Investing analysis of the software companies that power next generation digital businesses

Alteryx (AYX) – Q2 Recap

Alteryx announced Q2 earnings on August 6th. They delivered a slight beat to revenue estimates and a large beat on EPS. However, Q3 and full year revenue estimates came in below analyst targets. Currently, the full year revenue growth target sits at 11%, down from 65% in 2019. On the day following the Q2 report, the stock dropped by 28%. On the earnings call, the leadership team discussed a few challenges during the quarter, primarily attributable to the macro environment and the associated slowdown in enterprise spend. Over the course of the quarter, Alteryx launched several new product offerings. In this post, I review Alteryx’s Q2 earnings, customer adds and the expansion of the platform. I also take a look at the bigger picture in analytics and how Alteryx is positioned in the emerging competitive landscape. For more background on Alteryx, readers can review my prior coverage.

Headline Financial Results

  • Q2 2020 Revenue was $96.2M, up 17% year/year. This compares to the consensus estimate for $94.1M, representing growth of about 14.7%. The company’s prior guidance from Q1 was for $91-95M in revenue for growth of 10-15%. For comparison, Q1 revenue growth was 43%.
  • Q2 Non-GAAP EPS was $0.02 versus ($0.14) expected, representing a beat of $0.16. This compares to ($0.10) in Q1 2020 and $0.01 in Q2 2019.
  • Q2 Non-GAAP operating loss was $0.1M, for an operating margin of -0.1%. This compares to operating income of $0.8M in Q2 2019 for an operating margin of 1%.
  • Q3 Revenue estimate of $111-115M, representing growth of 7-11% year/year. This compares to the consensus revenue estimate of $119.3M or 15.3% growth. This represents a reduction by about 6% at the midpoint.
  • Q3 Non-GAAP EPS estimate of $0.09 – $0.14. This compares to $0.13 expected.
  • Q3 estimated Non-GAAP operating income of $8 – 12M, for an operating margin of about 9% at the midpoint.
  • Alteryx updated revenue guidance in Q2 for the full year. They expect revenue of $460-465M in 2020, representing growth of 10-11%. This compares to the analyst estimate for $505M, or growth of 21%. In February, with their Q4 results, Alteryx had initiated FY2020 guidance at $555-565M for 33-35% growth. Obviously, that was before the COVID-19 situation manifested.
  • As of June 30, Alteryx had cash, cash equivalents, and short-term and long-term investments of $974M.
Alteryx Q2 2020 Investor Deck

Other Performance Indicators

  • Alteryx provided ARR metrics for Q2 and a projection for the full year. In the past, leadership had been resistant to issuing ARR data, as it represented another metric for the markets to scrutinize. However, with the variability of revenue growth under ASC 606, leadership feels that ARR provides a more clear picture of the business. For Q2, Alteryx stated that ARR was over $430M, up over 40% year/year. They also initiated full year guidance for ARR, estimating it will reach $500M by end of year, representing growth of 30%. This does provide helpful perspective, as 2020 revenue is projected to grow 10-11% currently.
  • Q2 Non-GAAP gross margin of 91% versus 91% in Q2 2019 and 91% in Q1 2020.
  • RPO at Q2 end was $410M, up 72% year-over-year, from $239M at end of Q2 2019.
  • Breaking down Q2 Non-GAAP expenses by category, we see a year/year rise in R&D expense. This is due to increased hiring in engineering and investment in product development. The company saw a little operating leverage in S&M, dropping by 3% of revenue year/year. G&A expense was roughly inline.
    • R&D = 21% (versus 18% in Q2 2019)
    • S&M = 52% (55% in Q2 2019)
    • G&A = 18% (17% in Q2 2019)
  • Ended Q2 with 1,515 employees, up from 1,478 employees at the end of Q1, and 1,076 employees at the end of Q2 2019. This represents year/year growth of 41% and 2.5% sequentially.
  • U.S. revenue was $66M in Q2, an increase of 14% year over year. International revenue was $30.2M, an increase of 25% year over year. Growth in North America was negatively impacted by lower expansion activity, specifically within enterprise and global strategic customer teams.
  • Overall, average contract duration for Q2 was approximately two years, consistent with prior periods.

Customer Activity

  • At end of Q2, Alteryx had a total of 6,714 customers, representing 27% growth from Q2 2019. They added 271 net new customers in the quarter, up from 6,443 at end of Q1, for sequential growth of 4.2%. This compares to Q1 customer growth metrics of 30% annually and 5.8% sequentially. Hence, Q2 experienced a slight slowdown in customer adds. But, customers are still landing, which is a positive given the macro environment and transition to remote sales.
  • Alteryx added 6 customers from the G2K in Q2 for a total of 737, bringing their G2K penetration to 37%. In Q1, Alteryx added 12 customers from the G2K and 36 new customers in Q4.
  • Overall Dollar-based Net Expansion Rate (DBNER) was 126% in Q2. This compares to 128% in Q1, 130% in Q4 and 133% in Q2 2019.
Alteryx Q2 2020 Investor Deck
  • On the earnings call, management mentioned that DBNER for Global 2000 customers was 137% in Q2, compared to 140% in Q1. Since G2K customers make up a large percentage of total revenue, a DBNER of this magnitude still implies high levels of future expansion for large customers.
Alteryx Q2 2020 Investor Deck

In the earnings release, Alteryx leadership highlighted several customer wins:

  • Latin America: Levi Strauss do Brasil Industria e Comercio Limited (Levi Strauss subsidiary in Brazil), Petrobas, Banco Santander and Omega
  • APJ: Tassal Group Limited, Samsung Biologics, 5G Japan, Toyota Systems Corporation and China Construction Bank
  • North America: Suncor Energy, Match Group, Snap, BlackLine and ServiceNow
  • EMEA: Bayer AG, Mondelez, Statistics Center Abu Dhabi, Merck Chemicals, Qatar Airways, Southern Water Services and L’Oréal.

The leadership team also detailed two specific customer expansions in the earnings call that shed further light on trends in sales activity. The first was with SiriusXM, and the experience their Tax team had with the Alteryx APA platform. SiriusXM began using Alteryx in Q3 of last year with adoption licenses to create efficiencies in compliance and provision processes, along with developing cash models for federal and state tax apportionment. The savings from these use cases justified an expansion in Q2 2020 to extend Alteryx to address use cases in accounting operations, with a focus on automation. The business case for APA was defined across 157 use cases with potential savings of more than 9,200 hours annually. This ability for the customer to use discounted adoption licenses to calculate the ROI for the full purchase is an important sales motion for Alteryx.

The second example is with the Saudi Arabia Ministry of Health for a COVID-19 use case. The ministry manages local government hospitals and medical activities. They were in a trial experience of Alteryx when COVID-19 started. With the Alteryx APA platform, they were able to automate the collection of COVID-19 test results from national labs and integrate those with actual hospital and quarantine rates. They then built network analysis workflows to understand the virus reproduction rate and track its spread. The Alteryx platform gave them insights into how, when and where to respond to COVID-19 cases, what facilities to close and how to allocate hospital bed space.  This has translated into an expansion deal.

Alteryx Q2 2020 Investor Deck

In the Q2 investor deck, management also provided other examples from actual customer use cases that demonstrate the ROI of the Alteryx product set. As we consider competitive alternatives, it is these types of outcomes that will continue to drive growth. The examples of data insights reflecting new revenue opportunities, cost savings and operational efficiencies highlighted in the above slide justify the ROI for the cost of the Alteryx product set. Because Alteryx provides solutions that makes it easy for customers to achieve these kinds of outcomes with little technical overhead, I think the demand for the type of analysis that Alteryx enables will persist. The key question will be how broadly that opportunity can expand within customer organizations, beyond the core data analysis team.

Adoption Licenses

In Q2, Alteryx highlighted a significant slowdown in new customer spend, particularly in impacted industries and with large customers. Expansion business sales cycles slowed. Customers sought to leverage their existing investments in Alteryx rather than undertake new projects and consequently, postponed or downsized new orders. In order to facilitate customer on-boarding, particularly where spending approvals were constrained, Alteryx leaned more heavily on the use of adoption licenses in Q2. Adoption licenses allow customers to utilize Alteryx products for a heavily discounted rate as part of a short term contract, generally six months. This provides the customer with the ability to “test drive” Alteryx products for a period long enough to prove the value before making an extended commitment.

This type of streamlined on-boarding behavior in Q2 wasn’t necessarily limited to Alteryx. As other examples, MongoDB talked about bringing on new customers to MongoDB Atlas with a minimum usage commitment. DocuSign didn’t push their full CLM suite for customers who just wanted to get eSignature in place. With that said, Alteryx took a larger revenue hit from this behavior as these trial licenses contribute little to revenue and the impact is exacerbated by Alteryx’s revenue recognition structure for new business.

On the earnings call, Alteryx leadership stated that there was a 60% year/year growth in the use of adoption licenses in Q2 and a 100% increase from Q1 to Q2. They also clarified that adoption licenses are applied to large customers looking to enable a major digital transformation who can’t immediately size their anticipated usage – whether 200 or 2,000 users for example. With tighter spending budgets and likely more scrutiny around ROI analysis, it is understandable that CXOs would lean on the adoption licenses to help them make the ROI case. Keeping in mind the customer examples from the investor deck, it is easy to assume that if a CIO or CFO can see the ROI potential from using the Alteryx toolset, then the investment would be easier to justify. Typically, these kinds of conversations for new spending justification are framed something like “I want to spend $200k on new Alteryx licenses, because I know the operational efficiency gained will drive $5M in savings.”

Alteryx expects that at the end of the adoption license period, a large portion of the free licenses will convert into paid. They stated these tend to “lead to much bigger expansion.” If this spike of adoption license usage occurred in Q2 (and expecting more in Q3), then we could see a surge in revenue hit in Q4 and Q1 2021, as the average 6 month duration lapses and some portion of adoption licenses convert to full price. This will be an important factor for investors to watch.

As a real-world example of how this strategy worked, leadership highlighted the approach in their testimonial with SiriusXM. Their tax team started using adoption licenses in Q3 of last year. Those presumably lapsed in Q1 of this year, leading to a large expansion for Q2, as SeriusXM then committed to some portion of fully paid licenses. In a webinar in July, Neil Leibowitz, Vice President of Tax at SiriusXM, discussed their journey with Alteryx over the prior 3 quarters. He said that early use cases drove hundreds of hours of savings, which led to a third expansion planned for Q2 to support shared services and accounting operations. However, COVID has forced them to think more critically through a financial lens to justify technology investments. By being able to demonstrate the savings in analyst hours from using Alteryx tools over existing solutions, it was easier for him to make the ROI case to the CFO.

Also, we should keep in mind that use of adoption licenses aren’t “free” to the customer. There is also the cost of committing resources to learn and use the new toolset. This commitment of mindshare and quick wins generally creates an inherent momentum towards paid license adoption.

Partner Program

As investors will recall, Alteryx announced a strategic relationship with PwC in February. This represents a five-year relationship, which includes the designation of the firm as a “Global Elite Partner.”  As part of this, PwC will advise their clients in establishing strategy and governance around their data automation programs. This will include building automation solutions on the Alteryx Platform. Alteryx leadership expects the relationship to generate $1B in revenue for Alteryx over the next 5 years. On the Q2 call, Alteryx also mentioned that PwC is a major user of Alteryx solutions themselves, with over 55k users.

On the heels of announcing their relationship with PwC, in Q2, Alteryx broadened the partner network by adding two more technology partners in Adobe and UiPath. These partners will also promote Alteryx products as part of the broader solutions they craft for their customers. Alteryx leadership believes that having an open and vibrant ecosystem is an important element to bringing their analytic process automation vision to life. 

Go To Market

I think Alteryx was caught a little flat-footed by the rapid transformation to remote work due to the COVID-19 situation. My impression is that their workforce is more organized around physical proximity than some of the newer technology companies. This is likely an outcome of their founding in the late 90’s and headquarters outside of San Francisco, where office space isn’t quite as constrained. Also, their sales motion seemed more aligned towards in-person events and sales pitches. As a result, the transition to remote workforce and virtual sales meetings took a little longer for Alteryx than for other companies.

The Alteryx leadership team acknowledged needing to make adjustments to the sales and marketing process. Their primary investment area has been expanding their global go-to-market footprint. They are focusing on improving sales rep training and enablement, re-evaluating partner engagement, redeploying sales resources to focus on higher-value opportunities and adjusting marketing spend to be more digitally focused.

On the earnings call, Alteryx leadership reported a 68% growth in monthly active users of their Community site, and a 82% increase in engagement. In Q2, they launched a Virtual Solutions Center, where 1,000 customers continue to advance their learning and understanding of the APA platform. They conducted 235 virtual webinars, including a live stream for the APA event. Alteryx has also iterated a number of processes in the sales organization. They have invested in new programs for the Community offering, that take the place of on-site working days and in-person customer workshops. Alteryx leadership thinks these efforts will lead to more expansion and sales efficiency, once the COVID situation clears.

Analyst Reactions

Following Alteryx’s Q2 earnings results, 5 analysts provided updated coverage ratings. Of these updated ratings, all analysts lowered their price targets. Four analysts rated the stock at a Buy equivalent and one gave a Neutral rating. The average price target for these updates is $171, representing a 41% increase from the closing price after earnings of over $121 on August 7th.

DateAnalystRatingPrice Target
8/7OppenheimerOutperformLowered from $190 to $180
8/7GuggenheimNeutralLowered from $130 to $125
8/7NeedhamBuyLowered from $192 to $172
8/7Piper SandlerOverweightLowered from $195 to $185
8/7Goldman SachsBuyLowered from $216 to $195
Ratings Assembled from MarketBeat, YCharts

After the earnings results, Piper Sandler lowered their price target to $185 and maintained an overweight rating. Analyst Brent Bracelin provided this commentary.

Piper Sandler analyst Brent Bracelin lowered the firm’s price target on Alteryx to $185 from $195 and keeps an Overweight rating on the shares. The analyst reduced estimates to reflect higher than expected headwinds in the quarter that slowed existing customer expansion metrics and large deal uncertainty that he believes could further reduce second half revenue growth.

TheFly, Aug 7, 2020

Product Development Activity

Alteryx announced several major improvements to their platform over the course of the last several months. As investors may recall, Alteryx CEO, Dean Stoecker, talked about a robust product roadmap for 2020 at the JP Morgan Conference on May 12th. He said there would be more product innovation for Alteryx in the next four quarters than there has been in the last 10 years. He teased a few examples like a browser-based version of Designer and an intelligence suite. Another focus would be around lowering the bar in harnessing advanced data science capabilities for users. In another interview of the CFO by TheStreet, the CFO reinforced that M&A will continue to be part of Alteryx’s product strategy to selectively pick up strong development teams and unique IP.

On May 11, Alteryx announced the new Analytic Process Automation platform (APA). The intent is to unify analytics, data science and business process automation into one platform. Alteryx contends that current analytics and data science work is addressed through a disparate set of point solutions. They want to provide a platform that allows data workers to prosecute the full spectrum of tasks in one system.

Alteryx Q2 2020 Investor Deck

Beyond the technology consolidation, their approach is also very people centric. It is positioned to make analytics and automation available to a broad audience, essentially any employee with a need to better understand their company’s data. Alteryx primarily sells into the data analyst teams within business units versus centralized IT. Part of Alteryx’s intention is to expand the user set, by making data processing tasks easier through their “code free” approach. This is bolstered by the theme of empowering citizen data scientists.

The Alteryx APA platform provides an integrated solution that addresses all steps in the analytic process. It includes 260 automation building blocks that can be combined through a visual interface to create an analytic pipeline. These range from facilitating data import to cleansing, generating insights and exporting to external systems. APA enables five phases of the analytic process:

  • Automate data asset inputs. This allows data from many sources to be pulled into the pipeline for processing. The platform includes connectors for 80+ common data sources, like Salesforce, Oracle or MongoDB. These can be in the cloud, on-prem or even represent files on the user’s local machine like CSV, PDF, etc.
  • Data quality and preparation. Provides tools to cleanse, prepare, and blend the data. Both structured and unstructured data can be handled. The goal is to normalize the data across the different sources, adjust for errors or missing fields and combine it all into a single data set.
  • Data enrichment and insights. This layers in data from outside sources to complete the view. Users can add geospatial context or pull in data from third parties, like demographics. With this, the user can manipulate their view of the data to generate business insights.
  • Data Science and Decisions. To facilitate processing the data set to support predictive outcomes and repetitive querying, the system allows the user to build complex analytical models. For data workers without coding skills in Python or R, assisted modeling tools and step-by-step guides are available. Data scientists can still load their models or tweak the code generated by the systems automated tools. Machine learning and AI are injected into the process as well.
  • Automating Outcomes. This involves sharing the output and insights gained from the process. The sharing can be accomplished by writing back to a permanent data store, generating a spreadsheet or publishing a report. Once an automated process is built, it can be repeated to update the insight output on a scheduled basis. Also, APA supports lightweight dashboards for quick views, but is meant to export to more robust data visualization tools, like Tableau.

ML/AI capabilities will be layered into the data processing flows and leverages Alteryx’s October 2019 acquisition of Feature Labs, a data science company launched out of MIT. Feature Labs brought capabilities around automating the creation of machine learning models with a focus on feature engineering. These tasks are normally handled through manual, error-prone processes that rely heavily on skilled data scientists. That acquisition also lended Alteryx some street cred, as Feature Labs maintains a number of open source libraries for data scientists to automate feature engineering tasks. At time of acquisition, these open source libraries had been downloaded over 350,000 times and Alteryx plans to continue support for this work.

The APA launch was supported by a media blitz, including coverage in ForbesGartnerFast Company and Wired. Additionally Alteryx hosted a kick-off event that was live streamed on May 20th. Participants beyond Alteryx personnel were the Chief Product Officer at PwC (Alteryx’s global elite partner), the Research Director of Analytics at IDC and the VP Decision Science at Coca-Cola (a large Alteryx customer). Also mentioned in the release is a new customer, the Al-Futtaim Group, an operator of over 160 franchise brands across retail, financial services, automotive and health across the Middle East and Asia.

Adding to this, Alteryx announced a significant improvement to the core data processing engine in early June, dubbed the AMP engine. They added multi-threading and a method of breaking up in-coming data loads into packets that are processed in parallel. This results in data transformation jobs finishing several times faster than previously. This approach is similar to the MapReduce method used by other big data systems, like Hadoop. What is important for investors to consider is that this allows for distributed data processing (spread across multiple machines) and could provide a precursor to a future cloud-based, multi-tenant offering.

Alteryx AMP Engine Blog Post

Following APA, Alteryx later announced two new product offerings on June 16th. These were the Analytics Hub and the Intelligence Suite. The Analytics Hub greatly expands the user base for Alteryx products, by introducing new user roles that go beyond just data analysts with a Designer license. These users interact with the hub through a browser interface built on a modern Javascript-driven user interface. The open architecture also includes API interfaces to control all functionality. Data processing jobs run on 1-10 worker nodes which make use of the new multi-threaded processing engine (AMP) that allows large jobs to be broken up and run in parallel. All of these architectural constructs would support a cloud-based delivery model, if that becomes a future Alteryx product direction. The Intelligence Suite represents an add-on module for Designer. It enables assisted modeling to allow less technical users to create machine-learning predictive models through a wizard, without requiring code. The suite also includes sophisticated text mining capabilities to pull context from images, documents or customer sentiment. Both of these new products have separate pricing, which should drive incremental revenue. The Analytics Hub in particular seems to support a new user licensing model that would extend far into the enterprise organization, touching any knowledge worker with the need to view the output of an analytics job.

Feedback from users on these new products has been positive. Early customers include Siemens Gas and Power, Bell Canada, QDOBA Restaurant Corporation, Mars Incorporated, General Dynamics and the U.S. Navy’s Naval Research Lab. In future releases, Alteryx intends to push more integration with some of the capabilities of Connect and Promote, enable auto-modeling capabilities and launch a marketplace for third parties to monetize assets to the growing audience of Alteryx users around the world.

On the Q2 earnings call, leadership gave an update on recent M&A activity. Related to the last acquisition made, Feature Labs in October 2019, Alteryx has woven in their EvalML and the Python open source engines into their assisted modeling product. This is part of the predictive server solution that provides auto-modeling capabilities. Leadership said that a new product offering will compete with competitive products from DataRobot and H2O.ai starting next year.

Regarding the ongoing M&A activity, the CEO contends the the COVID-19 situation has put pressure on a lot of smaller entrants into the space who are still building their business and trying to find a good product/market fit. Alteryx is examining a lot of opportunities to purchase IP to plug into their broader platform, like they did with Feature Labs. He also mentioned that they see opportunities to acquire meaningful revenue streams for product offerings that are in proximity to their space. That will be something for investors to watch, as it could boost revenue inorganically.

Competitive Market Activity

Alteryx Q2 2020 Investor Deck

I won’t perform an exhaustive comparison of Alteryx’s solution to every vendor in the diagram above. For the most part, Alteryx’s position relative to these other vendors hasn’t changed in each category. Alteryx still enjoys a somewhat unique position as a popular tool for data analysts to dramatically simplify and lighten their workloads, as an alternative to performing exhaustive Excel VLOOKUP functions. Many of the customer value drivers for the purchase of Alteryx revolve around time savings for analysts, so that they can focus on surfacing insights and value creation. This positioning is still reinforced in every investor deck on the “Analytic Waste” slide, basically contending that analysts waste $60B a year doing repetitive manual work in spreadsheets.

As we saw with the SiriusXM example earlier, the customer was able to identify 9,200 hours of time savings for analysts in the Tax department by automating many of the data preparation and analysis tasks using the Alteryx toolset. For other examples, Alteryx maintains an active Community site for its customers. This includes the ability for them to post sample use cases of how they apply the Alteryx toolset within their organizations. Reviewing these use cases, we see many more examples of companies realizing time savings through this type of automation:

  • North American Airline. Using Alteryx, the Analytics team at a North American global airline was able to modernize its audit program, creating an automated solution that saved over $1M annually and reduced the audit process from 5-7 days to 15 minutes. The airline was able to streamline different manual processes involving SQL code, Access databases, Excel spreadsheets and other 3rd party software into four Alteryx workflows that can be run by any team member on Alteryx Server.
  • Publicis Groupe. Applied the Alteryx toolset to client report generation. Replaces manual processes of querying data from disparate sources, cleansing and running transformations. Also replaced an Excel-based system to generate client campaign data, involving the merge of 86 different Excel files. The result was significant time savings and faster repeatable run times.
  • Coca-Cola. Alteryx is used heavily within the analytics group at Coca-Cola. In one case, Alteryx was leveraged when Coca-Cola wanted to replace fountain machines for one of its largest customers with over 24,000 franchise restaurants across the US. With Alteryx, Coca-Cola regularly creates 700 personalized PDF reports files within 5 minutes to provide restaurant owners with insights about their sales and beverage usage, to help optimize inventory usage and reduce stock depletions. Previously data was generated manually using Excel.

As investors can appreciate, these customer use cases are full of examples of the benefits of the Alteryx toolset. These seem to orient around the primary use case of time savings in data prep and analysis tasks. Similarly, the customer set represents many traditional enterprises in mainstream businesses, as reinforced by Alteryx’s penetration into the G2K.

Within the framework of the new APA platform, Alteryx is attempting to expand the set of use cases into other analytical workflows within the enterprise. They are effectively pointing out to their customers that many more opportunities exist to apply Alteryx data analysis flows to other common business problems in resource optimization, forecasting, scheduling and pricing.

Alteryx Q2 2020 Investor Deck

With the PwC partnership (and others to come), it is easy to see how new applications for Alteryx will continue to surface. As the PwC Alteryx practice leaders engage with their existing clients, they will suggest uses of Alteryx products. They will recycle best practices and outcomes from other clients to generate maximum expansion of usage. This partner-enabled sales motion, particularly from a thought-leader like PwC, should drive revenue growth in the G2K.

I anticipate that once COVID-19 headwinds abate, Alteryx will be able to return to a predictable path of expansion within existing mainstream customers and incremental penetration in the G2K. Yes, there are competitive alternatives, but these don’t seem to have impacted Alteryx’s growth in a meaningful way, up until COVID hit. The current sales slowdown is primarily a consequence of economic headwinds and the position that Alteryx’s product valuation proposition occupies. It is a software tool that requires an investment in order to automate an existing process. That in turn generates savings or a new business opportunity to justify its cost. However, in the COVID-19 environment, that kind of investment will be rightfully de-prioritized by enterprises scrambling to adjust to the new operating constraints.

This doesn’t mean Alteryx tools don’t have value to an organization. It just reflects the urgency of IT spending needs that all CXOs are facing. To put a prioritization of IT services during COVID-19 in simple terms, we can assume something like:

  1. Allow employees to work remotely.
  2. Secure access for those employees and their enterprise assets.
  3. Move customer experiences online.
  4. Generate efficiencies and find other revenue opportunities.

Analytics tools that fall into the fourth bucket, particularly those that replace a working, but inefficient, process can be put off by a quarter or two. Once enterprises address priorities 1-3 above, there likely will be a return of investment in efficiency and optimization initiatives. We might even see a surge of pent-up spend initially after COVID-19 clears and then a long tail of expands within existing customers and the remainder of the G2K. This could lead to a spike in revenue in 2021 and then recurring revenue growth annually for Alteryx in the 25-35% range for many years. With high gross margins, Alteryx will be able to continue investing in R&D and go-to-market efforts to extend the product and improve sales execution.

There are however a few trends that we should keep an eye upon, which could limit the Alteryx growth thesis over the middle to long term (let’s say 3-5 years). This represents a competitive threat because I don’t think Alteryx is currently well-positioned to adjust for it. This doesn’t take the form of a competitor duplicating the same type of toolset as Alteryx. Rather, it represents a fundamental shift in how enterprise data will be stored, processed and shared in the future. I think a new ecosystem is forming around the centralization and processing of large data sets in the cloud.

Historically, enterprises kept their critical business data (customer, finance, sales, operations, etc.) in on-premise data silos. This was primarily a consequence of the applications that manage that data being hosted on a company’s local network. However, with the rise of SaaS and cloud-based infrastructure, this posture is shifting.

The cost, security and governance concerns for moving enterprise data to cloud storage are continuing to be addressed. Cloud adoption in general by enterprises has accelerated as a result of COVID-19, as more companies seek the economies of scale and resiliency that can be gained from cloud infrastructure. IDC recently highlighted this trend in their Cloud IT Infrastructure Spending Report published in June, 2020. Cloud IT infrastructure continued to grow in Q1 of 2020, while non-cloud environments experienced significant spending declines. While data storage is a sub-category within overall cloud spend, it is following the same trajectory (and logically would, as proximity between compute and storage is necessary).

Vendor revenue from sales of IT infrastructure products (server, enterprise storage, and Ethernet switch) for cloud environments, including public and private cloud, increased 2.2% in the first quarter of 2020 (1Q20) while investments in traditional, non-cloud, infrastructure plunged 16.3% year over year.

The broadening impact of the COVID-19 pandemic was the major factor driving infrastructure spending in the first quarter. Widespread lockdowns across the world and staged reopening of economies triggered increased demand for cloud-based consumer and business services driving additional demand for server, storage, and networking infrastructure utilized by cloud service provider data centers. As a result, public cloud was the only deployment segment escaping year-over-year declines in 1Q20 reaching $10.1 billion in spend on IT infrastructure at 6.4% year-over-year growth. Spending on private cloud infrastructure declined 6.3% year over year in 1Q to $4.4 billion.

IDC expects that the pace set in the first quarter will continue through rest of the year as cloud adoption continues to get an additional boost driven by demand for more efficient and resilient infrastructure deployment. For the full year, investments in cloud IT infrastructure will surpass spending on non-cloud infrastructure and reach $69.5 billion or 54.2% of the overall IT infrastructure spend. Spending on private cloud infrastructure is expected to recover during the year and will compensate for the first quarter declines leading to 1.1% growth for the full year. Spending on public cloud infrastructure will grow 5.7% and will reach $47.7 billion representing 68.6% of the total cloud infrastructure spend.

Within cloud deployment environments, compute platforms will remain the largest category of spending on cloud IT infrastructure at $36.2 billion while storage platforms will be fastest growing segment with spending increasing 8.1% to $24.9 billion.

IDC Cloud IT INfrastructure Spending REport, JUne 2020

This report underscores the migration of corporate data from on-premise storage solutions to private cloud and ultimately public cloud. Alteryx leadership maintains that most of their customers have a substantial portion of their data on-premise or in private data centers. They use this rationale as a defense for not moving quickly to product offerings that are cloud-based. Fundamentally, their product architecture utilizes a traditional client-server architecture, where the customer is responsible for managing the Alteryx software installation on both ends. As a further nuance, the Server software can be deployed by the customer onto a Windows server hosted in one of the public clouds and many Alteryx customers choose to do this. This gives the appearance of being “cloud” enabled, but isn’t really what we mean by multi-tenant SaaS.

The Alteryx CEO provided more color on the earnings call around this in response to an analyst question about Alteryx’s plans to move to a “cloud” offering.

So everyone knows, we actually have quite a few customers who are in the cloud. We have our server deployments for automation and analyst processing up in AWS and Azure for as little as $9 an hour that you can execute with just a few key strokes. We actually have that up there for three or four years with almost no activity. And that’s not to suggest that cloud isn’t important, but we’re hybrid. We understand that we want to be close to where the data lives. And in large organizations, especially in the Global 2000, most of their data hasn’t moved. In fact, earlier this week, I was on the phone with a Chief Data Officer of a Fortune 50 insurance company on the East Coast, and he indicated that they’ve been dabbling with the cloud for quite some time, but not a single bit of customer data was currently in the cloud. And so I think that we’ve been focused on cloud for a very long time.

We are in the process of a cloud-based Designer, mostly to ease the burden of deployment of large implementations in organizations around the world. We have — if you attended our APA event, you would have heard that PwC indicated that they have 55,000 users of Alteryx. Now it’s a hassle to deploy quarterly releases of an image for those users, so having a browser-based delivery would be better. The customers are not pushing us for a multi-tenant SaaS service, at least not today. The data is living everywhere. It’s going to be hybrid forever. And we’ll live where the customer tells us to live, and we’ll be prepared when the customer says that the data gravity has shifted. It just has not shifted. And I don’t think it’s going to shift for quite some time.

Alteryx CEO, Q2 2020 EArnings CAll

So, as the Alteryx CEO rightfully points out, the Alteryx server could be deployed in the cloud, but this would be a self-managed, single-tenant deployment model. This is different from the typical SaaS/PaaS deployment model, which is multi-tenant and vendor-managed. The transition from the former to the latter is not trivial, but is achievable with planning, determination and focus. However, Alteryx risks falling into an innovator’s dilemma scenario, where their existing customers create inertia against their migration to the next technology paradigm.

As I discussed in my Q1 review of Alteryx, they do seem to be preparing the technology organization for this transition. In early May, they began posting new jobs for software engineers that utilize modern software practices and development frameworks. The architecture description was cloud-focused, making use of browser-based UI on the front-end with scalable APIs and micro-services on the back-end. This does provide a sign that Alteryx is investing in modernizing their tech stack with an eye towards addressing some of the architectural limitations to bring their platform more inline with typical SaaS offerings. 

However, even if Alteryx migrates their data analytics platform to a multi-tenant, cloud-managed solution, the more recent emergence of centralized data clouds could diminish the importance of their offering. This trend is best encapsulated in the rapid growth of Snowflake (SNOW), which most investors are likely familiar due to their recent IPO. Snowflake describes themselves as a data cloud and goes beyond the traditional notion of a data warehouse.

Snowflake has leveraged the recent inflection in infrastructure costs, technology capabilities and security controls to make the aggregation of all data types into a single data store for an enterprise possible. Previously, this goal was prohibitively expensive and insecure. Disparate data types made ETL processes complicated and difficult to maintain. However, reductions in cost for cloud-based data storage and compute, as well as improvements in user authentication and authorization, have made a central store possible. In addition, Snowflake’s own data architecture design supports an inherent versatility of data types, from structured to unstructured. They facilitate data ingest from many sources through connectors, removing the overhead for customers to manage data loading and transformation.

Prior to these advancements, enterprises kept disparate data types in data silos, separating data for customers, finances, operations, human resources, etc. in different systems. Whether these were on-prem IT systems or SaaS, it didn’t matter, they were still separate. Because of this separation, data analysts needed tools that allowed them to pull data from these various systems, cleanse, filter, aggregate, and analyze it in order to generate insights (cost savings, business opportunities, reports, optimizations, etc.). This toolset is provided by Alteryx.

The Designer client application has connectors to hundreds of data sources. Once loaded, data can be cleansed, prepped and blended. Users can explore the combined data set and create visualizations for it. Or, they can run more advanced analytic jobs, including text mining and machine learning. Workflows can be stored on the Alteryx Server and scheduled for periodic runs. Insights and reports can be shared with other analysts and business users through the Analytics Hub.

Snowflake, and other cloud-based data platforms (I’m sure there will be others), provides the ability to converge all data silos into a single data cloud for an enterprise. While the system is multi-tenant, Snowflake applies strict controls to ensure data isn’t shared between customer instances and provides customers with extensive user authorization capabilities to manage data access both for internal employees and external partners. This can be done for a reasonable cost and delivers the flexibility to manage customer usage along the dual axes of storage and compute, which were not possible to separate in previous instances of data warehouses.

These advances finally enable enterprise data convergence. As Snowflake implores customers on their web site “Unify, integrate, analyze, and share previously siloed data in secure, governed, and compliant ways. Achieve all of these benefits via a single and seamless data experience that unites multiple clouds and their geographic regions.”

If we dig into the Snowflake web site on the Developer section, we find a sub-section dedicated to Machine Learning and Data Science. Snowflake offers the ability to “Quickly and cost-effectively extract the deepest insights from all your data to best serve your customers and reveal new market opportunities.” They provide a reference architecture for enabling this capability on a diagram that looks like the following.

Snowflake Web Site – Developer Use Case for Machine Learning

In the diagram, we see several areas that appear to overlap with Alteryx product offerings. Within the Snowflake symbolic runtime (the dashed blue box), they offer data transformation and cleansing tasks, scheduling and automation, data cloning and feature engineering. Additionally, model training and machine learning are facilitated with integrations to external packages and third-party software libraries (the gray box to the right). These services are provided by Snowflake Technology Partners, which include ML/AI tools (Dataiku, DataRobot, H20.ai). Other technology partners are the ETL providers (Informatica, Talend, Fivetran, Mattillion). As you will note, Alteryx doesn’t appear on this diagram and is not currently listed as a technology partner.

Alteryx does have a connector for Snowflake, as well as all the public cloud data warehousing solutions (Redshift, BigQuery, Azure Synapse), which facilitates data transfers and the ability to process workloads within Snowflake itself. Alteryx and Snowflake also share hundreds of customers in common and actively cooperate in sales deals. During the recent Citi Technology Conference, the Alteryx CEO discussed how even Alteryx themselves utilizes Snowflake as their data warehouse to replicate data from their various SaaS applications (Workday, Netsuite, Salesforce, etc) into Snowflake. They deploy Alteryx Server beside Snowflake and perform all their line of business analytics through the Alteryx toolset. Some of this analysis is done in memory on the Alteryx Server and other workloads are run on Snowflake itself.

So, in this regard, Snowflake isn’t a competitor to Alteryx per se, as some analysts keep asking. However, I think the right question is whether the ecosystem that Snowflake enables creates alternatives to Alteryx. That is the key consideration going forward. If enterprises continue to converge their data into a single instance on data cloud platforms, like Snowflake or even Databricks, does that supersede the position that Alteryx currently enjoys (which assumes enterprise data is in silos)?

Alteryx Q2 2020 Investor Deck

Going back to the Alteryx Investor Deck slide at the beginning of this section, Alteryx argues that their value is in providing the full platform of services spanning data catalog, prep, visualization, diagnostic, predictive and prescriptive analytics. However, Snowflake and its ecosystem of technology partners, could make the same argument. Even with APA, the diagram above appears similar at a high level to the same data flow diagrams of the Snowflake ecosystem.

My hope is that Alteryx evolves its product offering quickly to position itself against this risk. They should become a formal Technology Partner with Snowflake, so that customers gain more visibility into running ML/AI workloads on the Snowflake platform. They could also consider taking a similar posture with other emerging data cloud platforms and the large data warehouse solutions. I think the risk is that Alteryx leadership wants the APA platform to be viewed as the “glue” between disparate systems, when really it should evolve into a powerful data insight extraction engine that layers on top of a data cloud.

More broadly, Alteryx should acknowledge that the migration of enterprise data to the cloud is accelerating. The recent job postings and leadership comments about a rapid product development roadmap are encouraging signs that Alteryx has a strategy to address this. At this point, it is too early to draw definitive conclusions about direct impact of centralized data clouds on Alteryx’s business, but is something that we investors should monitor closely.

Alteryx Take-aways

Q2 was a tough quarter for Alteryx. On the surface, the revenue deceleration is concerning, dropping from 43% annualized growth in Q1 to 17% in Q2 and a projection for 7-11% in Q3. This deceleration has been exacerbated by Alteryx’s revenue recognition method, where new deals make an oversized contribution to revenue in the quarter in which it was booked, as a consequence of ASC 606.

In order to provide investors with an alternate way to measure business growth, leadership will report on ARR going forward. For Q2, ARR was over 40% year/year. Alteryx also issued full year guidance for ARR, estimating it will reach $500M by end of year, representing growth of 30%.

If you remember, we’ve spoken for quite some time, our revenue mechanics and ARR are disconnected. Revenue is driven by bookings, which is TCV and an upfront portion based on product mix, and ARR is really just the accumulation of ACV over time. So the two are very disconnected in that regard.

Alteryx Q2 Earnings Call, August 2020

Leadership discussed on the earnings calls how large organizations paused major digital transformation efforts, scrutinized spending and didn’t engage in the same level of expansions at renewal time. Some of this commentary sounds similar to other software companies reporting in Q2 that large customers slowed down spending, optimized their contract usage or lengthened sales cycles with more approvals. Examples included Datadog, Elastic, MongoDB and others.

We observed notable changes, such as higher levels of scrutiny on spending across all sectors, resulting in longer sales cycles, smaller deal sizes and less favorable linearity in the quarter. Based on what we see today, we do not anticipate a material improvement in business conditions during 2020. At the same time, we believe that COVID is creating a longer-term tailwind for our business. Companies that lacked analytic rigor or those with data challenges sought out a quick ROI solution to help them adapt to rapidly changing business conditions.

Many found their answer with Alteryx. We believe this dynamic provided a tailwind for new business during Q2 as we saw solid land activity in high-risk verticals, such as transportation, accommodations, food service and retail. We believe this illustrates that data and analytic capabilities are important, particularly in challenging times, although initial deal sizes were slightly smaller than they have been historically. The industries and functional use cases that Alteryx addresses continue to be quite broad.

ALTERYX Q2 EARNINGS CALL, AUGUST 2020

This impact on Q2 is understandable given the macro environment. However, the big question for Alteryx is how long this headwind will persist. In their case, leadership has indicated that it will impact growth through all of 2020. On the other hand, Datadog highlighted an uptick in large customer spending growth starting in July. Granted, these two vendors address different segments of enterprise software. It could be that enterprise IT spending will restart in waves. Services tied to software delivery, like infrastructure monitoring, would likely be prioritized earlier in the recovery cycle than analytics tools. This makes sense from the perspective of prioritizing based on risk, versus benefit. CXO’s would initially turn spend back up on items that minimize business risk (security, uptime, etc.) before they would invest in items that help grow the business, like analytical insights.

Customer growth for Alteryx also slowed in Q2. Total customer growth has generally been consistent quarter/quarter for several years. Yet, in Q2, total new customer adds slowed to 27% annualized growth, compared to 30% in the prior quarter. The same trend occurred with G2K customer adds, lowering to 6 for Q2, versus 12 in Q1 and 36 in Q4. DBNER dropped as well from a high of 133% a year ago to 126% in Q2. The key question will be whether and when the customer growth rate normalizes.

Assuming the business slowdown is just caused by COVID-19, then we should see growth resume once the COVID-19 situation improves. This would argue for a nice set-up going into 2021 for Alteryx. The backlog of investment in analytical tooling would surge at once, driving high growth, particularly after easy comps in 2020. Revenue growth would disproportionally benefit due to the revenue recognition methods. Additionally, the heavy use of discounted adoption licenses in Q2 will deliver further stimulus six months from now. As they expire, we can expect a high percentage of adoption licenses to translate into fully paid licenses.

The risk to the long-term revenue growth thesis for Alteryx is that some other factors emerge that dampen the need for their tooling. I think this is where competitive motions and the evolution of the analytics market could come into play. Next-generation data cloud platforms like Snowflake finally make full data convergence for enterprises feasible. These cloud platforms enable direct ingest of data from disparate sources and house toolsets for data cleansing, prep and aggregation. Additionally, they provide an ecosystem of third-party tooling to enable machine learning, automated training and visualizations. This removal of data silos and ecosystem of tooling risks rendering the Alteryx platform irrelevant. At minimum, as enterprises migrate data onto these platforms, they could siphon off some use cases that might have represented enterprise up-sells for Alteryx.

I am looking forward to more tangible evidence that Alteryx is rapidly evolving their product offering, hopefully to a cloud-centric delivery posture. The assumption that this type of effort is underway is based on cloud-oriented software engineer job descriptions from earlier this year and commentary from leadership on analyst calls referring to a robust product development pipeline.

Alteryx Q2 2020 Investor Deck

The profitability picture and operational leverage for Alteryx continue to be favorable. Non-GAAP gross margins are still at an eye-popping 91%. Operating margin was soft in Q2, but is projected to reach 9% or more in Q3. A long term target of 30-35% for FCF margin would support a higher than average multiple on the stock price.

Personal Investment Plan

I had initiated coverage on AYX in December 2019 when the stock was trading at $97.40 and set a 5-year price target of $325, which would land in 2024. I think this target is still realistic, considering AYX reached an ATH of $185 prior to Q2 earnings. I will keep the price target in place for now, and will revisit it in early 2021, once we have visibility into revenue performance following the COVID-19 situation and Alteryx’s evolving product positioning.

Prior to Q2 results, I had trimmed my position in AYX down to about 5% of my personal portfolio. Given the headwinds to sales raised by management coming out of the Q1 results, I took advantage of a substantial run-up in AYX stock price as we approached the Q2 report. I thought the risk/reward ratio made sense when AYX was trading in the $170 range. After the disappointing Q2 report, I sold the rest of my position after-hours in the $130 range.

While I prefer to hold stocks for long periods, the dramatic change in Alteryx’s sales execution gave me pause, even when accounting for the COVID-19 situation. I am still bullish on AYX for the long term and may re-enter a position next year, as we see the impact of the COVID-19 recovery and transition of the adoption licenses. These factors could re-accelerate revenue growth for 2021, making the potential increase in valuation more immediate. I will also look for further developments in the evolving competitive position and general big data processing ecosystem.

In the meantime, I had applied the proceeds from the AYX sale to an initial position in Cloudflare (NET). This was in reaction to their strong Q2 earnings report and my findings after covering their Serverless Week in a prior post. I will revisit AYX going into 2021.

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.

Useful Links:

For a great overview of Snowflake and analysis prior to their IPO, please see this write-up from peer analyst Muji over at Hhhypergrowth.

7 Comments

  1. Tim

    Hi Peter,

    Thank you as always for an amazing write-up. On the topic of selling out of AYX, I’ve also noticed you have dropped OKTA from your holdings.

    Will you be doing a review of OKTA’s latest quarterly and the reasons behind your decision?

    Thanks so much!

    • poffringa

      Thanks for the feedback. Yes – I am planning a full write-up on OKTA’s prior quarter and rationalization for my portfolio change.

  2. Joe

    Thanks again for your insights
    And thanks for the comments on snowflake. Having followed public companies for decades I have often seen scenarios where the narrative for a stock pullback is x and often repeated by many analysts then y comes out of the blue and in retrospect was more likely the issue for the stock price fall all along.

    I wonder whether the emergence of snowflake as a potential competitor is a big factor in why this stock has not performed as well as so many other Saas stocks.

    I know that you don’t cover snowflake but do you see their valuation as ludicrous or does this have the potential to be a trillion dollar company?

    • poffringa

      Thanks. It’s hard to say whether Snowflake’s rise explains the slowdown for AYX recently. Obviously, that correlates more directly with COVID-19. However, Snowflake’s ecosystem could create drag for future Alteryx upsells at enterprises, particularly for those that consolidate their data onto Snowflake or another cloud data platform.

      Regarding SNOW as an investment, I don’t think a P/S ratio over 100 is justifiable at this point. I prefer to observe IPOs for a few quarters to get a sense for where the revenue growth rate will trend, before considering a position.

  3. Gal

    Many tnx Peter.

  4. Luca

    Fantastic analysis. For some reason when it comes to AYX, I always find myself wondering about a potential buyout at some point.

    Curious about your thoughts on that and if you were to speculate, which companies would be a good fit (e.g., Microsoft, IBM, Salesforce or even Snowflake)

    • poffringa

      Thanks for the feedback. On buyout speculation for Alteryx, that could go in a lot of directions. Ideally, it would be a company that has a strong cloud presence already and is looking to add specialized data analytics expertise. A legacy player like IBM or Oracle would be interesting.