Cloudflare (NET) Q1 2023 Earnings Review

After setting an optimistic target for full year 2023 revenue growth in early February, Cloudflare management was forced to reset guidance lower after sales cycles unexpectedly elongated in March. While this can be attributed to exogenous effects like banking turmoil and macro conditions, investors are left guessing whether the Q1 results were a one-time speed bump or a precursor to further deceleration in growth.

This dilemma would be difficult to reconcile had we been left with just the Q1 report. That earnings call conflated a few contributors to the underperformance, including longer customer sales cycles and a restructuring of the GTM team. Fortunately, Cloudflare leadership scheduled an Investor Day a week later, which provided a much clearer picture of the performance drivers and a better understanding of Cloudflare’s plan to address these challenges.

Stepping back, Cloudflare’s history and future growth are grounded in disruption. An investment in Cloudflare represents a bet that their architecture choices and platform strategy will allow them to penetrate new markets through lower cost, better performance and a more robust feature set. They have applied these architectural advantages to pursue an enormous addressable market, representing some of the fastest growing segments of software and security infrastructure.

Cloudflare’s elevated valuation relative to peers hinges on the market’s assumption that Cloudflare will continue to innovate and increase enterprise adoption of their newer product categories like Zero Trust, network services and a distributed developer platform. Rapid expansion in these areas will offset the gradual decay of growth rates in their traditional product offerings like DDOS, application security and CDN.

Audio Version
View all Podcast Episodes and Subscribe

While the Q1 earnings report on its surface would indicate that future product expansion isn’t working, detailed product and customer metrics shared at Investor Day imply that it is. Customer take rates for products in network services, Zero Trust and the developer platform are increasing much faster than the base of application services. These Act 2 and 3 products now contribute more than 25% of customer ACV. Further, Cloudflare is emerging as a multi-product story, with customer adoption rates for 6+, 8+ and 10+ product subscriptions rivaling those of peers like Datadog and Crowdstrike.

Cloudflare’s $5B revenue target relies on gaining significant share of infrastructure spend from the largest enterprises, requiring $10M+ deals as common as those announced by competitors like Zscaler and Palo Alto Networks. Cloudflare is demonstrating traction here, with metrics from Investor Day revealing substantial spending growth from Cloudflare’s largest customers. These beachheads with the early adopters will need to become mainstream.

Arriving just in time is an overhaul of Cloudflare’s GTM effort. New leaders in sales and marketing have the opportunity to level up the performance of these departments to match that of the product and technology organization. While these kinds of transformations take time, we could see incremental sales productivity materialize as we transition into 2024. Better marketing messaging already appears to be filling up the sales pipeline, which management claims is at record levels.

In this post, I briefly revisit Cloudflare’s product strategy and the competitive advantages driven by their unique platform architecture. I’ll then tie this to the GTM restructuring to establish a baseline of the potential to penetrate the enormous addressable market that leadership has targeted. Finally, I will discuss the Q1 earnings report and contrast the dismal outlook with what we learned about Cloudflare’s growth mechanics as part of Investor Day.

For readers interested in a summary of the financial results with technical charts, please check out the Q1 earnings review published by our partners over at Cestrian Capital Research.

Cloudflare’s Product Strategy

I have written extensively about Cloudflare’s product strategy in past blog posts. Foundational to Cloudflare’s platform and their inherent advantages over incumbents is the extensive network they have built. Over their 10 year history, they have committed to several architectural decisions that make their platform unique in its ability to deliver lower costs, better performance and more customization than competitors.

This approach also created a longer path to growth and market penetration. These design principles run counter to prevailing approaches taken by the latest breed of SaaS companies that simply build their services on top of hyperscaler infrastructure. While that approach reduces time to market, over the long term, this dependence on hyperscaler infrastructure will limit their ability to match Cloudflare’s pricing, performance and feature set.

Cloudflare’s design tenets can be summarized as follows:

• Owned and Operated. Cloudflare built and manages data centers in proximity to 285+ cities around the globe. This puts a data center within 50ms of 95% of the world’s population. They have also assembled a global network with over 12,000 network interconnects with cloud providers, ISPs and enterprises. While this required significant CapEx and technology investment, Cloudflare is decoupled from reliance on other parties for their infrastructure. They incur no incremental cost to launch new services on their existing network (like Zero Trust).
• Run Everywhere. Every one of Cloudflare’s products runs on every server on every data center in parallel. This ensures that critical functionality is as close to the end user as possible and allows any excess capacity to be leveraged across the full network. Competitor approaches often involve backhauling traffic to central data centers or separating out products onto different networks. While Cloudflare’s approach is more complex, it provides the best performance and most efficient use of resources.
• Composability. Because Cloudflare chose to build a developer runtime (Workers) directly into their network, all services can be customized and even combined with other products. This work can be performed by developers within the customer organization, versus requiring professional services support from Cloudflare or a third party. This composability has also been a driver of new product offerings, as previous capabilities are extended to meet the requirements of new product categories or customer needs.
• Free Tier. While Cloudflare has over 168k paid customers, they offer services to millions of free users. With usage caps, these users consume a small amount of capacity. However, their benefit far outweighs the cost, as these users are leveraged to test new features and provide signals for product preferences. For security use cases, these millions of users represent a valuable source of real-time threat intelligence. Over 20% of the world’s top million busiest web sites are on Cloudflare’s backbone, allowing Cloudflare’s network services to avoid the public Internet when handling enterprise employee traffic.

An assumption that these architectural advantages will allow Cloudflare to disrupt existing providers in each product category underscores the investment thesis. Business performance may cycle through periods of strength and weakness as they traverse various product adoption curves, but Cloudflare’s architectural approach will eventually surpass competitors on price, performance and feature set. As with any technology disruption, change will often take longer than expected and will be faced with counteractive measures from incumbents.

Product Development Waves

Cloudflare has organized their product strategy into Acts. Each of these represents a product segment with a target buyer. Capabilities in later Acts often leverage technology developed as part of an earlier Act, reflecting the strong composability of the Cloudflare developer platform. As expected, the order of the Acts also correlates to maturity of the product and relative penetration in the market.

• Act 1 – Application Services
• Act 2 – Zero Trust Services and Network Services
• Act 3 – Developer Services

Cloudflare leadership held an Investor Day on May 4th, about a week after Q1 earnings. This event was well-timed, as management was able to elaborate on many of the issues that surfaced in the Q1 earnings report. Topics covered included a detailed overview of the product strategy, the GTM overhaul and more granular financial data. These helped form a clear picture of how Cloudflare is overcoming challenges in GTM and positioning for growth over the next several years.

While there were no new product announcements, the Investor Day presentations provided a very useful view of Cloudflare’s product landscape and how it is evolving. What struck me was the improvement in the polish and customer focus associated with product presentations, which included a number of new graphics and tables to describe the product offerings. For each product grouping (Acts 1-3), the associated product VP discussed the customer use case, value proposition, buyer persona and competitive set in very specific and articulate terms.

I don’t recall seeing Cloudflare’s product landscape presented in such a tight, market-focused way in the past. This is clearly the output of the recent hires of new heads of marketing and sales. Looking at the slides, investors can appreciate the improved alignment between marketing, product and sales. This should help as Cloudflare revamps their GTM effort.

During Investor Day, Cloudflare leadership provided another update on their addressable market. The primary change was the addition of Cloudflare Developer Services as a discrete market segment (Act 3). This encompasses both the serverless application platform (Workers) and object storage (R2). In the previous TAM diagram, the Workers product was parked in future growth and R2 solely contributed the revenue projection for the top market segment. Additionally, the estimated size of the market for 2023 was raised by $21B to $146B and now includes estimates out to 2026.

While these types of TAM slides for software providers can feel a little outlandish, they do serve to clarify the target market and set expectations for the size of the opportunity. If the software provider can’t sell their wares to customers, however, then the TAM slide is pointless. On the earnings call, Cloudflare’s leadership discussed some challenges in the sales organization. During Investor Day, Cloudflare’s new CRO addressed these head on.

Go To Market

During the Q1 earnings call, the Cloudflare team highlighted two sets of issues that are impacting revenue growth. These were somewhat conflated on the earnings call, causing more confusion for investors than explanation. During the Investor Day event, management was able to address each issue separately.

First, sales cycles lengthened. This factor had a more acute impact on Q1 revenue performance and was the primary driver of the reset in full year guidance. While the sales pipeline for new customer generation was healthy, the time it took to close the average deal in Q1 shot up unexpectedly by 27% as compared to the average of the prior four quarters. For existing customers, expansion deals took even longer with a 49% increase in the average sales cycle.

These extensions were primarily caused by hesitancy from large customers during the first couple of weeks of March. This period coincided with the failure of Silicon Valley Bank on March 10th, which raised questions about the banking arrangements of every company in the technology sector. Customers became increasingly cautious, scrutinizing deals more closely than before. This resulted in lengthened sales cycles, delays in collections and deals getting compressed into end of the quarter. According to the CEO, almost half of Cloudflare’s new business closed in the last two weeks of March.

New pipeline growth again remained strong in the first quarter, exceeding our internal plan for the second consecutive quarter in continuing the trend of accelerating growth in new pipeline generation for the third consecutive quarter. However, the significant incremental caution exhibited by customers resulted in a pronounced decline in our close rate and an extreme back-end weighting of ACV bookings during the first quarter, both of which were primarily impacted by the aforementioned longer sales cycles. More specifically, our average sales cycle during the first quarter was 27% longer than the average of the previous four quarters. Sales cycles increased most significantly in expansion deals with our contracted customers, which were 49% longer than the average of the previous four quarters.

Cloudflare Q1 2023 Earnings call, April 2023

This would obviously impact revenue growth by pushing deals into the following quarter. It would also explain a drop in the dollar-based net retention rate. Looking beyond Q1, Cloudflare leadership is assuming that these longer sales cycles persist through 2023. This drove the need to reduce the full year revenue guide, as the extended sales cycles would keep delaying deals into the following quarter until the impact averages out next year.

This circumstance of longer sales cycles wasn’t unique to Cloudflare. Security peer Tenable reported a similar pattern for March on their earnings call and lowered forward guidance for Q2 and the full year. They also called out high demand and record pipeline generation, in spite of the longer time needed to close deals.

Some of the macro conditions we’ve been discussing with back-end loaded quarters, longer sale cycles, and more scrutiny of large deals, particularly new business, continued during the quarter. The last two weeks of March exacerbated these conditions as the banking crisis came to light. Specifically, we saw longer lead times in purchasing and approval phases of our sales cycle in the first quarter versus what we had typically experienced. As a direct result, a number of highly qualified committed deals pushed out of the quarter, which is reflected in our net new enterprise and our 6-figure adds. Of the deals pushed, many were in banking and financial services, and technology and telecom, traditionally strong sectors for Tenable.

As we navigate the current macro environment, there are two points that are important to note. The first is that demand remains strong. New pipeline generation added during the quarter exceeded our expectations and was a record for Tenable.

Tenable Q1 Earnings CAll, april 2023

Those security companies with quarter close in April (Palo Alto, Crowdstrike and Zscaler) may have avoided some of this impact, as March fell into the middle of their quarterly sales cycle. Zscaler’s recent pre-announcement of earnings results may reflect this and even imply that sales cycles are improving.

The other factor that is impacting Cloudflare’s results, but not specific to Q1, is the performance of the sales organization. As investors will recall, Cloudflare leadership raised the GTM effort as an opportunity for improvement coming out of Q4. On the Q1 earnings call, Cloudflare’s CEO provided more details on sales rep performance, specifically pointing out that about 100 sales reps have been consistently underperforming.

As this issue with the sales organization isn’t new, it wouldn’t be the cause of the sudden drop in revenue delivery experienced in Q1. Rather, it represents a broader opportunity for improvement. While Cloudflare leadership could have started the process to address this issue earlier, sales were easy during Covid, talent was extremely expensive and Cloudflare’s growth priorities were firmly focused on product development.

Improvement Plan

Enter Marc Boroditsky. As investors will recall, Cloudflare announced the hiring of Marc as part of their Q3 earnings report back in November 2022. At that point, Marc had been on the team a little over a week. In the Q4 report in February, Cloudflare’s CEO referenced their 100 day review with Marc a week prior, in which the CEO shared that the sales organization had significant room for improvement and that he was “embarrassed” about the state of affairs previously.

Last week, Marc briefed me and Michelle on his first 100 days. My initial reaction, if I’m honest was embarrassment, over some of the basic things we should have been doing better, but my second reaction was excitement is there are so many opportunities for us to improve. Now we’re focusing on becoming a leader in the go-to-market side as well. I hear the excitement from our existing sales and marketing teams at the rigor and discipline Marc and Brent are bringing to those teams. And what I’m watching carefully is another important pipeline, the pipeline of new sales talent. We’re seeing incredible people from the leading sales team in the world apply to work at Cloudflare. We aim for nothing less than to build one of the leading sales organizations in the world.

Cloudflare Q4 2022 Earnings call, February 2023

Given this, the progression in the Q1 report to a more specific assessment of the state of the sales organization really shouldn’t have been a surprise to investors. If the CEO was embarrassed in Q4, the shift to singling out 100 underperformers isn’t that much worse.

During the Investor Day, Boroditsky presented his assessment of the sales organization and the plan for improvement. I encourage all Cloudflare investors to watch the replay. Having worked with a number of sales executives in the past, I found Boroditsky’s presentation clear, direct and actionable. He understands the sales process and appears very structured and objective. I liked the plan and came away agreeing that there are opportunities for improvement which appear achievable.

Here are the primary components of the plan:

• Talent Reset. Top performers make up 15% of the sales organization and consistently achieve 129% of quota. Of this segment, productivity only declined by 1.7% from 2022 to Q1 2023. On the flip side, about 100 sales reps have consistently missed quota and only contributed 4% of ACV. Further, their productivity is about 20% of even the average sales rep. This means that a significant gain could be achieved by just replacing the low performers with average performers, versus having to go recruit all top sales people. By reaching a more normalized distribution of quota attainment, Boroditsky estimates that AE productivity would increase by about 20% overall.
• Streamline the Sales Process. Boroditsky tracked the tools and steps required for an AE to create and close a customer deal. He counted over 20 individual tools and 100 steps. There are numerous opportunities to simplify this process. Additionally, product pricing and packaging options should be better structured and transparent across the organization.
• Organize the Sales Team around Specialization. Structure individuals and teams along geographies, industries and customer sizes. Map supporting roles to salespeople in customer service, sales support, product and finance. Make specialized resources available to assist in the sales process for technical subjects like security, network design and developer services.
• Improve sales materials and training. The marketing team has already delivered new pitch decks. Products are organized around solutions and targeted at personas. Solutions are aimed at customer pain points and product categories align with the customer org chart. They map out the customer journey and identify when to engage partners.
• Instrument the Process. Break the sales process down into discrete steps. Apply clear measurement criteria to each step and track progress over time. Identify areas for improvement and set goals. Instrumenting the sales process should improve estimates and consistency.

Given that we are 1-2 quarters into this reset, investors are left to wonder how long it will take. Boroditsky didn’t provide a timeline, but scoped the impact of the changes he proposed for 2023. He projected that the talent reset could improve sales productivity by 20%.

During the Q&A section of Investor Day, Cloudflare’s CEO shared that they gave Boroditsky 120 days to form his improvement plan, at which point they shared it with the Board. That implies it has been in place since late February (about 2-3 months ago), with process changes already underway.

In terms of length, the talent reset and new AE ramp will likely require the most time. During the Q&A section of Investor Day, the CFO shared that the org changes in North America should be happening within a week and in Europe after that. Fortunately, the leadership team claims that AE ramp is very efficient at Cloudflare. They usually see mid-market AEs reach full productivity within 4 months and Enterprise AEs typically need 6 months. If the process follows this track, the CFO’s comments imply they may see some benefit by end of year.

---

Sponsored by Cestrian Capital Research

Cestrian Capital Research provides extensive investor education content, including a free stocks board focused on helping people become better investors, webinars covering market direction and deep dives on individual stocks in order to teach financial and technical analysis.

The Cestrian Tech Select newsletter delivers professional investment research on the technology sector, presented in an easy-to-use, down-to-earth style. Sign-up for the basic newsletter is free, with an option to subscribe for deeper coverage.

Software Stack Investing members can subscribe to the premium version of the newsletter with a 33% discount.

Cestrian Capital Research’s services are a great complement to Software Stack Investing, as they offer investor education and financial analysis that go beyond the scope of this blog. The Tech Select newsletter covers a broad range of technology companies with a deep focus on financial and chart analysis. Additionally, they provided a useful summary of Cloudflare’s Q1 report with financials and technical analysis.

---

Q1 Earnings Highlights

In the run-up to Q1 earnings, NET stock was enjoying a 33% YTD appreciation. This primarily hinged on their bullish revenue growth outlook for FY2023 coming out of the Q4 earnings report. While peers like SNOW, DDOG and MDB all set FY2023 revenue estimates much lower, the Cloudflare leadership team felt like they had sufficient visibility into revenue trends to establish a baseline for 37% y/y growth. At the time, they described the target as “prudent” and responded to repeated analyst questions about the aggressiveness of the target that they were sufficiently conservative.

Historically, Cloudflare has set the full year revenue target closer to the actual outcome than many peers. Exiting 2021, their initial read for 2022 was for 41.6% growth. They eventually delivered 48.6%. While some level of outperformance is generally baked in, Cloudflare’s guidance has traditionally offered less expected upside. This contrasts with Datadog and Zscaler which finished their prior fiscal years outperforming their initial Q4 guides by over 20 points of annual growth.

So, one interpretation of the 37% lead was that management expected to reach this level of growth, with some small upside. By projecting forward their large customer revenue trends coming out of Q4, there was a path to $1.336B in revenue for FY2023. Cloudflare leadership had been providing the percentage of revenue from large customers for several quarters, making it clear that this was the source of outsized revenue growth.

Unfortunately, this rapid growth in large customer spend hit a speed bump in Q1, resulting in a drop in the rate of sequential growth. Carrying the impact of that drop forward through the remainder of the year required that the full year revenue estimate be reduced. To get back on track would depend on reacceleration of large customer revenue growth sequentially at a sufficient magnitude to make up for the gap in Q1. Unless elongated sales cycles return to prior levels, that assumption is unrealistic.

The question for investors will be whether Cloudflare’s revenue growth can remain on the revised track set for 2023. Further, they will need to maintain this growth rate (and probably accelerate) in order to reach $5B in revenue anywhere near 2027. Providing a little more support for Cloudflare’s valuation is their improving profitability, with operating income and FCF continuing to hit records. Cloudflare is not yet a cash flow generating story and the stock price is anchored in expectations for future revenue growth, dependent on their ability to meaningfully penetrate the large TAM they have targeted.

Revenue

For Q1, Cloudflare reported revenue of $290.2M, up 36.8% y/y. This result fell into the range for $290M-$291M that the company had projected from Q4. Q1 revenue was up $15.5M or 5.6% above Q4 revenue of $274.7M. A 5.6% sequential increase in Q1 was actually a pretty good result. Some peers in software infrastructure and security had smaller Q4 to Q1 increases. Examples include Datadog at 2.8% and Confluent at 3.3%. Even the hyperscalers had minimal sequential revenue growth, landing in a range of slightly negative (AWS) to about 2% (GCP).

With their pre-announced quarterly results, Zscaler delivered 7.6% sequential growth. As I mentioned, their quarter ends in April, so it is possible they were able to close some of the deals that might have pushed out of March. We will find out more when they provide their full earnings results on June 1st. As an aside, I think Zscaler is showing strong momentum, after the prior quarterly report disappointed the market with a lower than expected billings forecast. While Zscaler and Cloudflare compete, Zscaler’s growth durability reflects positively on market demand for Zero Trust solutions in general.

The challenge for Cloudflare coming into Q1 was that the Q2 revenue target necessary to keep the full year projection on track was for $320.0M, up 36.5% annually and 10.0% sequentially from the Q1 projection. A normal cadence for Cloudflare would be to beat the prior quarter revenue estimate by several percent, allowing for a similar sized sequential raise to reach the next quarter target. In this case, Cloudflare normally would have delivered Q1 revenue in the range of $295M-$300M (or 3-4% sequential beat) and then anticipate another 7%-8% of sequential growth to reach next quarter’s level.

With Q1 actual revenue only meeting their estimate, the next step to Q2’s revenue target of $320M would dictate a full 10% or $30M of sequential growth. This goes beyond Cloudflare’s normal cadence and is most likely not achievable in this environment. Even with the explanation of longer sales cycles, it would be highly risky to assume that sales cycles would shorten, allowing pushed deals to close in Q2 along with the previously anticipated volume.

This forced Cloudflare leadership to reset the revenue target for Q2 and FY2023. For Q2, they projected revenue in the range of $305M-$306M for 30.2% annual growth. This was far below the analyst target for $320M (36.5% annual growth), which would be needed to hit the full year target growth rate of 37%. To reach the revised Q2 target, Cloudflare will need to increase revenue in Q2 by 5.3% sequentially or $15.3M. Given that Q1 sequential growth over Q4 was $15.5M, this appears achievable.

To remain on track for the rest of the year, the big next step will be to beat the Q2 estimate sufficiently to reach the revised Q3 revenue target. That now stands at $329.5M or 29.8% annual growth. This is $24.0M or 7.8% above the revised Q2 revenue target. This means that Cloudflare will need a healthy beat of 3-4% or about $10M in Q2 to then hit the Q3 target. Prior to Q1’s sequential growth of 5.6%, Cloudflare was delivering about 8-10% of sequential growth each quarter in 2022, so it is possible. But, the magnitude of quarterly beats have been decreasing.

In the Q1 earnings report, Cloudflare lowered the full year 2023 revenue guide by $54M to a range of $1.280B – $1.284B for 31.4% annual revenue growth. The prior range set by management in the Q4 report was for $1.330B-$1.342B for 37.0% annual growth.

Last quarter, we highlighted our expectation for sales cycles to continue to lengthen for both the first quarter and full year 2023. However, the level of elongation experienced in the first quarter was unprecedented and far surpassed our forecast entering the year, pressuring revenue growth. As a result, for the second quarter, despite the continued reacceleration of our new pipeline generation and our high win rate against the competition, we’ve assumed sales cycle to remain at the elevated levels experienced during the first quarter and have, therefore, also assumed close rates remain at continued low levels.

Cloudflare Q1 2023 Earnings call, April 2023

The revised full year 2023 target is probably where Cloudflare should have set revenue growth coming out of Q4. Other software infrastructure companies established lower expectations with their Q4 results. Snowflake and GitLab even lowered the preliminary estimates they had verbalized in their Q3 results. Others, like Datadog and MongoDB, provided a target in Q4 that was lower than analyst estimates, which arguably wasn’t a true “miss”. In all cases, those companies provided themselves with a little more leeway for 2023. Their stock price reflected the lower targets as well, with most software peers registering a smaller YTD gain than NET coming into Q1 earnings.

The big question going forward is whether Cloudflare can get back to their previous cadence of quarterly beats to deliver the required sequential growth to maintain the full year target. While that was revised down to 30% annual growth, it still requires sequential quarterly revenue growth of about 8% for Q3 and Q4.

RPO reflected a similar slowdown as revenue growth. It reached $959M at end of Q1, representing an increase of 5.6% sequentially and 39% year over year. In Q4, RPO growth was a healthy 9.1% sequentially and 45% y/y. Current RPO was 75% of total RPO in Q1, up from 74% in Q4. If RPO growth can revert to the 7-8% sequential growth range, then 30%+ revenue growth would be supported.

Profitability

Profitability is an area where Cloudflare is making solid progress. In spite of the revenue growth headwinds, Cloudflare delivered record operating income and continued the trend of positive free cash flow. Both of these were investor complaints in the past.

Non-GAAP gross margin was 77.8% in Q1, ticking up from 77.4% in Q4, but down from 78.7% a year ago. Cloudflare’s published long term target range for Non-GAAP gross margin is 75%-77%. By managing their own data centers and network, Cloudflare is able to drive efficiency in their delivery infrastructure. The team can identify underutilized resources and promote products or services that capitalize on them. This was the genesis for R2, as an example, where extra disk capacity on the network is harnessed to offer a distributed object store.

Non-GAAP income from operations was $19.4M, representing 6.7% of total revenue. This compares to $4.9M, or 2.3% of total revenue, a year ago. It also beats the prior record from Q4 of $16.8M, or 6.1% of total revenue. Even on a GAAP basis, Cloudflare is making gradual improvement, with GAAP operating margin of -16.3%, as compared to -18.5% in Q4 and -18.9% a year ago.

In Q1, operating expenses as a percentage of revenue remained consistent sequentially and decreased 5% year over year to 71%.

Leadership has set a long-term target for Non-GAAP operating margin of 20%+. This would be achieved by continuing to lower G&A (target range of 8-10%) and S&M (target range of 27%-29%). R&D has already reached the long-term target range of 18%-20% of revenue. As Cloudflare focuses on innovation and product development, it’s nice to see that no further cuts to the R&D budget as a percentage of revenue are needed.

Net income on a Non-GAAP basis was $27.2M in Q1, compared to $3.5M a year ago. On a per share basis, net income was $0.08, up from $0.01 a year ago. This doubled the company’s prior guidance for a range of $0.03 to $0.04 and analyst expectations for $0.03.

Cash flows were also favorable in the quarter. A year ago, Cloudflare delivered terrible free cash flow, with FCF dropping to -$64.4M for a FCF margin of -30%. After delivering positive FCF of $8.6M in Q4 of 2021, the subsequent decrease of $73M in Q1 2022 was disappointing. This year, the story was better. Cloudflare ended Q1 2023 with FCF of $13.9M or 4.8% margin. This is down from the seasonally strong Q4 2022 FCF of $33.7M for a FCF margin of 12.3%, but remaining positive in Q1 is a good sign. This was helped by a lower than normal CapEx allocation of 5% in the quarter, versus the target for 11% – 13%. Cash flows were pressured by collections in Q1, particularly in March, similar to the trend reported for revenue.

The positive performance in Free Cash Flow has benefited Cloudflare’s cash balance, which has increased for the past two quarters. Investors can take comfort in knowing that Cloudflare is cash flow positive now and isn’t in a position to raise more money to fund operations.

Cloudflare added 174 employees in Q1, representing a re-acceleration sequentially over their all-time low of 40 additions in Q4. As Q1 starts the calendar year, we would expect to see a larger number of employees starting with the company. Going back a year, this quarter’s additions were below the 310 employees gained in Q1 2022, but above the 143 headcount increase in Q1 2021.

I think the continued hiring indicates that management is still confident about the opportunity ahead. We also know that there is substantial churn in S&M, but an increase in total headcount indicates that they are hiring incrementally beyond the AE reset. On the earnings call, the CFO stated that they would continue the pace of hiring for the year, based on market conditions.

Looking forward, in spite of lowering revenue estimates, Cloudflare raised their profitability targets. For Q1, they had estimated Non-GAAP operating income of $11.5M to $12.5M and delivered $19.4M. Looking to Q2, the estimate for Non-GAAP operating income notched up to a range of $14M – $15M. For the full year, the initial target for FY2023 operating income in the Q4 report was for a range of $54M – $58M. In Q1, they raised this to $73M – $78M for a 5.8% operating margin. Note that the raise was much larger than the Q1 beat, implying that Cloudflare expects further outperformance on profitability through the year.

Customer Activity

As discussed, in the Q4 earnings report, Cloudflare’s preliminary full year revenue guide for 2023 appeared aggressive. While peers in software infrastructure were dialing back projections from analyst estimates, Cloudflare set the revenue bar higher than expected. Analysts were looking for revenue of $1.312B (up 34.6%), and Cloudflare set preliminary guidance at a range of $1.330B – $1.342B for 37.0% annual growth.

In trying to reconcile their high revenue guidance in the Q4 report, I dug into customer spending trend data. I discovered that revenue growth isolated to just the large customer segment was much higher sequentially than for the company overall. Additionally, this sequential growth had been consistent through 2022 and even accelerated in Q4, in spite of deteriorating economic conditions. As management had been publishing the percentage of revenue represented by large customers over the course of 2022, investors could extrapolate out growth for just this segment.

Projecting a similar level of sequential growth forward, I found that the increasing revenue contribution from large customers would allow Cloudflare to reach their full year revenue target of about $1.336B at the mid-point. Based on those assumptions, the preliminary full year guide coming out of Q4 appeared optimistic, but not impossible.

In Q1, growth in the large customer segment slowed down substantially, as a consequence of the sudden lengthening of sales cycles during March. This applied to both new and existing customers, but the delay of expansion deals with large customers would have a more acute impact on revenue delivery within the quarter.

This manifested in the dollar-based net retention rate (DBNRR), which dropped by 500 bps sequentially to 117% in Q1. This had been 127% a year ago. Cloudflare’s method for calculating DBNRR is based on annualizing customer spend in the current quarter and comparing that to annualized spend for the same customer in the year ago quarter. This differs from the method employed by other SaaS companies, which use trailing 12 month revenue.

Cloudflare’s limit to just the current quarter’s revenue can cause larger movements of DBNRR between quarters. Additionally, Cloudflare includes the impact of attrition (which is low) and excludes the contribution from customers that upgrade from the free tier (which is common). On the earnings call, management pointed to delayed expansion deals with larger customers as the cause of the continued deceleration in DBNRR.

Cloudflare ended Q1 with 168k total paying customers, an increase of 13% annually. This is up 6k or about 3.7% from 162k in Q4, which also added 6k total customers. While this level of total paying customer growth is slower than some other SaaS companies, the base of 168k provides a large pool of customers for spend expansion. Large customers make up less than 2% of this total, so the focus should rightfully be on increasing share from existing customers, versus expanding the total customer base. Additionally, current macro conditions are likely causing some small paying customers to transition to the generous free tier temporarily.

Cloudflare defines a large customer as spending more than $100k in annualized revenue within that quarter (not trailing 12 months). Cloudflare reported 2,156 large customers at the end of Q1, up 40.3% y/y. In Q4, there were 2,042 large customers, meaning Cloudflare added just 114 large customers in Q1 for 5.6% sequential growth. This is lower than the 134 or 7.0% sequential adds in Q4, and also represents the lowest count in the last 4 quarters. Going back to Q1 2022, large customers increased by 121 sequentially, which was also lower than the Q4 additions.

This slowdown in large customer growth reflects the same pattern of delayed expansion deals that impacted DBNRR. Customers on the threshold of passing $100k in annualized revenue would not do so in the same quarter, if their expansion deal took longer to close than normal. The 49% increase in sales cycles for expansion deals with contracted customers mentioned earlier would have this effect.

Customer and Product Highlights

As part of the earnings call, management shared a number of customer wins. The general theme was continued adoption of multiple products and particularly the newer Act 2 and 3 offerings. As I will discuss in a bit, Cloudflare’s CFO provided some very useful metrics associated with multi-product adoption and revenue contribution from Act 2 and 3 products during the Investor Day presentation. These customer wins reflected how Act 2 and 3 products have emerged to exceed 25% in ACV.

• A Fortune 500 media company expanded their relationship with Cloudflare, signing a three-year $840k deal and bringing their total annual spend to $2.1M. The customer had been using products from the application services category (Act 1) since 2017 and recently expanded into Zero Trust (Act 2) by purchasing Cloudflare Access and Browser Isolation for thousands of contractors. This deal represented a competitive displacement and could open the door for future expansion to the full employee base.
• A leading e-commerce technology company in Europe expanded their relationship with Cloudflare signing a three-year $780k deal. The company is migrating their network access and security to the Cloudflare One suite of products, including Gateway, Access, DLP and Magic WAN. The customer was attracted to the ability to consolidate all network access and security services onto one provider, which is a key competitive differentiation for Cloudflare over some other providers who only address Zero Trust (SSE) without fully combining that with network access (to achieve SASE).
• A global industrial machinery manufacturer expanded their relationship by signing a three-year $648k deal. The company is also migrating to Cloudflare One with Access, Gateway, Area 1 and Magic WAN. With this move, they are displacing several vendors and consolidating on the Cloudflare platform to reduce costs.
• A Fortune 1000 SaaS company expanded their relationship with Cloudflare by signing a three-year $8.4M deal. This deal was primarily for application security and performance products, including data localization support. This shows that Act 1 deals can still drive significant upside.
• A leading IoT security company expanded their relationship, signing two three-year deals for a combined $4.2M. The company has been a Cloudflare customer since 2020, using core application services and Cloudflare Tunnel for IoT. The customer is now contracted for over 25 products across application services (Act 1), Zero Trust (Act 2) and the developer platform (Act 3). This deal includes use of R2 and Durable Objects to counteract high egress fees from AWS. They are also leveraging Cloudflare’s new FedRamp authorization.
• A Fortune 1000 retailer signed a $1.3M multi-year deal for Cloudflare’s core application services to replace their existing vendor. They also added R2.
• A leading gaming company expanded their relationship with Cloudflare, moving from a pay-as-you-go customer spending $200 a month to a contracted customer signing a one-year $388k deal for R2 and application services. The company was looking for a more scalable platform for storage to support their rapid growth and reduced costs. They were impressed with R2’s lower latency and better throughput, as well as how R2 tightly integrates with workers and core application services. This underscores the customization competitive advantage that Cloudflare provides by supporting the ability to glue several solutions together with Workers code.

Additionally, Cloudflare leadership sprinkled in some product updates for Q1 that were useful. These represented various spot metrics on new product adoption, particularly around Developer Services. While the revenue contribution is building (25% ACV contribution from Act 2 and 3 in 2023), the rate of growth appears much faster than for the company overall. As these services increase their contribution to revenue, their higher rate of expansion may drive a reacceleration in Cloudflare’s overall growth rates.

• There are now 4.9M applications running on the Workers platform, representing an increase of 146% over the last 6 months. That is certainly a rapid increase. I can’t point to a specific catalyst beyond greater awareness in the developer community and an improved marketing message. The vast majority of these users are pay-as-you-go developers.
• R2 has 33k paying customers as of Q1. They are storing more than 7 PB of data, which has increased 25% q/q. While I like the growth rate, the amount of data is still fairly small. R2 has a lot of potential and will likely see further rapid growth as customers get comfortable with the performance and start moving more data onto it. Additionally, the CEO has been highlighting new demand from AI companies using R2 to locate training data in close proximity to compute.
• Revenue generation from AI companies has grown “north of 20% quarter-over-quarter”. Leadership claims this spans many AI companies, not just one or two. These companies are attracted to the Cloudflare developer platform as their starting point, particularly because it integrates tightly with the application security and fraud prevention services (Act 1) also offered by Cloudflare.

What we’re hearing from AI companies is that as they look to who they’re going to use for their infrastructure and they can start with a clean slate, that Cloudflare is a part of that, and that we’re helping them go fast, compete, get the most efficiency, be able to protect themselves from some of the significant cybersecurity and fraud risks that they face. So, I think that that has continued to be a real positive for us, and that has only accelerated from even Q4.

cloudflare Q1 Earnings call, April 2023

Regarding R2 usage, Cloudflare leadership did clarify that they have been refining the use cases they want to target. They admitted that it isn’t a good solution for large-scale S3 cold storage, like log files that are accessed very infrequently. However, they are seeing strong interest from AI companies. These customers use R2 to store AI training data and make it available at whatever location offers inexpensive compute to run their training process.

Next week, investors will get further insight into the product roadmap for R2 and AI-related services. Cloudflare will be holding another innovation week, which they are referring to as Developer Week. Leadership is signaling that investors will receive updates on the Developer platform and AI-specific offerings, along with some customer stories. This may provide more signals for the growth path of the Act 3 product category.

Investor Day Financial Updates

As part of the Investor Day, Cloudflare’s CFO shared some additional insights into Cloudflare’s performance and growth strategy. By publishing metrics on multi-product adoption, large customer expansion and contribution from Act 2 and 3 products, he provided the underpinnings of support for Cloudflare’s continued growth in the large TAM they are targeting.

One update that seemed lacking was a definitive confirmation of the $5B in 5 years revenue target. As investors will recall, this was introduced in Q3 2022, as Cloudflare’s revenue run rate crossed $1B annually. The leadership team then set the projection that they would reach a $5B run rate within 5 years, bringing Q3 2027 quarterly revenue to $1.25B (from about $254M in Q3 2022).

This long term growth target implied a revenue CAGR of about 37% for that 5 year period. When the Cloudflare team reported Q4 2022 earnings, they reiterated this target and even aligned the FY 2023 revenue estimate to it for a 37.0% growth rate. With the Q1 earnings report, they lowered the target growth rate for 2023 to 31.5%. This means growth in 2024 – 2027 would need to be higher than 37% to reach the original $5B target within 5 years.

During Investor Day, the CFO referenced the $5B revenue target, but didn’t explicitly connect it to the original 5 year timeline. I was hoping for some clarity here or an indication that they had adjusted the expected timeline, as the $5B revenue target represents a large component of Cloudflare’s investment thesis. If leadership feels confident that the $5B revenue target by 2027 is still largely on track, then it would remove some of the disappointment of the 2023 revenue adjustment.

I suspect that as 2023 progresses, we may get an update on this long-term target. Like most software infrastructure providers, Cloudflare has a few unknowns in motion right now, which make any long-term prediction like this difficult. We don’t know the path of overall IT demand and whether sales cycles will eventually recover. The team also needs time to gauge the likely impact of the GTM transformation. While these changes make sense and seem like they would result in a reacceleration of revenue growth, the timeframe is unproven.

With that said, the CFO focused his presentation on what is known, namely the drivers of revenue growth currently. He also shared signals that give confidence that Cloudflare’s revenue trajectory can continue for many more years.

He started by addressing the Q1 shortfall, which as I mentioned earlier was attributed to elongated sales cycles. In Q1, the average number of days required to close a deal increased by 27% overall and by 49% for expansion deals with existing customers. This change has a magnified impact on Cloudflare because deal cycles are typically short, normally closing within the same quarter. Slower closes push deals out into subsequent quarters.

The flip side is that new customer demand has been increasing, as evidenced by sales pipeline growth. The CEO explained that these represent qualified leads at the top of the sales funnel, which progress towards revenue generation eventually. This increase in sales pipeline has been occurring for the past 3 quarters, which may fuel revenue acceleration into the second half of 2023 as a greater portion of these leads transition into sales.

Shifting to large customers, Cloudflare currently enjoys penetration into 30% of the Fortune 1000 and has 6 of the Top 10 global companies as paying customers. As Cloudflare’s product offering expands and the GTM motion improves, a major contributor to the growth story will be increased spend from their largest customers. As an example from one shared data set, customers that reached the $5M threshold in Q4 2022 had increased their spend on average by 18x over the four year period since Q4 2018.

The CFO also showed the minimum spend needed to land in different segments of the largest customer cohorts. By the end of 2022, the smallest Top 10 customer was spending $6.2M annually. This is up almost 5x since 2018. It also implies that Cloudflare’s very largest customers have pushed past the $10M mark. Similarly, Cloudflare has 25 customers over $2.6M in spend and 100 customers over $888k. In Q4, they reported 85 $1M customers, up 52% y/y.

As Cloudflare drives towards the $5B revenue goal, a large contributor will be from $10M+ customers. The growth rate of these largest customers has been higher than Cloudflare’s overall revenue growth. As Cloudflare continues to build out the product offering, these large customers have more products to adopt. This larger product footprint provides more opportunities for consolidation of vendor spend and the ability to enhance the combined product subscription through customization via Workers

While Cloudflare has historically employed a product-led, bottoms-up sales model for Act 1 services, engaging customers of this size indicates that they are penetrating the enterprise. With the GTM restructuring discussed in the prior section led by Cloudflare’s new CRO, I think we will see further expansion of these large customer cohorts. The other benefit for Cloudflare is their foothold with the Fortune 1000, mostly through Act 1 products. Often representing a visible component of a company’s security strategy, these products provide an entry point to promote other Act 1 offerings and introduce Acts 2 and 3.

The growth of large customer spend is being driven by an increasing number of product purchases. As part of his presentation, Cloudflare’s CFO shared the percentages of contracted customers that subscribe to 8+, 9+ and 10+ products. These percentages have been increasing over the last 4 years. This is similar to the multi-module attach rates exhibited by software and security infrastructure peers, like Datadog and Crowdstrike. Multiple product attachments are also driving a larger share of total revenue. Customers with 6 or more product subscriptions contribute the majority of Cloudflare’s total revenue.

Related to the opportunity for Act 2 and 3 product offerings, the CFO provided updates on attach rates for each Act and the overall contribution to total revenue. In the Investor Day presentation from May 2022, he shared similar attach rate metrics by product type. We can compare the two sets of percentages for year/year improvement. Further, we can plug in the total number of customers for both periods to calculate the overall increase in product adoption by segment. All metrics provided were as of December 31st in each year.

Between December 31st 2021 and 2022, Cloudflare’s total customer count increased from 140,096 to 162,086 or 15.7%. This is compounded by the growth in attach rates for each product segment. Most notable is the rapid increase in adoption of Act 2 products, namely Network Services at 74% annual growth and Zero Trust more than doubling year/year to 131% growth. While many of these customers are small, this level of Act 2 and 3 product adoption is encouraging.

As we would expect, this rapid growth in Act 2 and 3 attach rates is showing up in revenue contribution. While the percent of revenue for Act 2 and 3 products was negligible in 2016 and 2018, it became material in 2020. By 2022, Act 2 and 3 products contributed more than 25% of total ACV. This is based on the annual contract value of customers over the full year of 2022. The percentage contribution may be even higher in the most recent quarter.

Finally, the CFO shared the long term operating model for the company. This was largely the same as that shared in 2022, except for the addition of a Free Cash Flow margin target of 25%. The CFO feels that a FCF margin of 25%+ is achievable given the unit economics in the business and opportunity to scale the operation on Cloudflare’s highly efficient platform. Investors have debated whether Cloudflare’s CapEx heavy model could ever produce an appealing FCF margin. Cloudflare leadership believes it can.

While Q1’s 5% FCF margin is still far from a 25% target, I like that management has asserted that a much higher FCF margin is possible in the future. If Cloudflare indeed reaches $5B in revenue exiting 2027, total free cash flow could be approaching $1.25B.

Investment Plan

The combination of a high valuation coming into the Q1 earnings report and a reset of full year revenue projections caused an outsized adjustment to NET’s stock price following earnings. The stock dropped 26.6% the following day, and continued sliding to a low of $40.48, before getting a recovery bid on May 5th. The stock has clawed back some of those losses, briefly crossing back over $50 on May 11th.

While investors were disappointed with the Q1 performance, we have to keep in mind that NET stock was up 32% YTD prior to earnings. With the reset, it is now up 9.3%. Interestingly, this puts it closer to peers like DDOG (up 19% YTD) and SNOW (up 17%). ZS stock is up 2% YTD, after dipping over 20% into the negative for the year, but then recovering nicely after their strong pre-announcement of this quarter’s results.

Cloudflare’s Q1 earnings report certainly threw cold water on the growth story. Revenue projections backtracked and large customer spend is under pressure. Profitability metrics are improving, demonstrating that Cloudflare is starting to achieve operating leverage. The key question for the remainder of 2023 is whether the revised revenue target can still be reached. It appears achievable, if macro conditions and customer purchase patterns do not deteriorate further.

Investor Day provided useful updates on the GTM restructuring plan and the expected benefit if executed successfully. This will likely require the remainder of 2023 before the impact is reflected in revenue growth, but could drive an acceleration in 2024. The new CRO projected that sales productivity could increase as much as 20% by transitioning underperforming AE’s to average output.

Additionally, the CFO shared encouraging data for multi-product take rates, large customer growth and adoption of the newer Act 2 and 3 product offerings. Customer additions of network services, Zero Trust and developer services is growing at a much faster rate than application services (Act 1). These Act 2 and 3 products now contribute more than 25% of customer ACV. As their growth rate is higher than for the company overall, we should see their contribution to total revenue continue to increase, providing a greater upward push on overall revenue growth rates.

While we don’t know the trajectory of macro conditions, we have indications that the rate of interest rate hikes is slowing. Currency exchange rates will become less of a headwind for the remainder of 2023 as well, and might become a small tailwind. While Cloudflare (and some other software infrastructure companies like Datadog and Snowflake) don’t perform a currency adjustment on their revenue numbers (because they charge in dollars globally), they have commented that a strong dollar can pressure sales contracts outside of the U.S. In Q1, Cloudflare generated 47% of revenue internationally.

The other potential incremental tailwind for Cloudflare could come from the new wave of investment in AI services. The leadership team has cited strong adoption by AI companies across a broad spectrum of size. These companies are drawn to Cloudflare’s platform for application services and newer developer products. They tend to start with application security and DDOS protection, and then evolve to use of Workers and R2. Storage of training data on R2 appears to be a popular, but unexpected, use case for these companies as they try to locate the most cost-efficient compute for generating their training models.

Cloudflare leadership expects to continue to evolve their developer platform and embrace these newer use cases from AI companies. They emphasize the seamless adoption of the Cloudflare platform for these start-ups that lack significant legacy infrastructure investment. This highlights the crux of the growth opportunity and challenge for Cloudflare. While the architectural choices at the core of their platform will yield better cost, performance and breadth of product reach, enterprise adoption is often a disruptive motion. Product growth curves of these sorts can often take longer than expected, but also move quickly if new breeds of customers emerge with different requirements.

While the Q1 report represented a setback, I am still optimistic about Cloudflare’s potential to realize meaningful penetration into the TAM they have targeted with their product offerings grouped around Acts 2 and 3. Similarly, the restructuring of both sales and marketing should provide additional positive benefit, as Cloudflare finds a more healthy balance between world-class product development and their GTM motion.

For my personal portfolio, I haven’t reduced my allocation to NET stock following earnings. I also don’t plan to increase the allocation, acknowledging the greater execution risk coming out of the Q1 results. If it appears that performance gets back on track as 2023 progresses, I may take advantage of price dips to add to the position.

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:

• Our partners over at Cestrian Capital Research published a Q1 earnings review which includes detailed financials and some technical analysis. This provides a useful supplement to my coverage.
• Muju over at Hhhypergrowth has published a review of Cloudflare’s Q1 results on his premium site. He also put out a preview of Developer Week with a prediction for a couple of new AI-focused products. Well worth the cost for additional perspective and deep product analysis.