help enterprises design, develop and deploy customized communications solutions using the Twilio platform. As Twilio products become more software-driven, they allow for extensive customization and integration with other business systems. These types of implementation projects are more meaningful for integration partners, attracting renewed interest in building practices around Twilio software products like Flex. As the Partner Network ramps up, this should create a new channel for customer additions.
• Direct Sales. As discussed earlier, Twilio has invested significantly in their S&M function, effectively doubling headcount over the past year. A large portion of these additional employees will be performing direct sales functions. This would include new customer marketing and outreach. I expect these efforts to contribute meaningfully to new customer additions. While the bottoms-up developer adoption provided Twilio with an inexpensive growth vector, direct sales will supplement this motion, layering on the benefit of outreach.

Also notable are some of the customer wins announced as part of the Q2 earnings report. Many of these were large enterprises that adopted higher-end software products. Enterprise wins are obviously important, as those customers would generate larger spend over time. Additionally, adoption of higher-end software products will help drive up gross margin.

Customer Wins

Twilio has been investing heavily in their go-to-market effort, particularly in driving sales growth at the enterprise level. This appears to be paying off, as they announced a number of large customer sales and expansions for the quarter. Interestingly, many of the new deals involved packaged software products, versus communication channels like SMS, voice and email by themselves. This may reflect deliberate content curation by the Twilio leadership team, but the increasing mix is refreshing to see regardless.

• Expanded their relationship with a Fortune 500 global healthcare company.  The customer selected Frontline (still in Beta technically) to help field reps interact with their clients in a secure, compliant manner.  The company plans to roll this out to 12k sales reps across the world.  As you’ll recall, Frontline provides an application that allows company reps to conduct business on their personal phone.  This app ensures communications are captured by the company, appear to originate from the company’s communications channels (versus the employee’s personal number) and are kept secure and private, all while giving the employee the flexibility to be mobile and not have to manage a separate company-issued device. This would represent the second large enterprise adoption of Frontline, following Nike at product launch.

• Expanded usage with a Fortune 500 airline looking to provide cross-channel messaging experiences for flight delays, gate changes, check-in reminders, etc.  They airline selected Twilio SMS for these use cases.

• Expanded their relationship with a Fortune 500 energy company.  The company will use multiple Twilio products including Conversations, SMS and Email to make their customer communications more proactive.  

• Signed several Fortune 2000 deals for Flex in Q2.  Leadership highlighted a leading retailer that will be using Flex, Conversations, Video and other products to build a virtual store. In this case, customers can initiate a chat over messaging or video to connect with an advisor to guide them on specific product selection.  This type of solution should help customers get better service and drive more digital sales.  More broadly, once one retailer in an industry segment starts offering support like this, others will likely follow.

• On the international front, Twilio signed a deal with TK Elevator (one of the world’s largest elevator companies) to power the voice, video and chat features of the elevator’s cabin assistance function.  An elevator passenger can initiate a voice call to the support team, allow video monitoring of the cabin and send chat messages back and back.  

• Signed a new deal with Philippine Airlines (PAL) for Flex.  PAL is migrating off of their legacy contact center and selected Flex for the programmability and multiple channels.  This deal was driven by a partner, Tata Consultancy Services, which is an IT service provider based in India.

• Segment expanded its relationship with AB InBev, the world’s largest brewer.  AB InBev is integrating Segment into their internal data platform to construct a single view of their customers and improve engagement.

At a high level, we can conclude that digital transformation itself is providing a strong tailwind for the adoption of Twilio products. As new customer experiences are created on digital channels, enterprises are discovering that they need to compete as heavily online to delight customers as they did in physical locations. Additionally, to differentiate themselves, enterprises are discovering they can’t just employ an off-the-shelf e-commerce or customer service package to support their online experiences. They need to emphasize the creation of bespoke customer engagement solutions.  Customers expect personalization and customization. For communications, they expect to be notified or reminded about all aspects of their relationship with that company. Particularly, if something they care about changes, they better be proactively informed.

A few observations about customer highlights from Q2:

• Companies are trying to be more proactive and helpful for their customers. This is likely due to competitive pressures driven by digital transformation.  If one airline or energy company starts sending out proactive notifications to customers when a flight is delayed or power is out, then customers will expect it from all companies in that segment.  This is forcing successive “bar raising” around customer service that will benefit Twilio in a non-linear way. If one airline does something unique, then the other 10-15 airlines will be expected to do the same.

• Several Flex deals were highlighted, versus just one in Q1 (with a Fortune 1000 luxury retailer).  Q2 brought another Frontline deal and included several references to Video, Conversations and Chat.  These are all high margin product offerings.  As these scale up, the incremental revenue should help boost overall gross margins.

• The investment in building out partner channels for software implementations appears to be gaining traction.  The deal with Tata Consultancy and Phillippine Airlines provides an example of this.

Besides landing new customers, the other major contributor to Twilio’s revenue growth is expansion of spend from existing customers. While Twilio doesn’t disclose the revenue share between new and existing customers, I estimate it is probably 1/3 from new and 2/3 from existing. This is based on leadership commentary in the Q1 earnings call and comparable metrics from peers. As I mentioned in my Datadog Q2 summary, Datadog disclosed that 70% of their annual revenue increase in Q2 came from existing customers and 30% from new customers. Twilio and Datadog likely share similar expansion drivers and have comparable net expansion rates.

As investors know, Twilio’s customer spend growth is measured as the Dollar Based Net Expansion Rate (DBNER). For Twilio, DBNER represents the percentage growth in revenue generated from the cohort of active customers in the year ago quarter as compared to the present quarter for the same cohort. This calculation would include decreases in spend and churn – which is worth highlighting as some peers exclude churn from this calculation.

For Q2, DBNER increased to 135%. This is up from 133% in Q1 and 132% a year ago. Given that it looks back a year, it would not include any of the revenue contribution from Segment or other acquisitions.

What investors will notice from this chart is how durable Twilio’s DBNER has been historically. As I pointed out earlier in the discussion around revenue, most companies hit a point of “large numbers”, in which both net expansion and revenue growth start to taper as a consequence of absolute size of revenue. This general rule doesn’t appear to be kicking in yet for Twilio, even as they pass a $2.5B organic revenue run rate. Most SaaS companies start to taper around $500M-$1B. So, we are left to wonder when deceleration of DBNER should occur, if at all.

At some point, Twilio’s customer spend expansion should bump up against a ceiling. In the near term, however, I could see it continuing at a high rate due to the following drivers:

• International Expansion. The same expansion rates recorded for Twilio in 2019-2020 for domestic customers should apply to international as they ramp.
• Increased Usage. Customers tend to onboard with one use case for customer communications for a single product and then multiply that across many use cases and new applications over time. For example, they may start by sending an SMS to remind a customer of an upcoming appointment and then expand to other types of SMS prompts.
• New Products. Customers may start with one product, like Email, and then add new products like SMS, Voice, Video, etc. over time. Additionally, Twilio’s new software-driven offerings, like Flex and Frontline, have a multiplier effect on usage. Flex has its own subscription fees, but also generates revenue for Twilio based on usage of the underlying communications APIs like Messaging, Voice and Email.

Beyond these factors, I think a bigger expansion opportunity is being driven by some of Twilio’s recent acquisitions. These will increase customer usage through deeper engagement of that customer’s end users. Many of Twilio’s communications channels are currently used by customers in a “single shot” mode. This means that the customer will send a one-way communication to a customer, without expecting or tracking a response or next action. One example might be an SMS that just notifies the customer of a status, like an upcoming appointment. Another might be an email that announces a sales promotion. These generate usage fees for Twilio through the single communication.

However, many of Twilio’s newer offerings, aided by acquisitions, are encouraging a longer back-and-forth engagement with the customer’s customers. A simple example is two-way messaging, where the recipient can respond to the business’ message. In the example of the appointment reminder, a customer could confirm the appointment, or ask that it be rescheduled. This type of threaded messaging is supported by the capabilities in Twilio’s new acquisition of ZipWhip, which provides an application interface for businesses to manage a full messaging thread with customers.

Also, Segment is making the ability to track the end user’s response to a communication easier. Their new Journeys product, released in June, makes this easy. The idea is that a Twilio Segment customer can lay out a path (journey) of communications and follow-up actions for users that encourages them to move towards a business goal, like a subscription, purchase or sign-up. Each step in the journey generates a communication, which would be sent over Twilio’s network, driving more usage fees.

I expect this kind of threaded communication to significantly increase usage for Twilio customers. We can think of it as another dimension of growth which could be non-linear. If you consider how many messages you receive from businesses today, versus what you could reasonable handle, I would posit it could go up by an order of magnitude. Personally, I get 1-2 SMS messages from businesses each day, which represent useful engagement. This could balloon to 10-20 a day and I probably wouldn’t be bothered.

This isn’t spam, mind you. These are businesses which I have given permission to contact me, whether my health care providers, fitness partners, e-commerce vendors or delivery services. Further, ZipWhip provides the ability for these businesses to use an identifiable toll-free number for these communications. That further enables trust in these communications, raising the tolerance level for receiving them.

I expect Twilio to benefit from the expansion of trusted communication channels. A Twilio customer will often test one path of communicating with their users and measure the ROI. Once they discover the lift to their desired business outcome, they duplicate this many times over on other channels. Trusted communications will make consumers more receptive to business outreach. And Segment closes the loop, allowing for multi-step series of interactions to be implemented and tracked.

Other Metrics

I provided some updates earlier on headcount growth in certain segments. I will cover it more broadly here. I think that employee growth can be a very useful signal for investors to predict future sales growth. For example, Datadog went on a hiring spree in 2020 (when many companies were pausing hiring) and increased headcount by 56% year/year. Now, in 2021, we are seeing the impact of that, as revenue growth has accelerated in Q2 to 67% y/y, higher than the prior 3 quarters. I suspect that this rapid growth is partially fueled by the increase in headcount.

Aggressive hiring often precedes growth. It’s logical, of course, that an increase in sales people, particularly direct sales, will drive more revenue. As investors know, most salespeople have a quota to hit. So, more sales people translates into more sales. This is why analysts often ask leadership about sales hiring plans on earnings calls.

If we look at Twilio’s headcount growth over the last 12 months, we see a similar pattern of rapid expansion of the employee base with a heavy emphasis on sales and marketing. A good percentage of this is from acquisitions, but not all of it. Headcount from acquisitions should be as productive, regardless. Often salespeople at the acquired company welcome the opportunity to have a broader set of products to sell, as they can realize higher commissions.

At the end of Q2, Twilio reported 6,334 employees. This is up 93% from 3,284 employees at end of June 2020. Twilio has nearly doubled headcount in the last 12 months. Further, in the 10-Q, they shared that international headcount grew from 924 employees to 2,250 in the last 12 months. That represents a 144% increase, more than doubling, off of a good sized base. As I mentioned in the profitability section, headcount in S&M grew by 102%.

We are seeing the impact of this hiring, as international revenue doubled in the last 12 months from $108.7M a year ago to $216.0M in Q2. This was obviously higher than the overall revenue growth of 67%. Given the rapid increase in headcount, much of which was allocated to regional sales offices outside the U.S., I think we can expect this pace to continue. With $216M making up 32% of Q2’s $669M in total revenue (up from 27% a year ago), another double from here would push up overall revenue growth rates.

This international growth isn’t isolated to just Twilio. Several peers in software infrastructure have called out strong international uptake for their services. Okta saw international revenue increase by 103% year/year in their most recent quarter. International revenue now makes up 21% of their total, versus 16% a year ago. DocuSign is seeing a similar trend. As they have been leaning into international growth, they have seen its contribution increase to 22% of total revenue in the most recent quarter, growing by 71% year/year.

The takeaway for international seems to be that companies outside the U.S. are fast following the migration of their operations and customer experiences to digital channels, much as the U.S. did over the past couple of years. This will provide a nice tailwind of growth for software service providers, including Twilio, as this transformation plays out. Given that the rest of world economy is about 3x larger than U.S. GDP, international should eventually make up a greater component of Twilio’s business.

Twilio doesn’t regularly report metrics on large customers (like spend over $100k, $1M, etc.). In the last investor presentation from March 2021, we learned that Twilio had a relationship with only 18% of the Global2000 as of Q2 2020. This number increased by 100 or 39% in the preceding 2.5 years. However, spend from that customer set increased by 7.5x, from $4M to $30M. That growth underscores the long road ahead for Twilio. They can continue to focus on adding new customers in the Global2000 and then expect a long tail of expansion from those customers over time.

Developer engagement provides another view into a future growth driver. As mentioned previously, while Twilio has been investing in top-down sales resources, they still have a strong bottoms-up GTM motion. Developers are often the ones to introduce a Twilio product to their organizations. This is usually just an inexpensive use case as a POC. Once the efficacy is demonstrated, then usage starts to ramp up quickly. Additionally, as developers move from one company to another, they will bring their preferred third-party services, including Twilio, with them.

At IPO in 2016, Twilio reported just over 1M developer accounts on the platform. At the end of 2020 (about 4.5 years later), this number exceeded 10M for a 10x increase. Statista reported about 24M developers worldwide in 2019, growing to 28M by 2023. Twilio still has room to expand here.

Product Development

Twilio’s platform is evolving rapidly. As I have covered Twilio for a couple of years, I have noticed a shift in their product marketing positioning. In the early days, the story was about enabling developers to easily add communications to their applications through programmable APIs. These APIs were easy to consume, with low implementation overhead. Use cases tended to revolve around a single shot communication – like sending a one time SMS or email to notify a user of some event. Little data was collected about what actions the user performed after that communication and Twilio maintained limited history of the user’s interactions with the business.

As Twilio began adding new software-driven products, their messaging expanded into enabling communications across multiple channels and required Twilio to track a user’s interactions over time. This is because an engagement might start with an email, but later transition to an SMS or a voice call. The creation of higher-level customer experience products like Flex and Conversations required maintaining a history of user interactions across multiple channels. However, that history still lacked depth of context. Twilio might have a record of sending multiple emails or SMS’s to a user, but had little insight into why or what business outcome was achieved. Without having access to the business context and full history of that user’s interactions with the business, Twilio could offer little insight to help optimize a business’ relationship with their customers. Sure, they could tell the business that a user seemed to respond best to SMS messages sent in the evening, but couldn’t tie any of that insight to the user’s actual interests, intent or past activity.

Of course, that all changed with the acquisition of Segment. Segment fills in the missing context of the “why”. It provides Twilio with access to first party customer data. Segment allows the business to instrument every digital channel they manage to track every detail of the user’s engagement. Because data is collected directly from the business’ own applications and reflects actual actions taken by the user, it is categorized as first party data. This kind of customer data is considered the most valuable and most reliable source for understanding how to best meet a consumer’s needs. This is in contrast to third party data, that observes their behavior on other applications or tries to fit users into generalized interest groups. More importantly, privacy concerns are making reliance on third party data sources a business risk.

First party data allows the business to create a truly personalized customer experience. This implies that a business remembers what interactions a customer had taken previously, and uses that data to modify the experience going forward to reflect that user’s interests or needs. This modification applies not just to the business’ web site, mobile apps and other applications, but also their communications with user, whether through marketing campaigns, notifications or updates. And this personalization needs to span all communications channels, whether email, SMS, voice, chat, video, etc.

Twilio Segment published a “State of Personalization” report in June 2021. More than 3,000 companies, consumers, and marketers worldwide participated in the study to analyze their attitudes, expectations, and experiences with personalization. In the report, 60% of consumers said they would become a repeat buyer of a business’ services if they were provided a personalized experience. And over a third of consumers said they would return to a brand following a personalized experience, even if there was a cheaper or more convenient option available elsewhere. These numbers are up significantly from the last report published in 2017.

The take-away is that Segment allows Twilio to address the full lifecycle of a business’ relationship with their customers. This capability makes Twilio’s platform many times more valuable to enterprises. Twilio moves far beyond a CPaaS provider with some useful APIs for sending an SMS or email, to a platform upon which a full-featured customer engagement program can be built. This does two things for Twilio. First, it will significantly increase the ROI that businesses realize from their customer interactions, whether to drive new business or ensure satisfaction. Second, it leapfrogs Twilio ahead of CPaaS competitors who can only offer single shot communications with little insight into the business outcome that interaction actually accomplished.

In the Q2 earnings prepared remarks, CEO Jeff Lawson used a large portion of his section to emphasize this product inflection. This wasn’t highlighted in the Q1 report and certainly wasn’t the full story before the Segment acquisition.

The move to digital creates a virtuous circle for companies, whereby the more consumers engage over digital channels, the more data is available for companies to learn about how their customers are engaging with them. Companies can learn from this data, and continually make improvements to deliver highly personalized interactions, across any channel, at scale.

This is where Twilio’s platform will play a significant role. We are enabling companies to build direct-to-consumer engagement across nearly every digital channel, and helping to leverage first-party data to develop a single view of the customer. There are billions of consumer records — from website visits and in-store purchases to customer service requests and app interactions — that companies have to sort through, trying to connect them to an individual across multiple systems, all with the goal of building a better understanding of who that customer is and what they like.

Our platform enables companies to navigate this complexity and build incredible B2C experiences that are highly personalized, at scale. We started with offering different channels, like messaging, voice and email, to help companies connect with their customers. Then, we built Flex, a contact center platform that allows companies to build a solution that fits their specific business needs, in turn driving a better customer experience. And we acquired Segment, adding an incredible data platform that can pull together data from hundreds of different systems to create a single view of the customer.

Jeff Lawson, Q2 EArnings prepared Remarks

I realize this sounds a little like a marketing pitch, but this shift really is a big deal. It elevates Twilio’s platform to a higher level, representing a broader playing field for their product offerings. I think it sets the stage to focus on building more value-add packaged software products that combine customer data with communications.

Twilio Segment Journeys which was announced in June is just the first step in what I think will be a much broader product offering. Journeys enables the weaving together of customer activity data across multiple digital channels to create a single view of their experience with a company. Experiences can be modeled into a series of steps, called a journey, that nudge a customer towards a desired business outcome (like a purchase). These nudges are often communications, like an email or SMS, which would be delivered by the Twilio platform.

I think Twilio is working on more capabilities like Journeys that will help enterprises optimize their interactions with customers. Twilio leadership and the Segment CEO have hinted at big announcements coming during the Signal user conference in October. Since announcing the Segment acquisition in November 2020, the two teams have been busy combining capabilities into a new, unique offering. I suspect that will be unveiled at Signal.

Importantly, the two teams aren’t hypothesizing this new combined offering in a vacuum. Customers have been asking for it. In fact, due to customer demand, the Twilio leadership team decided to begin co-selling Twilio and Segment as of July 1st, which is 6 months earlier than planned. Additionally, during an analyst event with Wells Fargo in mid-August, Segment’s CEO described how the real impetus for agreeing to the Twilio acquisition occurred after Segment’s own large customers encouraged the combination of the two companies.

Product Releases

With that overview, let’s take a look at some of the product announcements made during the quarter. While we have Signal to look forward to in late October, Twilio pushed out a couple of product enhancements in the last several months.

First in July, they announced achieving full compliance with the STIR/SHAKEN protocol. This is an industry standard intended to ensure that phone calls transmitted across a telecommunications platform are legitimate and authenticated. Going forward, all of Twilio’s calls are stamped with an approval attesting to the fact that terminating service providers and end users can trust that the call is not illegally spoofed.

This is all part of an industry effort to reduce robocalling and rebuild trust with consumers that voice calls will originate from desired sources. The STIR/SHAKEN protocol, as well as other industry and FCC programs, allow participants to raise the bar on regulatory compliance and eliminate unwanted inbound calls. The outcome will be that caller identity is verified and that callers are contacting consumers for a legitimate reason. After the abuse is cleared out of the system, voice calling can become a more effective method for businesses to reach consumers. In its current state, most consumer won’t take calls because they can’t verify that the caller is a legitimate and authorized relationship.

This represents a similar transition that email went through over time. The identity of the sender is now generally verifiable through protocols like DKIM, DMARC and SPF, as well as the spam filtering through activity monitoring and data sharing done by email providers. As a result, email is generally functional as a marketing outreach channel.

Once these industry standards and best practices are rolled out across the voice calling industry, we may see a similar improvement in efficacy. The benefit to Twilio and other industry participants (including all CPaaS providers) will be more usage of voice calling as an outreach channel. For Twilo, that would drive incremental revenue.

Also in July, Twilio announced a new Beta product offering, called Twilio Live. It is a cloud-based service that allows businesses to easily embed live, interactive audio and video streaming solutions into their applications. Like other Twilio programmable APIs, the product is targeted at developers, providing access to a low-latency and secure live streaming platform. It is built for high scale, supporting broadcasts to millions of users. To make the service programmable and easily integrated into other applications, the starter kit includes server-side API code modules and client-side SDKs for iOS, Android and JavaScript.

As part of the release, Twilio is providing reference applications for multiple clients that support audio-only and video streaming. The applications come with built in features for speakers and audience members, as well as interactivity capabilities like chat, screen sharing, polling and the ability to bring audience members onto the stage. Developers can freely use and extend the source code from these reference applications. This starter code simply jumpstarts the developer’s work.

The Twilio Live service could be applied to a number of business use cases including marketing events, concerts, fitness classes, educational sessions, live shopping and sporting events. As part of the product announcement, Twilio revealed that Reddit will be using the service to power a new live audio feature for community conversation, called Reddit Talk, which Reddit appears to be positioning as a Clubhouse competitor.

Twilio’s existing Video product is geared for one-to-one or small group interactions. Twilio Live significantly expands the the number of users that can participate in an event, making it more akin to recent social broadcast services like Clubhouse and Twitter Spaces. Currently, Twilio is inviting customers to try out the product as a Beta and will likely release it to GA with pricing at some point in the future. I think this offers Twilio another revenue stream and access to a different set of application use cases.

In May, Twilio expanded their IoT portfolio by offering Super SIM as an eSIM Consumer Profile for UICC-enabled IoT devices. For IoT devices that make their connection to a cellular network through a SIM card, they are bound a particular carrier based on the data in the physical SIM card. Switching carriers requires updating the SIM card. An eSIM allows the device’s profile information to be stored electronically on the UICC, which enables updates to be made over the wire.

When IoT fleet managers need to switch network providers, they traditionally would physically swap SIM cards out on IoT devices. This can be a cumbersome process. Twilio’s support of the eSIM standard allows companies to load Twilio’s multi-carrier Super SIM profile onto their devices over-the-air, and then switch between the different SIM profiles as needed. This eliminates the need to physically interact with the devices, all while improving the end user experience by delivering best-in-class IoT connectivity.

Twilio is continuing to build out their IoT capabilities. The IoT market is still nascent, and Twilio is ensuring they can occupy a position as a leading provider of IoT communications services. This builds on their Electric Imp acquisition from July 2020. eSIM support for SuperSIM is being offered to existing IoT customers as a private beta and later will be rolled out more broadly.

In a related development, Twilio added Super SIM Connection Events to their Event Streams product.  This allows users to track activity for IoT devices through Super SIM.  It captures a number of events, including device registration, attachment worked/failed, cell tower activity, data session start/end. This is will be useful for IoT fleet managers to optimize the performance of their IoT devices.

state of IoT was raised as an analyst question on the Q2 earnings call. Twilio leadership replied that 5G would be a strong driver of demand and that roll-out has been delayed a little by the semiconductor shortage. However, IoT is expected to represent a large market opportunity for Twilio in the future and they have laid a strong foundation to capture that demand as it builds.

Segment

As I mentioned above, I think the addition of Segment to the Twilio communications platform represents a game-changer. While Twilio’s other acquisitions have helped round out its reach and feature set within CPaaS channels, Segment is a more foundational addition that moves Twilio into a new level of customer engagement product offerings. This enables Twilio to provide enterprise marketers with the tools to create personalized outreach campaigns to their customers.

Segment’s CMO was recently interviewed by Diginomica. She highlighted these advantages and the timeliness of the Segment / Twilio combination as it relates to industry trends with third party data. The general backlash against relying on third party data sources for customer targeting is creating a renewed interest in leveraging the data that a business already has access to, their own first party data. To make this data useful isn’t trivial – it has to be collected from all digital touchpoints, cleansed, normalized and combined into a single view of each customer. In her view, connecting this customer data with tools to reach out to customers provides a powerful combination.

So marketers have access to a lot of technology. What they don’t have direct access to is data and good data. Data that has been cleaned and unified into this single view of the customer. And what’s powerful about connecting a CDP with actual tangible tools that marketers can use, such as building a customer journey, is the next era – is being able to connect the two.

Without that great data, accurate data, compliant data, you’re not going to be able to deliver the right message at the right time across the right channel for your customers.

Segment CMO, Diginomica, August 2021

To begin leveraging these combined capabilities, in June, Twilio Segment announced a new product called Journeys. Segment’s Customer Data Platorm provides the ability to weave together customer activity data across multiple digital channels to create a single view of a customer. Not only does the CDP store data about the customer’s preferences and profile, but it also records the history of interactions they have made with the business. Journeys sits on top of this data, allowing marketers to organize it into a series of steps (called a journey) that combine to achieve some business outcome (like a product purchase, sign-up or subscription).

As part of modeling a Journey, marketers can create nudges that help move the customer along the Journey’s path. Many of these nudges take the form of a communication, like an email or SMS message, but can also represent an ad or content module with the web site. For communications, these are delivered by Twilio’s communications channels.

This capability will allow marketers to create user engagement campaigns that boost conversion rates, lower cart abandonment and improve customer retention. Marketers can design multi-step interactions with a user-friendly visual builder. Twilio Segment also provides a set of starter templates. These represent common scenarios for digital engagement channels, like new customer onboarding, repeat purchase campaigns and trial to paid conversion. Or, the marketer can create their own.

To make integration easier for enterprises, on August 19, Twilio Segment unveiled a new CDP Developer toolkit. This provides developers with the tools they need to customize and extend the Twilio Segment platform. The toolkit consists of a set of code modules, SDKs and integrations for common tasks in user data collection, cleansing, aggregation and normalization. These tools are all designed to save developers time, so that they don’t need to write code for these common functions from scratch.

As an example, new users will often have multiple profiles on different business digital channels. Yet, a marketer will want these consolidated into a single persona. The Developer Kit provides tools for user identity resolution and profile aggregation. If an engineer tried to write this code themselves, it would add a lot of complexity to a customer engagement project. Yet, Segment provides these standard libraries as part of their offering. This creates stickiness for Segment’s CDP and lowers the barrier to adoption.

Beyond these product announcements, we know that the Twilio and Segment product teams have been working on bringing the two platforms together into a single offering. Leadership is hinting that we should look forward to some new features and product introductions at the Signal user conference in late October.

Given customer demand, as part of the Q2 Earnings announcement, Twilio leadership said they have started co-selling Twilio and Segment together. This started on July 1st (beginning of Q3) and represented an acceleration of the prior go-to-market target by 6 months. I think this bodes well for the opportunity for Twilio to begin realizing the value from this acquisition, as they can at least cross-sell the Segment solution to their existing customer base of over 200k businesses.

Interestingly, the Twilio web site now lists Twilio Segment under the Applications section of the Product menu. This links to a product page describing the Segment offering and the opportunities it provides in combination with Twilio’s core communications capabilities. They now show a graphic highlighting some notable brands that are using the combination of Twilio and Segment. I think this shows some evidence of the immediate traction that the new co-selling effort will drive.

ZipWhip

On May 17th, Twilio announced the acquisition of Zipwhip. Originally, they expected that the deal would close by the end of 2021. However, Twilio was able to work through the deal quickly and announced on July 14th that they had completed the acquisition. Zipwhip will operate as part of Twilio’s Communications Platform Unit with the current CEO of Zipwhip reporting to that business unit’s GM.

Completion of the acquisition in July means that Zipwhip’s revenue contribution and customers will be included in the Q3 results. As part of the Q2 earnings report, Twilio leadership did not include expectations for Zipwhip in their Q3 estimates, as they are still working through the integration plan and purchase accounting. They will provide an update on Zipwhip performance in the actual Q3 report.

I think this provides some opportunity for a little upside to Q3. We don’t know Zipwhip’s actual revenue, but I have seen estimates ranging between $20M – $50M a year. At a 20 P/S ratio, the acquisition price of $850M would imply a revenue run rate of about $40M. In a company blog post in April 2021 announcing a new developer hub in Calgary, Zipwhip’s CEO included a stat that the company doubled its revenue in 2020 and increased texting traffic over its network by 129%.

Additionally, Zipwhip has a customer set of over 30,000 businesses.  Some portion of those would be new to Twilio and would therefore add to Twilio’s existing 240k customers.  Zipwhip lists some marquee customers on their site like Comcast, Barre3, LawnDoctor, Harry & David, Stanley Steemer, Massage Envy and the Detroit Tigers.  To date, Zipwhip has enabled over 4M landlines with text message capabilities, primarily in the US and Canada.  In the acquisition announcement on Zipwhip’s blog, the Zipwhip CEO mentioned a stat that “66% of Americans received at least one text from our network.” Zipwhip has about 275 employees, primarily based in the Seattle area. These would join Twilio’s employee base.

On the product side, Zipwhip offers a potentially valuable and unique capability for Twilio’s platform. Their product is similar to Twilio’s messaging product, in that they provide tools to allow businesses to send messages to customers. These are primarily over SMS. Messages are initiated through a CRM like application, with a UI that allows the business to send any customer a message, reply to responses and view the prior communication history. The business can also set up automated messaging in the tool, like sending all clients a reminder who have an appointment that day.

The key differentiation for Zipwhip’s service is that their messages are sent from the business’ actual phone number.  To accomplish this, they created unique integrations with the major wireless carriers, like AT&T, Verizon, T-Mobile and Sprint, so that the incoming and outgoing text messages could be intercepted by Zipwhip’s platform.  Zipwhip claims to be the first company to enable texting on a business’ existing phone number, whether a landline, VOIP or toll-free number. They are also well positioned competitively in that they have the most direct carrier connections of any provider and are the sole channel for toll-free messaging in North America.

Because Zipwhip’s messages are tied to the business’ actual phone number, the recipient can identify the sender and have confidence that it is a legitimate communication.  It also promotes two-way messaging, allowing the recipient to respond to the text message. Zipwhip’s platform makes this easy, as the message response is intercepted from the carrier and directed into the CRM-like app provided by Zipwhip to customers.

Coupled with STIR/SHAKEN protocols for voice calling, Zipwhip’s ability to identify the sender of a message clears the way to re-introduce trust into these communication mediums. When voice calling and text messaging were first introduced, recipients generally trusted that the sender was legitimate. Then, spammers discovered how to abuse these channels through automation and identity spoofing. Now, Twilio has the opportunity to make these programmatic communications trusted again.

Making every call, every text message, coming from a trusted identity, not just some phone number but a business name. And with a padlock that’s like your web browser where you know that it’s trusted. When they say that they are the pharmacy or your kid’s school, you trust that that’s true. That’s the world we’re building toward.

Jeff lawson, TWilio Q2 EArnings Call

That return to legitimacy should make these channels more viable for business outreach, effectively raising usage across the board. As the largest provider of programmatic communications services, it is in Twilio’s interest to help establish these standards and clean up the industry. As trust returns to these channels, Twilio’s business will benefit immensely because the frequency of trusted communications could increase by an order of magnitude.

Competitive Landscape

I have spent a fair amount of time analyzing Twilio’s competitive landscape in past coverage. In terms of direct competitors in the CPaaS space, not much has changed. Twilio is still the largest and most influential provider by far. They are also continuing to extend their lead, growing faster than publicly traded peers. There are a couple of new developments that I will explore below, but for the most part, Twilio can still be viewed as the “gorilla” in the CPaaS space.

Before the Segment acquisition, Twilio offered customers the most complete solution. They covered more communications channels and geographic regions than any competitor. Twilio provides developers with the ability easily add voice, video, SMS, email, WhatsApp, FB Messenger and Chat integration to their applications. Additionally, these services are available in over 180 countries. For an enterprise with global reach and an omni-channel posture, Twilio checks the most boxes.

Twilio’s breadth of coverage makes a compelling developer resource efficiency argument, relative to competitive point solutions. To save money, a sufficiently staffed development team could try a DIY approach. They could create their own integrations to address each communications channel individually. They might code directly to Facebook’s WhatsApp API, stand up their own email servers and use separate CPaaS providers for SMS and voice. To achieve global reach, the development team might also have to integrate with a few other international providers. However, this work all represents low-value reproduction of plumbing already available from Twilio. Developers should be focused on building unique software solutions that deliver incremental value for the enterprise.

IDC recognized the same benefits of using Twilio in their CPaaS MarketScape report published in May 2021. Twilio occupied the Leader segment with the furthest position up and to the right. This reflected the broadest scope of capabilities and most complete strategy for Twilio as compared to competitive offerings. The fact that they are higher on both axes than any competitor is significant. Finally, they are attributed the largest market share (size of circle).

With the addition of Segment to the platform, Twilio will take the competitive comparison to a whole new level. In addition to covering the most channels and geographic regions, Segment allows Twilio go deep into the consumer experience. Prior to Segment, Twilio (and CPaaS competitors) were fairly limited in the amount of consumer tracking post communication that they could provide. While they could measure if a consumer opened an email or answered a call, they had no visibility into the consumer’s future activity on any of the enterprise’s digital channels. They couldn’t “close the loop” and measure the business outcome associated with any communication.

Segment, of course, changes all that. Once the integration is complete later this year, Twilio will be able to not just record the immediate consumer engagement with the communication itself, but also capture all follow-on consumer activity within the enterprise’s digital channels, whether their e-commerce site, call center, mobile app or chat. All this activity can be correlated and analyzed, allowing the Twilio/Segment platform to build a 360 degree profile of each consumer related to their engagement with the enterprise’s digital channels and the outcome of each communication.

This “closing of the loop” will significantly improve the effectiveness of Twilio’s communications platform. It underscores the evolution of their product marketing messaging to focus on the words “customer engagement”. That is because the combination of Twilio’s communications channels and Segment’s customer data platform (CDP) will enable the creation of a true “Engagement Cloud”. Plus, the recent industry shift for marketers to relying more heavily on first-party data provides a timely tailwind for adoption of Segment’s approach.

Going back to the DIY approach of stitching together point solutions, in order to collect the consumer engagement data across these different channels and providers will require the development team to gather event data from each provider’s communication and then correlate that to the activity data on their digital channels (assuming they have those sufficiently instrumented). This effort will quickly balloon into an enormous headache. The marketing team will want to know whether consumers are more likely to purchase from an SMS message or an email. Without the full lifecycle of measurement in place across all channels and countries, the development team will not be able to provide that simplest of insights.

Viewed across the three dimensions of channel coverage, global reach and depth of activity tracking, Twilio can claim the most complete solution. They address more communications channels than any CPaaS competitor, most of which are missing at least one or more channels. Twilio has integrations with sufficient carriers to distribute communications across every major country globally. Competitive CPaaS providers have traditionally been regional, available just in North America or Europe. The addition of Segment’s CDP has evolved Twilio’s platform along a third dimension of engagement tracking that isn’t even available from competitors.

For these reasons, Twilio is well positioned to continue to dominate the $62B CPaaS market. Following the Segment acquisition, Twilio leadership published updated estimates for their TAM as part of an Investor Deck in March 2021. Including the Customer Data Platform (CDP) market that Segment addresses, they estimated a TAM of $79B in 2020, expanding to $110B by 2023. These numbers are obviously getting very large. Few software companies can project a TAM over $100B.

Within this enormous TAM, Twilio is already the largest player by a large margin and continues to grow faster than (or as fast as) their smaller competitors. Some cloud vendors have stuck a toe into the space (like the Microsoft Azure Communications Services), but will not represent a meaningful threat due to less coverage across the three dimensions above. Additionally, enterprise technology organizations pursuing a multi-cloud strategy prefer to work with an independent provider where possible, diminishing the likely reach of any hyperscaler offering.

Bandwidth

Bandwidth (BAND) represents Twilio’s largest domestic competitor. Bandwidth experienced strong momentum and outperformance in 2020, as growth of their voice solutions accelerated due to heavy demand from UCaaS providers like Google, Microsoft, Cisco, Zoom, and RingCentral. This drove organic revenue growth of 48% in 2020, after logging 14% growth in 2019.

In October 2020, they announced the acquisition of Voxbone. The transaction closed in early November 2020. Voxbone’s financial performance was included in Bandwidth’s quarterly results starting in Q4 2020. Voxbone expanded Bandwidth’s offerings and reach in CPaaS, enabling the combined company to service 60+ countries representing 93% of global GDP. The primary benefit to Bandwidth was acceleration of their international expansion.

Bandwidth announced Q2 results on August 5th, a few days after Twilio. Breaking down the top-line numbers, we have the following:

• Total revenue of $120.7M, up 57% y/y from Q2 2020 revenue of $76.8M.  Q2 2021 included $26.9M of revenue from Voxbone.  Organic revenue was $93.8M, for growth of 22% y/y.
• CPaaS specific revenue was $105M, up 57% y/y, compared to $67.1M in Q2 2020.  Foxbone contributed $26M to this. Organic CPaaS specific revenue was $79M, which was up 17.7% y/y.
• Non-GAAP gross margin was 50%, up 2% from a year ago.
• Non-GAAP net income was $8.6M or $0.32 a share.
• Active CPaaS customers was 3,051, up 61% y/y from 1,900 in Q2 2020.  This includes the contribution from Voxbone.  At time of acquisition, Voxbone had about 900 customers.  Backing these out leaves 2,151 organic customers for a growth rate of about 13.2%.
• DBNRR was 114% in Q2 2021, compared to 133% in Q2 2020.  Voxbone results aren’t included in NRR.

As we can see, the rapid organic growth that Bandwidth experienced in 2020 does not appear to be sustaining and has decelerated into the low 20% range in 2021. DBNRR also slowed down almost 20% of growth year/year. Voxbone by itself doesn’t appear to be growing much faster. At the time of acquisition, Bandwidth announced that Voxbone’s 2020 revenue was expected to be over $85M, up more than 25% y/y.

These growth metrics are obviously much lower than Twilio’s on an organic basis across the board. Given that Twilio’s total organic revenue is over 6x larger than Bandwidth’s, and gross margins are roughly the same, Twilio will have much more gross profit to continue investing in go-to-market and R&D. At this point, I don’t see how Bandwidth can catch up. They may have some new product announcements lined up for later in the year and certainly may start to experience some synergies/acceleration once the two platforms are fully merged. Looking forward, however, we have to assume that Twilio will continue to lead the market relative to Bandwidth.

Zoom and Five9

In July, Zoom announced their intent to acquire Five9. Together, they plan to build the “customer engagement platform of the future.” The intent is to combine Five9’s leading cloud-based contact center solution with Zoom’s communications platform to improve how businesses connect with their customers. The combination of the two would provide the ability to address all communications channels (voice, SMS, video, email, chat, etc.) for engaging with consumers (at least in the U.S.) and offers a feature-complete contact center solution.

This will be a development to watch and would certainly fall into Twilio’s market sphere, although from a different angle. Zoom and Five9 have traditionally sold their services into the corporate IT department (CIO led) versus the engineering organization (CTO / VP Eng). Their solutions are primarily packaged for out-of-the-box utilization versus providing discrete communications building blocks to be consumed by developers in building custom applications. While Zoom and Five9 offer APIs for accessing communications channels programmatically, these have been primarily leveraged by other technology providers to create integrations, versus powering stand-alone applications.

I am not saying one approach is better than another – they are simply different. Twilio’s go-to-market is very much developer-focused, with a bias towards building a custom solution. Twilio’s product marketing is infused with the word “build” – a quick search of Twilio’s home page yields 12 instances of “build” or “building”. The home pages of Five9 and Zoom have 0. Twilio’s focus is to enable developers to build a custom solution with Twilio’s platform. From this perspective, the product positioning of the companies are different.

Even with their packaged contact center solution, Flex, the intent is to support significant customization by enterprises, versus being an all-in-one pluggable solution. This underscores Twilo’s product marketing position that enterprises need to build bespoke experiences for their customers, by starting with pre-packaged building blocks, and having the flexibility to snap them together and add custom code to create a customer engagement solution that is unique to that enterprise.

Given these different strategies, I think the two sets of companies can easily co-exist. Additionally, both Zoom and Five9 have had their products in market for several years, yet we haven’t really seen much competitive overlap between Twilio and these two providers to date. Even after Zoom and Five9 complete the acquisition and integrate their platforms, I don’t see this changing, at least for the next couple of years. Still, the combined company will be competing for the same enterprise spend over time, so it is worth monitoring.

Zoom has their annual Zoomtopia event coming up. This will provide more insight into their future plans. Additionally, we will get a sense for where the focus of their product development for the next year or two will be. For Zoom as a stand-alone company over the last year, it appears the focus has been on extensions to the core platform, like Zoom Phone, Rooms and Apps. These don’t really infringe on Twilio’s core market. The addition of Five9 would more closely align the Zoom platform with Twilio’s Flex offering. But, again, I think the difference will primarily be in the targeted audience for product sales.

Final Thoughts and Investment Plan

I think Twilio provides an opportunity for investors to own a leading provider of tools for programmable communications and customer engagement. Their Q2 report demonstrated that growth is not slowing, as enterprises continue to create new channels for interacting with their customers. This is all being driven by the migration of offline to online experiences and higher consumer expectations for convenience and real-time information. While COVID provided an impetus for this digital transformation, it doesn’t appear to represent a one-time bump as we move through 2021. I expect this growth to continue, at least representing a long tail of migrations and digitization projects that last several years.

In this demand environment, Twilio was already the largest and fastest-growing provider. This expansion was driven their breadth of channel coverage, global reach and strong developer-led sales motion. The acquisition and integration of Segment will further set Twilio apart from competitive solutions, as they close the loop on customer engagement, moving from single-shot interactions to driving multi-step consumer journeys towards tangible business outcomes. With their annual user event, Signal, coming up in October, we will get more insight into this exciting combination.

Twilio’s gross margin performance and operating leverage need to improve. Gross margin has been negatively impacted by rapid growth in low margin offerings, like international messaging. Operating expenses have been ballooning with multiple acquisitions and absorption of incremental headcount. Looking forward, the gross margin profile should benefit from relative growth in higher-margin software products (Flex, Segment, Frontline, etc.). The surge in operating expenses should temper as we move through acquisition integration. On the upside, the rapid increase in headcount, particularly the year/year doubling of go-to-market employees, should support continued sales growth.

As discussed earlier, my target for 2021 revenue is $2.859B, representing growth of 62%. I think this is reasonable, given that we have the first two quarters booked and can make fairly predictable estimates for the Q3 beat and Q4 sequential growth. Looking to 2022, I think Twilio will maintain a high rate of revenue growth. This is based on a few assumptions:

• Leadership has already told us that they will deliver at least 30% revenue growth each year through 2024. This is not an average over those years, but an absolute value greater than 30% in each year.
• They have not shown a slowdown in growth from 2020 to 2021, even as they push past a $2B run rate. This is important to appreciate for two reasons. First, Twilio doesn’t appear to have been a one-time COVID beneficiary with a pull-forward of revenue. Second, most software companies start to see growth decelerate after $1B in revenue (law of large numbers). Thus far, this boundary doesn’t seem to exist for Twilio, leaving us to wonder where it might be.
• Their penetration of the projected TAM is still low. Even as their revenue approaches $3B for 2021, Twilio estimates a TAM of $110B across all of their services by 2023. This has been increased multiple times, demonstrating that investors and analysts often underestimate the TAM for software products.

With that in mind, I am modeling 50% revenue growth for Twilio in 2022 for total revenue of $4.3B. This is far above analysts projections for about $3.5B representing growth of 30.4% over their current 2021 projection. I doubt Twilio’s growth will drop to that level, given the current momentum in the business. Looking back to the 2020 to 2021 transition, analysts initially modeled 2021 annual revenue growth to be 32.5%, coming out of 2020. As of September 1st, this estimate has been raised to 52.1% growth. At the end of 2020, the Segment acquisition was known, so that wouldn’t explain the inflation in growth estimates. Given that analysts have currently modeled 2022 for 30.4% growth, it is conceivable that a similar pattern repeats next year.

Coming out of 2020, we expected Twilio’s growth to temper. Yet, organic growth has ticked up in the first two quarters of this year. COVID provided the catalyst for businesses of all sizes to accelerate the migration of their offline experiences to digital channels. This migration will likely prove to offer a long tail of secular growth. Additionally, the limited use of programmable communications to date, arguably limited by abuse in voice and text, provides a long tail of increasing communications engagement. As communications with key vendors and lifestyle providers become trusted, consumers will tolerate much more frequent communication outreach.

Assuming my $4.3B revenue target for 2022 and the current market cap of $60.6B (close of September 9), the P/S valuation ratio for end of 2022 would drop to about 14. Twilio’s P/S ratio is currently at 25. This valuation metric has been over 25 for the past year and even hung around 20 for much of 2019. This implies that TWLO stock could gain 40% – 70% over the next 16 months. Improvement in gross margin and operating margin would push this higher, as Twilio’s valuation multiple is depressed compared to peers due to the marginal profitability. A very optimistic view would be that Twilio’s market cap doubles from here to the end of 2022.

On the pessemistic side, revenue growth in 2022 could slow down substantially, as the revenue becomes larger (law of large numbers). Annual growth of 50% is certainly unusual for a $3B run rate. Additionally, gross margins could deteriorate further if new high margin software offerings don’t gain traction and low margin messaging products continue to be the growth driver. If these occur, Twilio would likely “grow into” its current valuation, with the stock price continuing to flatline for the next 12 months. Given Twilio’s growth drivers and management’s own commitments, I would expect revenue growth at least stay above 30% and gross margins to remain in the current range, in the worst case.

Given these views, I think the risk/reward profile is appealing. I don’t see significant downside risk from the current valuation for TWLO and a favorable upside opportunity if current execution and demand conditions persist into 2022. Given that the stock currently trades around $340, a 50% gain to $510 by late 2022 is conceivable. The stock reached a high of $457 earlier in 2021, which provides further support that a new high just 15% above that is a reasonable assumption.

For this reason, I will maintain my current allocation of about 10-11% to TWLO in my personal portfolio. I may increase this if the price of TWLO drops further from here or continues to flatline through 2021. I have a cost basis of $144 with the majority of my holdings in a taxable account in the U.S. Therefore, any reduction to this allocation would incur a 20% long term capital gains tax. I think it is important to take capital gains into account when conducting re-allocation exercises, as a 20% reduction in a realized gain means the next investment has to return 25% just to break-even. Obviously, investors outside the U.S. or trading in a tax-free account (IRA, 401K) don’t need to take this into account.

NOTE: This article does not represent investment advice and is solely the author’s opinion for managing his own investment portfolio. Readers are expected to perform their own due diligence before making investment decisions. Please see the Disclaimer for more detail.

Additional Reading / Research

• I have covered Twilio since late 2019 on this blog. Interested readers can review my past coverage.
• The Colossus Business Breakdowns podcast conducted an analysis of Twilio in April 2021.
• A deep dive on Twilio published by Daniel Rubin on a popular stock investment board.
• Liberty publishes a weekly newsletter that provides digestible updates on many of the “builder” companies that this blog covers.
• Hhhypergrowth published a great overview of Application Development. This serves as a useful primer and includes many of the building blocks used by developers to construct modern applications. Twilio, along with many other companies covered on this blog, are discussed.