For passive investors, this has been an easy year to be a bull. The market is up over 17% YTD in what is one of the lowest volatility years since 2017.
In 2017, we were in the first year of the Presidential Cycle, the market finished the year up over 19.4% without showing a single 5% dip. For comparison, the S&P 500, on average will experience a 10% or more correction almost every year, and has shown an average annual return over the last 30 years of 10.9%.
Similar to 2017, the S&P 500 is currently up just over 17% and has yet to provide a drawdown greater than 5%. However, unlike 2017, this year is not a broad market uptrend. In fact, the divergence we are seeing with a high number of stocks not participating in this year’s uptrend is important to note.
The above chart shows the price returns of the S&P 500 next to the percentage of stocks in the index that are above their Simple 200-Day Moving Average.
In a healthy market, the two graphs should be trending upward together, which would indicate more stocks across a wider variety of sectors are participating in the bull market. In an unhealthy market, we tend to see the majority of stocks in a downtrend, while the broad market keeps moving higher with narrower leadership. Note the pattern in 2020 – as the indexes makes a new high, the % of stocks above their 200-Day MA makes a lower high. This was a warning that the markets were weakening under the hood. Starting in March of this year the same divergence began, and is still playing out right now.
Sign up for I/O Fund's free newsletter with gains of up to 1100% – Click hereSign up for I/O Fund's free newsletter with gains of up to 1100% – Click hereClick here
Because the weighting of the Index is market cap weighted, which is just price times float, it allows for narrow leadership to hold the market up while the underlying stocks move into a correction. As of now, nine of the top 10 names in the S&P500 are Big Tech. Collectively, they account for 25.18% of the entire index weighting. Some may view this concentration as a negative; however, this concentration in Big Tech is what has allowed for the underlying index to complete a covert and deep correction in 2021, without bringing the market down with it.
The above graph compares some of the most important companies in the US Economy since December of 2020 to now. The red area indicates when these names topped and began their corrections. Many names topped in Dec/Jan of last year while a number of names topped in May/June of this year.
The top row, Microsoft and Apple, collectively account for 11.8% of the S&P 500. The combined weighting of Exxon, Caterpillar, JP Morgan, Walmart, Home Depot and Johnson & Johnson is only 4.97%. This shows how major stocks, as well as multiple sectors, can be in a correction while just a handful of key names remain strong.
2021 Has Tested Even Seasoned Investors
The eye-opening consumer price index (CPI) numbers in February/March announced that inflation was potentially here. This led to the indiscriminate selling of risk-on assets, focused specifically in your high beta names that are priced with future cash flows in mind.
We saw this as a buying opportunity within a much larger uptrend, which we expressed both publicly and within our premium service. For one, the renewed uptrend within the bond market as well as a collapse in many commodity names was suggesting that inflation fears may be over blown. Also, the technical analysis work that we do on broad markets was further suggesting that we still had higher to go.
Over seven months after the growth sell-off, we still hold true to our original thesis.
We spoke in length about the M2 money supply and why it is key to measuring actual inflation in the economy here. In short, it is the layer of the money supply that filters into liquid assets like money market funds, savings accounts, CDs etc. This is the layer that the FED does not directly control, and it measures actual credit/money entering the economy.
Even with a 34% increase in the M2 money supply, we believe that the two one-time events of excessive global stimulus as well as the re-opening of global economies will likely outweigh the increase in M2 on a meaningful level. These are two extraordinary events that are not being factored into current CPI numbers. Inflation is likely here, as suggested by the increase in M2, however we do not believe that it will have a meaningful and immediate impact on the current bull market that is underway.
Last year was arguably one of the more extreme anomalies in market history, while this year is forcing investors to address the consequences of global policy decisions. Rising rates and commodity prices, coupled with an unexpected dovish FED, has affected growth in different ways.
Many popular high fliers from last year, are negative for the year, as we see a rotation out of Covid names and primarily into tech stocks that surprise to the upside with strong fundamentals.
At I/O Fund, we raised cash going into the end of August due to the technical analysis that we perform on various broad market indexes. Our Elliott Wave work was suggesting that we were approaching a big top or minor top as the NASDAQ100 approached the 16000 level.
The above scenarios are what we based our broad risk management on in August. The red path had us topping in what would likely be a relatively deep correction, while the blue count had us only in a minor dip, which would be relatively brief. Keep in mind, both scenarios still had us in a much larger uptrend, which we see moving into 2022.
Through our cycle work, we had identified that Aug. 30 – Sept. 7 would likely mark an inflection point. This information coupled with fading momentum and a complete 5-wave pattern (in red), had us raise cash going into the inflection point.
In our August 19th report I stated that It’s my belief that the market is marching towards a large degree correction within a much larger uptrend. Whether that large degree correction has started or not will depend on what supports hold. Below is a visual of what I generally believe is playing out.” In other words, I was leaning towards the red count playing out, and so we prepared accordingly.
However, as we progressed in the initial dip, what had shift in real-time towards the more bullish blue count was a few data points. The primary one being sentiment, followed by seasonality trends and relative strength in key economically sensitive sectors. All of these data points were suggesting that a low was in, or at most, one more minor low was possible.
Sentiment
Implied Volatility vs. Realized Volatility
Prior to big unwinds in the market, we will typically see decelerating implied volatility while realized volatility starts ticking up. In short, sentiment is characterized by a belief that the bull market has much more room to go, which causes an increase in leveraged bullish bets.
On September 9th, with the S&P 500 down just 1.1% from its high, we were seeing an unusually large separation from actual volatility over a 30-day period, and what investors were expecting within the next 30 days.
The S&P 500 had a low realized volatility of just 7.5% while the implied volatility was well over double at 18.12%. What this means is that investors were willing to pay high fees for downside protection before at the very beginning of this drawdown.
On Sept 20th, with the S&P 500 down just over 4% from its high, this trend intensified to an extreme we usually see at major bottoms.
The S&P 500 had over 200% gap between its 30-day realized volatility and the implied volatility going forward. In other words, the premium between what investors were expecting and what was going on had investors willing to pay anything for downside protection.
Put/Call Ratio vs Realized Volatility
Another way to view the unusual level of fear in the market surrounding the recent drawdown is in the Put/Call Ratio vs. Realized Volatility.
The above chart takes the 30-day moving average of the Put/Call Ratio (green) and compares it to the 30 realized volatility of the S&P 500 (Blue). Note the pattern going into the February bear market. As realized volatility was actually increasing, investors couldn’t buy enough calls compared to puts. This is the type of sentiment that usually results in overleveraged long bets, and the unwinding of this leverage is usually what fuels large drawdowns.
Sign up for I/O Fund's free newsletter with gains of up to 1100% – Click hereSign up for I/O Fund's free newsletter with gains of up to 1100% – Click hereClick here
Compare that pattern to today. We are seeing the opposite unfold – as realized volatility continues to trend down, investors are buying more puts. This is signaling that sentiment is rather bearish to cautious as the market continues to climb higher in a low volatility environment.
The Importance of October
From our estimation, October should be strong. September is known as a seasonally weak period in the markets, while October is historically a strong month. This becomes evident when you look at the average monthly returns for the S&P 500 going back in time.
According to the same data, October has historically been a seasonally strong month. However, there is a caveat to how October tends to play out, and it is determined by how the market is trending into this month.
October has historically been a pivotal month in market history. We have seen an outsized number of major lows, and some major tops in the month of October.
The above chart shows the history on major turning points within the month of October. We’ve had 8 bear market/deep corrections bottom in October with only two major tops occurring in the month of October. In other words, how a market is trending into October is a key factor to the year-end trend.
This pattern also seems to play out on smaller scales. In 2019 and in 2020, September proved to also be weak months. We then saw the market find a bottom in October, which led to a year-end rally.
Conclusion:
When you note seasonal trends and historic patterns, coupled with an unusual amount of negative sentiment and the recent relative strength in economically sensitive sectors, we believe a similar trend will unfold into year-end. For this reason, we have been accumulating high conviction names that are showing excellent relative strength. Many of these names are embedded in strong tech microtrends that we anticipate to continue into the foreseeable future.
Disclaimer: This is not financial advice. Please consult with your financial advisor in regards to any stocks you buy.
Aehr Test Systems is a small cap semiconductor company that is nearing an inflection point. The company has developed a unique technology that provides tangible benefits for testing emerging semiconductor components, such as silicon carbide and silicon photonics. Silicon carbide is increasingly being used in EVs while silicon photonics is being integrated into edge computing data centers. I explain the tailwinds driving adoption of these materials in greater detail below.
Aehr’s key markets: Silicon Carbide and Silicon Photonics
Aehr’s wafer level burn-in testing systems are specifically built for testing silicon carbide and silicon photonics. Silicon carbide has recently been adopted by the automotive industry, which is driving demand for Aehr’s testing systems. Aehr disclosed in its 10K that “silicon carbide (SiC) semiconductors have emerged as the preferred technology for battery electric vehicle power conversion in on-board and off-board electric vehicle battery chargers, and the electric power conversion and control of the electric engines.
These devices reduce power loss by as much as greater than 75% over power silicon alternatives like IGBT (Insulated-Gate Bipolar Transistor) devices, which has essentially changed the entire market dynamic. With this development, the Company sees most, if not every automotive company that is working on electric vehicles, moving to silicon carbide-based powertrain and charging systems in the near future.”
Tesla was the first to start using SiC in its vehicles with its Model 3, and more EV manufacturers are quickly following suit, due to SiC’s ability to withstand hostile conditions, improve efficiencies and lower failure rates. CEO Gayne Erickson explained on the Q1 call that while the SiC market had been around for years, “it really started to take off when Tesla introduced their Model 3 with a silicon carbide traction inverter in it, and then quickly shifted all of their products to it.” The adoption by Tesla was a ‘jolt’ that has spurred further adoption by others. For example, German chipmaker Infineon Technologies launched an SiC inverter for electric vehicles in May 2021. According to Yole Research, “the silicon carbide power semiconductor device market is expected to increase over 500% between 2020 and 2026, growing at a compound average growth rate (“CAGR”) of 36% to $4.5 billion”.
All of these silicon carbide chips need to be tested, and Aehr’s wafer level burn-in testing systems are the most cost-effective way to test these chips at scale.
On top of the robust growth expected in the automotive SiC market, the company also has favorable tailwinds from the rise of silicon photonics, which is also starting to ramp. In its 10K, Aehr disclosed that its lead silicon photonics customer had moved to full volume production during 2021. It added that “we also saw three different silicon photonics customers move from engineering testing to high volume production test and burn-in of their devices using our FOX-P systems. We expect all three of these customers to ramp production during fiscal 2021”. Aehr’s customers must first perform lengthy qualification tests before ramping up orders. Aehr’s silicon photonic customers are nearly done qualifying and are finally expected to start putting in large orders this year.
During the Q1 call, CEO Erickson explained how we are still in the early days of silicon photonics. He stated that “companies like Intel and Nvidia are talking about integrating fiber optic transceivers into their core and graphics processor units or CPUs and GPUs. This is very exciting and we believe an enormous opportunity for Aehr Test with our unique position of having a cost effective and proven multi wafer solution for testing and burning-in or stabilizing silicon photonics devices at a massive scale while still in the wafer form”.
Silicon photonics are being used to increase communication speeds, which is critical for edge computing as it links 30-megawatt data centers within a 120 km distance to function like a 120-megawatt data center. This enables 100G Ethernet services for cloud operators and enterprises. Microsoft and telecom operators are both customers of Inphi’s silicon photonics. Beth had previously discussed this micro trend in her in-depth analysis of Inphi. She explained that the 100G switches were launched in 2020 but were more of a 2021 story as this is when they will be deployed in volume. Importantly, these 100G switches use silicon photonics.
We see this trend starting to ramp, as Inphi accounted for 10% of Aehr’s sales in 2021, the first time it was mentioned as a significant customer. As Inphi’s thesis plays out, Aehr will also benefit from the increased demand for its silicon photonic testing systems.
As shown below, Yole Research anticipates silicon photonics to rapidly grow at a CAGR of 49% through 2026. The market is relatively small right now, as customers complete lengthy qualification tests, but once the test systems are qualified, bookings of Aehr test systems will likely significantly ramp as silicon photonics gains market share. I discuss how Aehr’s order have started to ramp in 2021 in greater detail next.
Aehr’s testing technology, branded as FOX-XP test systems, was first introduced in 2016, and the company had initially sold only small orders of its testing systems (~$2 million to $5 million a quarter) as customers qualified the equipment before placing large orders.
During the fiscal Q4 Earnings Call in July 2021, CEO Erickson announced that the company’s lead silicon carbide customer had finally qualified Aehr’s FOX-XP test systems for “high-volume production” for wafer level burn-in testing for electric vehicles. This was a significant development, as it meant that orders for its test systems may soon start to accelerate.
Just five days after the Q4 earnings call, Aehr announced that it had received an $11 million order for its silicon carbide test systems with a “leading Fortune 500 supplier of semiconductor devices with a significant customer base in the automotive semiconductor market.” To put this order into perspective, it represented more than half of the company’s total orders in all of FY2020 ($16 million). In fact, since the close of Q4, the company has announced a total of $40 million orders to date (June through September), which is more than the aggregate bookings amount over the prior nine quarters.
The company’s strong orders led to an 80% increase in the company’s fiscal 2022 sales guidance, which Aehr now expects to be at least $50 million, or 3x the amount in fiscal 2021. Furthermore, attached to Aehr’s test system sales are high margin WaferPak and DiePak consumables, which are purchased multiple times over the life of a FOX testing system. Over time, management expects that sales of its consumables to be 4x the level of sales of its test systems.
This strategy of selling test systems and recurring consumables is similar to an ink and printer sales model, where the sale of the product leads to recurring sales of higher margin consumables. Aehr explained in its 10K that “as we continue to build our installed base of FOX systems, our WaferPak and DiePak [consumables] business continues to grow and is becoming a much more significant part of our business … our high margin proprietary WaferPak contactors and DiePak carrier consumables … can be up to four times the annual sales of systems. The systems typically are used for longer periods, with annual needs for new contactors and consumables. This is why we are confident that our consumable business is likely to exceed our overall systems business over time, even though both will grow in absolute dollars.”
To highlight the upside to the recent growth in orders, management explained on the Q1 Earnings Call that the $19 million order it recently received in September will likely be accompanied by a separate ~$10 million order for consumables, pushing the true order size up closer to $30 million. Importantly, these consumable purchases will also be recurring over the lifetime of the test systems, so the “lifetime bookings” value of this $19 million system sale may be closer to $60 million once consumables are considered
Continuing with this logic, Aehr’s $40 million in bookings YTD may be understated. Using Aehr’s disclosure from its 10K that recurring consumable purchases can be ~4x system purchases, then the $40m in YTD systems orders can result in a high-margin recurring sales stream of ~$160 million in consumable orders. Stated differently, the firm’s $40 million in test system bookings should also yield an additional $160 million in consumable purchases over the life of the systems, resulting in “lifetime bookings” of ~$200 million.
Importantly, orders are expected to continue to ramp going forward. CEO Erickson stated during the Q1 Earnings Call that “we believe we will add several new silicon carbide customers over the next 18 months that will ramp into production on our solutions.” He explained further that its leading silicon carbide customer is the “fourth largest in the space and not even half as large as the next largest”, meaning that new customer orders could be much larger. If Aehr’s test systems provide a competitive advantage (discussed below) then new customers will likely soon place large orders, further bolstering Aehr’s growth rate. I discuss what is driving the adoption of Aehr’s test systems in more detail next.
What is driving the adoption of Aehr’s test systems?
Aehr has a unique technology that is just now starting to ramp called FOX-XP test systems, which are used for wafer level burn-in testing of silicon carbide and silicon photonics. The main advantage of wafer level burn-in testing is that it reduces “infant mortalities”, or early failures in semiconductor equipment. Burn-in testing attempts to lower the failure rates from stage 1 of the “bathtub curve” (shown below), which increases the reliability of semiconductors.
Wafer level burn-in testing reduces chip failure, which is critical in certain industries such as EVs, 5G and datacenters. For example, if a chip fails in an EV’s drive train, then the passenger is walking home. The high costs of chip failures in EVs is driving the industry to push to zero failures and wafer level burn-in testing helps to achieve this.
“Bath Tub Curve” Representation of Chip Failures
CEO Erickson explained further on the Q1 call that he anticipates “that wafer level test and burn-in will become the industry standard for quality and reliability screening of silicon carbide devices”. He added that Aehr’s patented technology allows “customers to screen devices that would otherwise fail after they are packaged into multi die modules, where the yield impact is 10 times or even a 100 times as costly. With the most cost effective solution in the market to address this opportunity, we believe that Aehr has the chance to achieve a significant, perhaps dominant market share for silicon carbide wafer level burn-in.”
Assuming a yield impact of 10x to 100x, then early adopters of Aehr’s technology will gain a significant advantage over their peers. CEO Erickson added that customers with wafer level burn-in technology are “differentiating themselves against their competitors because they … [are] shipping a higher quality module … So other customers have figured this out, but they're still doing package per burn-in. So the discussions we're having with them in many cases is to move from package to wafer.” The 10x to 100x yield benefit awarded to early adopters can lead to a rapid acceleration in orders of Aehr’s test systems, since competitors will need to also adopt the new technology or risk falling behind. This dynamic rapidly moves Aehr up the “S-Curve” as customers rush to adopt the new technology and orders rapidly increase.
The qualification of the lead customer in July 2021 may have been the beginning of Aehr’s growth story (pictured below). While bookings are relatively low right now at ‘just’ $40 million, the company may be entering a period of growth, which is why the I/O Fund has taken a position. I discuss Aehr’s most recent financial performance and forward growth projections in greater detail next.
As mentioned above, Aehr reported a surge in bookings, as fiscal year YTD bookings (June through September) have grown to $40 million, or more than the prior nine quarters combined. These bookings are expected to convert into sales this year, which led management to raise its FY2022 sales guide to at least $50 million.
In the most recent Q1 quarter, sales increased 181% YOY to $6 million, while gross margin came in at 40%. Operating expenses are largely fixed, and operating income was a slight loss of $1 million during the quarter, while EPS was a loss of -$0.02, which beat by a penny.
Aehr has a history of losses, which makes sense as customers have been placing small orders for the company’s test systems as they qualify the equipment for high-volume use. With customers starting to ramp orders, the company has forecasted that it will be profitable going forward.
Specifically, CEO Erickson stated that the company breaks-even at around $28 million in annual sales, and for every dollar in sales above $28 million, it expects ~50% of that to flow down to earnings. Simply put, the $50 million guide for FY2022 sales should lead to ~$11 million in earnings. Moreover, since Aehr has a history of operating losses, the company has significant net operating loss (NOLs) carry forwards, which lowers the amount of taxes that the company must pay.
Taking this math a step further, the $11 million in earnings will equal about ~$0.43 in after-tax earnings per share (EPS). At ~$13/share, Aehr is valued at 29x forward earnings and 6x forward sales, which seems reasonable for a company expected to grow 200%+ this year. As discussed above, the company has numerous tailwinds as the SiC and silicon photonics markets rapidly grow at a 36% and 49% CAGR through 2026, respectively.
Above is a table of comparable semi-conductor equipment providers that are expected to grow at similar rates to Aehr. Relative to the above comparable companies, Aehr trades at a premium valuation based on forward sales and earnings. However, Aehr has exposure to markets that are expected to rapidly grow going forward (silicon carbide and silicon photonics) which supports a premium valuation. Aehr is also expected to grow faster than the peer median.
Below is a table of peers that Aehr competes against. Aehr is competing against some very large companies, however, Aehr is well positioned to benefit from the above-mentioned tailwinds. If Aehr can capture a dominate position in these rapidly growing markets, its market cap may close in on some of its peers. For reference, just six years ago in 2015, Teradyne’s market cap was much smaller at ~$4 billion, highlighting how semi-conductor testing equipment providers can grow into large companies in a relatively short period of time. It is also noteworthy that Aehr’s CTO used to work at Teradyne.
What sets Aehr apart from the competition is that its wafer level burn-in testing equipment can scale better than competing products. CEO Erickson explained that Aehr’s systems are “architecturally different and unique” as the systems test 18 wafers at a time, while competing products test one to eight wafers at a time. This is significant as each test system has the footprint of a Toyota Prius, so customers who are scaling and need to test 500+ wafers a day will need 500 Prius’ of footprint in a wafer fab, which is expensive and difficult to accomplish.
CEO Erickson explained further that “we've made an enormous investment in this platform over the last decade, and particularly with the last handful of years. We have IP and patents protecting that capability, both in the tester and the proprietary contactor that enables it and we intend to defend it.” To give you a better sense of how long Aehr has been developing its unique, proprietary technology, the company received funding from DARPA back in 2001 to develop wafer level burn-in testing, and its first product wasn’t introduced until 2016.
While Aehr trades at a premium to some of its peers, the company’s technology has the potential to be a homerun if it can quickly scale with customers and provide the promised cost savings. If early adopters realize strong cost savings and higher quality products, then this will likely lead to more customers rapidly adopting Aehr’s technology. Furthermore, if Aehr can continue to capture market share in the rapidly expanding silicon carbide and silicon photonic markets, then its market cap will likely grow and mimic some of its other semi-conductor testing peers.
Risks and conclusion
A key risk going forward is supply chain risk, which isn’t unique to Aehr, but nonetheless may impact Aehr’s ability to convert the recent ramp in bookings into sales. The company has $10 million in inventory on its books, which helps get the ball rolling but is well below the $40 million in orders it has received in the past few months. The firm also has just $6.5 million of cash on its balance sheet, which helps explain why Aehr recently announced a $25 million share offering to raise more cash.
CEO Erickson explained on the Q1 Call that customers were happy to see the cash raise as it added conviction that Aehr will be able to have the necessary working capital to support its rapid growth in bookings. However, the capital raise will weigh on Aehr’s stock price in the near term.
In regards to the supply chain concerns, CEO Erickson stated that “Aehr has the manufacturing infrastructure and supply chain in place to ramp to significantly higher revenue levels. We have been ordering long lead components for systems and WaferPak, particularly for the enormous opportunity we see for silicon carbide that is gaining momentum, and we have been able to maintain reasonably lead times to meet customer requests.” The company likely has enough capacity to honor the orders that have come in thus far, and it also has the vendor relationships in place to support a further increase in customer orders going forward.
Looking forward, the company has two strong tailwinds propelling it: silicon carbide and silicon photonics. The recent qualification of Aehr’s lead silicon carbide customer may have been the tipping point that drives rapid adoption of its technology, as early adopters can gain a significant cost saving advantage.
Silicon carbide is quickly becoming the material of choice for EVs, which itself are expected to rapidly grow going forward. Furthermore, silicon photonics is also in the early stages of ramping and has exposure to strong microtrends such as datacenters, 5G and edge computing. Taken together, these two markets should help support sales of Aehr’s test systems, which in turn will lead to recurring sales of Aehr’s high-margin consumables. The company’s stock has increased following strong results, but its valuation is not overstretched. In fact, given the firm’s strong growth rate and forecasted profitability, it appears cheap relative to peers. It is also noteworthy that forward growth estimates do not fully encapsulate sales of consumables, meaning that the company’s true sales multiple is cheaper than reported.
Disclosure: Bradley Cipriano and the I/O Fund own shares in Aehr Test Systems and do not have plans to change their respective positions within the next 72 hours. You can access the I/O Fund’s positions herehere. The above article expresses the opinions of the author, and the author did not receive compensation from any of the discussed companies
Two years ago, the I/O Fund wrote about Apple’s mobile privacy changes and kept our newsletter readers aware of these changes as they rolled out. Going into earnings, we are seeing headlines that Facebook may be affected. We think when Big Tech goes up against Big Tech, that investors should watch the outcomes closely. Our stance for the past two years is that Apple owns the real estate on iOS, and everyone else is renting. The hierarchy is straight forward yet many critics question Apple’s decisions, often feigning concern for the impact to small businesses. We do not think Facebook cares about small businesses at all, per se, but rather about ad dollars.
This upcoming quarter will be the first full quarter to reflect the change. Some models suggest about 7% decline if 20% of iOS users opt-in. The opt-in rates quoted here match what is being reported (about 1 in 5 users opt-in for Facebook to track them). Flurry also stated about 20% were opting in. Meanwhile, according to Bloomberg, some agencies are reporting that companies went from spending “nearly all” of their budget on Facebook to more around two-thirds or half of their budget due to the iOS tracking changes.
The reality is that Apple built the ecosystem and it’s theirs to monetize as they see fit. In this equation, consumers matter too, and data should not have been collected without permission in the first place. Although we’ve been covering privacy concerns since 2014, we specifically called out Facebook in 2018 during Cambridge Analytica to discuss the various ways Facebook was collecting data without permissions.
We’ve also maintained that Apple is running out of near-term growth markets so it makes sense they’re looking for ways to expand their revenue. Below is a snapshot of Apple’s growth pre-Covid in 2019. Due to an increase in time spent indoors, even sleepy segments like personal computers exploded overnight. However, these segments could return to pre-Covid levels (or even lower if consumer hardware saw a pull forward). This helps us to understand Apple’s motivation taking back its real estate. My only question is … why didn’t Apple do this sooner?
Pictured below: 2019 revenue for Apple and it’s iPhone segment
Below are excerpts and links to our previous analysis, which was written for our free newsletter subscribers over the past few years.
Governments Won’t Be Able to Stop Facebook and Google — But Apple Could
In April 2018, Congress tried to piece together how Facebook’s platform works. It ended up being a disaster. Anyone who works in the mobile-ad industry knows that the mobile device, notorious for its massive data leakage, could be used to collect thousands of data points daily to reveal personal thoughts, behaviors and political preferences.
When Facebook CEO Mark Zuckerberg answered a question on how Facebook makes money — “We sell ads, senator” — he wasn’t fooling the ad industry. It’s well aware that Facebook sells audiences and identities, as the company’s ads would be worthless without extracting data points from the mobile device and aggregating them for targeting.
This isn’t your typical targeting of pizza (or beer) ads during football games. This targeting knows you better than you know yourself, as it monitors your actions with data science and look-alike modeling.
The only force that can stand up to the complex tracking methods used by Google and Facebook will be an opposite, yet equal, force. It will not come from governments, which think that paying for search results is the problem. Rather, the problem is the pervasive code and software that continually tracks people, which no competitor can compete with.
Turns out, there is an opposite and equal force in magnitude that has chipped away at the anti-competitive tracking that occurs in the browser with Intelligent Tracking Prevention (ITP). Yet it has not done so on the leakiest device of all: mobile. And that would be Apple.
Facebook and Google aren’t the only companies that track users on mobile and browsers. They simply have software and code in more places. For instance, Facebook’s software is in 32% of the top 500 app market — and up to 800,000 applications. They track billions of non-Facebook users with software that can track you whether you have navigated one of their digital properties or not.
There is no way to opt out of Facebook or Google from tracking you, as their tracking is simply everywhere. In fact, security experts, including Bruce Schneier of the Berkman Center for Internet and Society at Harvard, call such tracking outright surveillance.
The incredible depth of information those giant companies have on mobile and internet users is the “moat” that generates unprecedented cash flow in advertising.
Advertising Stocks Face New, Major Challenge with Apple’s iOS 14
Excerpts:
Tim Cook has publicly criticized [Facebook with its software development kit “Audience Network” installed in 300,000 applications on iOS and Android combined and has seen nearly 200 billion downloads. Google’s AdMob is even worse with installation in 1.5 million applications and 375 billion downloads. (Now consider that users did not authorize or download this software on purpose!)
[Despite Apple’s privacy advertisements] what happens on the iPhone most certainly does not stay on the iPhone. Mobile has become a free-for-all in data collection over the past ten years. The device leaks volumes of information through software development kits (SDKs) installed inside every application. Most applications have 18 SDKs, which extends beyond Facebook and Google to include a mix and match of ad software companies although the most pervasive being Google and Facebook who are inside the far majority due to the depth of their data for cross-targeting.
The concept of Apple pioneering privacy at the client level is not new. Apple began to restrict tracking on the Safari browser through iterations of Intelligent Tracking Prevention (ITP) from 2017 to 2019. As I covered previously in depth, Apple implemented strict requirements, such as having a relationship with the customer within the last 24 hours to place a cookie, and companies have continued to find loop holes.
Unlike cookies on the web, where there is a tag on the browser, mobile identifiers have much stronger tracking capabilities. Apple’s IDFA enables the following: user tracking, marketing measurement, attribution, ad targeting, ad monetization, programmatic advertising including DSPs, SSPs and exchanges, device graphs, retargeting of individuals and audiences.
What investors may not realize is these advertising cash machines are largely dependent on tracking software for the high CPMS (cost per thousand views) and CPIs (cost per install) they charge because they can track actions on a granular level even days after a mobile user has seen an advertisement. The mobile users are not aware they are being tracked by many companies they do not have a first-party relationship with (but the developer or publisher does). These developers and publishers must now obtain permission. Without permission, the inventory on mobile becomes less valuable.
Mobile applications, such as Spotify, Uber, Lyft, and mobile gaming, for example, are also dependent on the ability to track and identify cohorts for user acquisition. This is one reason we see the top line grow rapidly in ridesharing at the expense of the bottom line; these companies are crunching customer acquisition costs and lifetime value (LTV) across specific demographics and then using lookalike modeling to target the demographics with the best LTV.
Facebook’s Surveillance-Like Software is Called Audience Network
The betrayal was two-fold. On one hand, a first-party data company boldly entered the third-party data marketplace to broker data, risking the trust of its social media users. Secondly, Facebook did everything in its power to make sure social media users would not find out. Audience Network, which fuels a substantial portion of Facebook’s revenue from the social network, has not been disclosed to the public to this day. It is eerily absent from PR releases and discussions around privacy. Unless a Facebook user was a detective, they would have no reasonable way to know that Facebook operates Audience Network and is selling private data across hundreds of thousands of applications at an estimated 40% of the app market.
Built in 2012, Audience Network went live in 2014, and caused Facebook to stage a remarkable turnaround on the stock market. Facebook has posted consistent returns ever since. This is in marked contrast to the years prior to Audience Network, between 2012 and 2014, when Facebook faltered quarterly, often losing 50% of its stock value due to frequent, disappointing earnings.
Disclaimer: Beth Kindig and I/O Fund does not own stocks mentioned in this article. This is not financial advice. Please consult with your financial advisor in regards to any stocks you buy.
There are three important trends we weave together in this report to draw conclusions around potential winners in big data and analytics. We’ve recently covered MongoDB, Elastic, and we’ve discussed Confluent. What do they all have in common and why are these companies important right now? That’s what we aim to answer in this write-up.
Before we go into where we are with big data and analytics right now, I’ll quickly touch on cloud IaaS and especially why hybrid and multi-cloud are leading this space and why investors should not be concerned with tech giants that offer competing products in the data and analytics space.
When we talk about Big Data, the main driver is machine learning, which is performed through supervised learning with the use of historical data, or unsupervised learning, with clustering models and associations to identify rules. There is also reinforcement learning that is trained from feedback. Here are the three trends we are going to weave together to form a full picture of Big Data and Analytics.
The migration to the cloud — but more specifically multi-cloud and hybrid
Why multi-cloud drives demand for best-of-breed, i.e., generally speaking, we do not need to be overly concerned when tech giants that offer competing products
How Apache Spark helped catalyze the AI/ML market with efficient data processing
How we plan to invest right now given that #2 and #3 are prepping the market for us
Hybrid and Multi-Cloud are Driving the Cloud IaaS Market:
For cloud IaaS, we don’t want to only focus on CAGR but also the budget allocation that cloud IaaS is capturing. According to IDC, the IaaS market will reach $112.9 billion at a CAGR of 11.3% through 2025 and will account for 66.1% of total compute and storage infrastructure spend. Two-thirds of IaaS spend is on the public cloud.
To compare, the on-premise market (i.e., not hybrid) will grow at 0.3% CAGR for a total of $57.9 billion. According to the most recent Denodo survey, hybrid cloud drove 35% of the workloads worldwide. Private cloud expanded from 16.6% of workloads to 24% percent of workloads worldwide. Meanwhile, the public cloud had flat growth.
Hybrid cloud is a mix of public and private clouds or a mix of cloud and on-premise. Enterprise companies that choose hybrid deployments are motivated to not share intellectual property or data with a vendor, known as data residency, plus other security implications that come with storing data on another company’s servers. Other companies find moving to the cloud to be time and resource-intensive and prefer to keep some workloads on the servers they own.
Recently, a report came out that repatriation, or moving some workloads back to on-premise, has resulted in quite a bit of cost savings for companies like Dropbox, Crowdstrike and Zscaler, who use hybrid approaches. The report is quite surprising as the conclusion is that $100 billion to $500 billion in market value is lost on cloud deployments in terms of margins. One use case that is detailed is Dropbox, a company that reported savings of $75 million in two years after repatriation, which in turn, helped the company’s gross margins increase from 33% to 67%. Meanwhile, companies like Asana and Datadog spend about 60% of their revenue towards committed cloud spend. This report, among others, shows why hybrid is likely to be the chosen deployment for many enterprises into the near future.
We had previously formed a Microsoft thesis in 2018 based on the trend towards hybrid cloud and why a focus on a hybrid strategy for governments and enterprises was important to Azure’s growth rate. Microsoft is especially well suited to serve the hybrid market because of the company’s deep roots with on-premise enterprise software. When the I/O Fund first covered hybrid cloud as a major driver of cloud IaaS in 2018, Amazon’s AWS did not even have a publicly available hybrid product. The company later publicly released Outposts in 2019 to compete with Azure. If you want more information about how these two compete on hybrid on-prem deployments specifically, the in-depth analysis I published in the past is found on Seeking Alpha and also Forbes.
Multi-cloud refers to using more than one cloud provider, which is usually done to avoid vendor lock-in and to choose best-of-breed products. It also helps to avoid downtime should one cloud provider go down or become overwhelmed with demand.
Multi-cloud is the dominant strategy today and is used by 80% to 90% of organizations. In 2019, Gartner stated 81% of respondents were using two or more cloud providers. The top reason was to avoid vendor lock-in by the “megavendors.” Therefore, this is why investors should not be concerned with tech giants offering competing products in the data and analytics space. The far majority of companies are taking strides to avoid vendor lock-in as multi-cloud technically requires more work yet increases agility and flexibility. The end result is these companies will use best-in-breed products.
According to IBM, 98% of companies plan to use multiple hybrid clouds and 85% operate in multi-cloud environment. There is substantial evidence that organizations are preferring a mix of cloud providers. Most importantly to our thesis and this particular analysis, only 40% use management tools and/or have implemented DevOps practices. The migration to the cloud was happening slowly over time and this migration is under-served in terms of management tools, data and analytics. This would be a sufficient tailwind on its own yet we also have the additional tailwind of data-intensive industries that are moving into machine learning.
The motivation behind cloud IaaS growth and especially hybrid and multi-cloud growth is partially driven by the need for analytics, and also newer trends, such as stream processing. Stream processing is a continuous stream of events that is processed in real-time as it’s received. This allows applications to respond to events as they occur. It combines real-time analytics, inferencing and monitoring to achieve things like optimizing transportation routes, understanding traffic patterns, anomaly detection in cyber security, making real-time predictions powered by machine learning, and even location-based advertising.
In terms of architecture, we’ve covered how microservices and containers are also driving the multi-cloud trend as microservices often span multiple clouds. You can find this write-up here on Forbes and also here on Medium, where we discussed a background on Google Cloud and how the company was the first to automate orchestration across containers. This write-up provides a great overview of where the major cloud IaaS providers are today and where they might go next strategically speaking.
Big Data and Analytics will Explode because of AI/ML Applications
There is an oft-quoted statistic that 90% of the world’s data was created in the last two years – and this stat is from 2018. The world produces 44 zetabytes of data across the digital universe as of 2020 and there is expected to be 200+ zetabytes of data in cloud storage by 2025. Each zettabyte has 21 zeroes or is 1,000 bytes to the 7th power. By these estimates, we can expect to see up to 5X growth specifically in data centers. Statista places the number at 181 zetabytes by 2025 up from 64.2 zettabytes in 2020.
In regards to data integration in the cloud, this spans from data lakes, to ETL pipelines, cloud data warehouses and object storage. Data fabrics and data virtualization is key to both hybrid and multi-cloud strategies.
Here's how Datadog’s CEO describes what is going on in terms of big data in the most recent earnings call: “it's almost a given that there will need to be a different way of charging for capturing some of the value provided to customers that can't just be attached to the straight volumes of data that are being exchanged because those volume of data are exploding exponentially while our customers' revenues are not going to explode exponentially.”
Generating the data is not the issue (clearly), and distributed data storage has been largely solved with Hadoop. I think it’s worth going through what Hadoop is and how it came about, and then we can look at how Apache Spark helped accelerate data processing, including for Machine Learning. Notably, most open-source projects are not “easy” and this is why companies do well that simplify how to work with Apache Spark and other frameworks, like Kafka.
Background on Hadoop and Data Storage:
Hadoop became instrumental in helping companies store large amounts of structured data, semi-structured and unstructured data through distributed storage and compute. The result was that data storage became cheap enough to retain any/all data that was generated rather than only the essential data due to its distributed file system. The distributed file system was designed to store and process billions of search engine pages across thousands of nodes. The project was created in 2006 by a team of engineers at Yahoo, who had worked previously on a search engine in the early 2000s with the goal of indexing 1 billion pages.
You can think of search engines as some of the first projects that needed to utilize Big Data. The original search engine project “Nutch was limited to 20-to-40 node clusters, and for this amount of data, more clusters were needed. At Yahoo, the team separated the distributed computing parts from Nutch and renamed the project Hadoop, which successfully worked on thousands of nodes. Parallelism was key for the data processing model as Yahoo’s algorithm would need to be run on multiple nodes at the same time and it had to scale linearly. It was then released in 2008 as an open-source project with up to 4000 nodes with distributed capacity with contributors such as Facebook and LinkedIN.
Distributed systems and parallel computing didn’t begin with Yahoo, of course, it began with Google. The paper “MapReduce: Simplified Data Processing on Large Clusters” is considered a defining moment in how programming models handled large data sets. MapReduce was a key moment because it was specifically designed to handle Big Data in terabytes and petabytes due to its framework for parallel computation using a key-value pair.
By 2012, Hadoop’s clusters were up to 42,000 nodes and the number of contributors had reached nearly 1500. Apache Hive is a ETL and data warehouse tool that uses SQL, but Hadoop can manage and process large volumes of data that are structured, unstructured or semi-structured data depending on the database that is chosen. Therefore, you can use many tools with Hadoop, such as Spark.
Background on Apache Spark and Data Processing for Machine Learning:
In 2014, Apache Spark was released which took over the MapReduce model primarily because of its speed. By working with data in-memory, the parallel processing framework can push queries 100X faster and on-disk queries run 10X faster. After the extract, transform and load the data (ETL) process, with Spark you can run a training algorithm on the same in-memory data. This helps Spark reach peak performance over competitors for ETL and relational queries, but also for machine learning. Spark’s goal was to become (and now remain) the general platform for distributed programmers where many specialized systems have one interface and one system to install and manage. Apache Spark also reduces code volume by using APIs for Scala, Java and Python. The framework offers a unified API for fault-tolerant stream processing, which reduces the number of APIs to learn. Spark ML and SparkML are the two APIs that are offered for machine learning pipelines.
Hadoop helped solve some of the data storage issues and reduced the cost for expensive storage and compute. Therefore, the next issue is who can work with these databases and can this be simplified. Apache Spark simplifies who can work with the framework by supporting libraries, which can be executed to interact with data shared across many libraries. The data processing engine is extremely fast because it processes and keeps the data in-memory without reading or writing to disk. This has resulted in Apache Spark becoming popular for machine learning and AI applications with the support of Apache’s very large community of contributors.
Overview of Public Companies in the Big Data and Analytics Space
Databricks and Snowflake:
I’m starting with Databricks and Snowflake simply because we discussed Apache Spark in this analysis. The founders of Apache Spark are from Berkeley and later went onto become the founders of Databricks. We covered this company in-depth on our Snowflake analysis because we feel this is Snowflake’s strongest competitor (i.e., not traditional SQL warehouses or Big Tech). Databricks is not public right now but plans to go public soon.
Here is a summary of the explanation we published in April as to how these two companies compare:
The major difference between Snowflake and Databricks from a customer standpoint is that Snowflake is laser-focused on the public cloud/cloud native while Databricks is differentiated in that it can build information pipelines across silos, including on-premise and hybrid architectures. As we know from this analysis, hybrid is key moving forward.
Snowflake's main value proposition is to reduce the time required to prep and monitor data so that a customer does not need to manage software or hardware. Even if a team has the technical skills, they may not want to spend the time required for Databricks, which is perhaps one reason why Snowflake is reporting decent growth in the Fortune 500 and other key accounts.
The architecture of a data lakehouse allows for business intelligence and machine learning through a more open paradigm. The idea is to combine the best of data warehouses and data lakes to span unstructured and semi-structured data while keeping costs low. By combining both, teams can move faster and without duplicating the data. This is a key benefit to Databricks DeltaLake, and this is especially important for data analytics and machine learning. With that said, Databricks is more advanced and expert-level.
I want to point out that Snowflake is very clear as to why it's done well – which is that it handles migrations to the public cloud from legacy on-premise systems better than the competitors. Snowflake's priority is to compete with other SQL databases right now, although the company will need to eventually compete with Databricks. Management has discussed rolling out support for unstructured data, for instance, but no timeline has been set.
Looking longer-term, what Snowflake needs to answer is how will it compete with Databricks on machine learning? Databricks is superior here for ML as it’s built on top of Apache Spark and supports Spark, Python, Scala and also SQL. This was discussed in the thread on the forum here.
The forum thread points out that Databricks is more complex to upload the data, monitor and manage, but there are benefits to going through this hassle. One of the primary benefits is support for Python and Scala, which are programming languages for machine learning. For now, you must use an outside vendor or tool as connectors or integrations in order to support these programming languages and libraries with Snowflake. It’s also worth mentioning that Databricks is cheaper for processing a lot of data at petabyte scale.
Growth is the great equalizer when comparing products and my preliminary understanding is that Snowflake is growing much faster than Databricks and expects to continue to outpace the competitor. I will need to look into Databrick’s financials and see an earnings report or two to determine more about the competitor’s sustained growth rate.
What I find to be very intriguing is what Snowflake will do to compete on ML. This gap in product capability is not lost on the Snowflake team and management. Being laser-focused on the public cloud/cloud-native lends itself well for Snowflake to compete here theoretically, yet its laser-focus on SQL is getting in the way strategically speaking. The company is aware of this and plans to roll out support for unstructured data.
We have two strong products here yet the valuation on Snowflake is stretched and I imagine Databricks will be, too. It’s rare to see a company sustain higher than a 40 or a 50 forward P/S for an extended period of time. Right now, Snowflake is at a 79 forward P/S.
MongoDB:
Big Data applications require a flexible data model, which NoSQL supports. MongoDB is a database that can handle unstructured and semi-structured data, whereas SQL competitors require data to be structured and stored in tables. The predefined schema of the relational database is correlated due to common characteristics. SQL is well-supported as the original database management type yet NoSQL is also reaching critical mass.
The reason NoSQL has risen in popularity is because as data grows, there are more data types to work with outside of Excel spreadsheets/CSV or tabular structures. MongoDB and its competitors are a good choice for Big Data because NoSQL databases can process unpredictable and unstructured data. The most popular types of NoSQL databases include graph, key-value pairs, columnar and document.
Moving forward, we think NoSQL is going to take more market share, simply because it saves steps when dealing with Big Data as the unstructured data does not need to be converted and this is preferred for some machine learning models. This is why NoSQL is used by companies that generate the most data, like Amazon, Facebook, LinkedIN and Google. The extra bonus is that the JSON documents in NoSQL databases can be prepared for machine learning. Because you do not have to define a schema, this allows data to be directly loaded from any new source without changing lines of code. SQL is used in training machine learning models with most of this data coming from on-premise servers. Therefore, the migration to the cloud and various types of data that are generated is also helpful for companies like MongoDB in growing market share. This is because the cloud produces various forms of data.
MongoDB has a query language and secondary indexes for specific values to filter, sort and aggregate data. The leading NoSQL database also allows for the storage and retrieval of trained models as JSON documents. In this case, you can query MongoDB to pull up a previous model.
In the multi-cloud trend, MongoDB is a leader here as the company was the first cloud database to run applications simultaneously on all major cloud providers. The multi-cloud clusters allow developers to deploy applications across multiple cloud providers without having to manage the complexity. In addition, the technical team at MongoDB maintains that you can forego Hadoop and Spark, which requires complex functions and logic, and instead rely on Tensorflow.js, MongoDB and a browser for the same level of machine learning but with less complexity. In an example, a MongoDB representative was able to write a ML program with 88 lines of code. With that said, NoSQL requires more expertise than the universal language of SQL.
The takeaway is that Big Data companies prefer NoSQL for many reasons, and we think in the era of ML and AI, that more companies will lean towards having similar requirements as Big Data companies. This isn’t to say that SQL isn’t alive and well due to the sheer amount of support for structured data across various database systems. Financial transactions for instance fit well into SQL. This is not a “SQL will die” discussion, instead it’s a “NoSQL may see a bigger market thanks to big data and the sheer amounts of unstructured and semi-structured data that will continue to grow” discussion.
Although the SQL and NoSQL debate has lingered for some time with SQL being the leading database today, requirements may change and we think MDB is positioned well for this shift.
Also, refer to the fact that MongoDB is fifth in terms of database market share yet is tied for first place for most wanted database skills among software developers. Notably, MySQL and Oracle are the top database systems globally yet MySQL is fifth in terms of most wanted database skills. The demand for talent is typically an important indicator of where we are now and where the puck is going.
The founding team of Apache Kafka worked at LInkedIN before leaving to start Confluent. Apache Kafka is used by thousands of companies for message streaming, such as LinkedIN, where a publish/subscribe model allows applications to share and create data in a serverless and microservices architecture. What Kafka solved for is the ingestion of events data in real-time and with low latency.
At the time that Kafka was developed, LinkedIN was ingesting 1 billion events a day. The company is now ingesting 1 trillion per day. Kafka does this through a log that writes messages to a topic and is able to retain messages for a long time. Kafka is also used in stream processing by parallelizing the pipelines. Kafka Streams were built to increase simplicity while retaining the same amount of performance as a Spark streaming job.
As with Spark and other open-source projects, there is a marketplace for making the frameworks easier to use. Confluent Kafka opens up the amount of data that can be integrated, for example, to combine transactional data (orders, inventory) with sentiment-driven data (likes, page clicks). This helps with predictive analytics and also machine learning because the “data flow” allows for algorithms to work as they are intended to. This is what is meant by the title slide of the S-1 filing “Set Data in Motion.” In order for data to be in motion, Confluent’s platform connects data from many different sources.
The end result for Confluent is that the company allows large amounts of data to be moved very quickly. This is needed for machine learning algorithms that are very data hungry. Kafka can be paired with Apache Spark and Apache Samza to route data and then load it into ElasticSearch, for instance, so it’s a bridge (or a nervous system according to Confluent’s marketing department).
The goal of Confluent is to reduce operational complexity. In the case of Kafka Streams, this is done by not requiring a cluster to be spun up, offering a single framework for streams of events, and reducing the number of pieces in a stream architecture. Confluent Cloud is growing rapidly at 200% year-over-year, primarily driven by event streaming.
Please note, that Confluent is on a partial lockup schedule. The partial lockup dates are 15% on the day of the IPO (June 24th), 25% on the second day of trading (August 09th) after the Q2 earnings, with the remaining at the earlier of the second day of trading after Q3 earnings and 181 days of the IPO.My note: Already up to 40% of the shares have already been released by the eligible employees. The full lock-up expiry is between November and December..
Elastic:
Elastic is a best-of-breed search company that has other benefits, as well. Elasticsearch is the core product that allows for the searching, storing and analyzing of data. This allows developers to build search features that pair Uber passengers with drivers, recommend grocery items on Instacart based on your history, match online data profiles for Tinder, or log events for Fitbit at a rate of 250,000 logs per second. In addition to searching and storing data, Logstash and Beats are ingestion tools to ingest data from applications and to query external systems. Kibana is an open-source tool for visualizing the data. We’ve covered Elastic Stack in more detail here.
Since 2018, the Elastic License has been free and open source with paid proprietary features. As Bradley detailed in this write-up, Amazon began to profit from Elastic’s open-source software and did not contribute back. According to Elastic, over 90% of new downloads choose Elastic’s License. As of January 2021, the company dual-licensed Elasticsearch and Kibana under SSPL or “Server Side Public License,” which requires Amazon or any others to publish modifications and the entirety of their source code. We think the multi-cloud trend is one reason that Elastic has been able to overcome Amazon as the primary driver is to avoid vendor lock-in. Notably, Elastic is cloud neutral so it does not rely on any specific external services for machine learning like AWS’s OpenSearch. Basically, this goes back to the points we made about multi-cloud earlier in this analysis.
We also discussed Elastic’s move into XDR is important because security is a primary concern for those who are on multi-cloud deployments. The SIEM and XDR space is not without its competitors yet it could be Elastic’s combination of already having ingestion tools for thousands of applications and sensors that lends itself well to monitoring and detection. SIEM is security, information and event management while XDR stands for extended, detection and response (XDR). SIEM was first used as a compliance product and often works alongside endpoint and network security products in order to offer a narrower yet deeper set of activity. This last piece has become critical over time. For Elastic’s product, XDR builds on the SIEM and EDR (endpoints) combination for more accuracy and applies machine learning models to detect anomalies.
Where there is data, there will be new opportunities for growth as the AI/ML landscape goes from nascent to mature (i.e. not all uses cases have arrived for big data and analytics companies). Due to Elastic being essentially a pretrained model for extracting keywords and synonyms and “term co-occurrences”, it lends itself well to natural language processing (NLP). With Elastic, terms can be filtered by significance and offer out-of-the box shortcuts to Python with its REST API. Cognitive search is a new form of search that uses AI to improve search queries and to extract information from multiple data sets. Cognitive search can combine a traditional search engine with NLP to extract more useful information since keyword search is limited in the variety of data that can be searched. Cognitive search uses machine learning algorithms for its greatly improved search results and will be a $6 billion market by 2025. We think it's impressive that Elastic was named a Leader in the Gartner Magic Quadrant for cognitive search in the first year it was added as a new entrant, blowing past Microsoft, AWS and even Google.
Conclusion:
I wanted to cover Big Data and Analytics broadly and horizontally rather than vertically by company because it paints a better picture of what we are positioning forand why. It’s easy to get lost in the jargon when discussing companies individually especially with technical companies like these. But what really separates each of them? We think the side-by-side comparison can be more conducive at times when setting up a microtrend.
We had a few goals with this analysis that I hope we accomplished:
Bring to your attention this trend (and the common thread) and pull-out names from the general “cloud” list to discuss why they may have a unique catalyst. There will be many winners in this space and we are limited in terms of number of positions we can enter. It’s easy to get caught up in “stock picks” yet we also want to offer you microtrends to help inform your individual portfolio decisions.
We think big data and analytics from best-of-breed companies could become a solid post-covid cloud play due to the sheer number of companies that migrated to the cloud yet have multi-cloud and hybrid deployments
Third, I want to make sure and elaborate on where the MongoDB, Confluent and Elastic positions are coming from that the I/O Fund recently entered. We offer deep dives on companies but we also want to anchor our readers with the underlying microtrends that we are investing in. For instance, Snowflake is a great choice, yet the valuation is high and that range above 50 has not treated us well in the past (i.e., personal choice). Perhaps for your investment profile, you prefer Snowflake right now, etc.
This is a big space and it’d be impossible for me to cover everything but we pulled out the critical pieces. We think it’s important to simplify the key drivers of a microtrend and illustrate the ways that specific companies are serving the trend. You can expect to see MongoDB and perhaps Confluent added to the LTBH portfolio as the thesis should take about 3-5 years to fully play out. The main thing to know is this means we will have to remove a name or two from the current LTBH portfolio. We will keep you in the loop as we weigh these decisions.
Despite Bitcoin’s recent decline, volatility is actually decreasing and the asset is beginning to stabilize in terms of historical performance. The last crypto peak to trough in 2019 saw a decline of 84% in price while this sell-off’s peak to trough was 55%It took three years to recover the all-time high from 2017 for Bitcoin yet the I/O Fund thinks Bitcoin will recover its current all-time high quicker this time as the long-term pressure for Bitcoin is up.
Naturally, with the steep drawdown of 55% (yet the very promising future of crypto – see our price targets below) we were wondering how crypto trading apps are faring after the Q2 sell-off now that Coinbase and Robinhood have gone public, two of the biggest names in the space. Below, we highlight data that is offered by Apptopia (and is also available on the Bloomberg Terminal) to show that in some cases crypto trading apps are down sequentially while others are down even year-over-year. Robinhood is stock trading app too, of course, yet we think any Q3 declines will be largely contributed to crypto trading. We also discuss Voyager, a company we have done a deep dive on before and currently hold a position in.
Two Popular Crypto Trading Apps Went Public in Q2
The I/O Fund figured it would be a good idea to check on the health of crypto trading apps to see how the asset class underperformance may have affected crypto trading apps. When Bitcoin and crypto go through a rally there will be a correlation with the crypto trading apps, yet how much of a fall-off do crypto apps see when there is a sell-off?
Coinbase and Robinhood seized the crypto rally to go public as the companies posted 1080% growth and over 309%+ revenue growth, respectively in the March quarter. You’ll see below in the downloads and sessions that there was unusually high activity during the period the two companies went public.
Coinbase was listed on April 14th at $381. The shares were off to a good start on the day of direct listing, but the shares sold drastically a month later. Lock-up periods are standard for Initial Public Offerings (IPOs). However, Direct Listings usually don’t have lock-up periods. There could be exceptions like Palantir, which despite the direct listing, had lock-up restrictions. Coinbase, which did a direct listing, did not have lock-up agreements. This is important because shares can come under pressure following a soft quarter.
Robinhood shares were listed on July 29th at $38, which was at the lower end of the offering range. The shares peaked at $70.39 and are currently trading above the IPO price at $47.15. Many IPOs have 180 day lock-up periods yet Robinhood has a partial lockup schedule. In Robinhood’s case, 15% of the employees were able to sell their shares on the day of the company’s listing. The next 15% of the shares will be eligible for sale on October 27, 2021. There is mention of a final lockup expiring on December 1st in the amended S-1 filing here. There will likely be volatility during this period as insiders tend to sell after lockups expire. A soft quarter or two could exacerbate this. Doge Coin also peaked around $0.60 and is now trading around $0.20. Per Robinhood’s financials, this coin drives 62% of Robinhood’s crypto trading activity.
Quick Glance at Q2 Financials for Crypto Trading Apps:
Before we look at the current downloads and sessions for Q3, we want to review the Q2 financials for Coinbase, Robinhood and Voyager Digital.
Coinbase reported $1.3 billion in revenue in 2020 compared to $533.7 million in 2019 for growth of 143.6%. Revenue growth in Q2 was impressive at $2.2 billion compared to $186.4 million for 1080% growth. A similar trend was seen in H1 2020 compared to H1 2021 for 936% growth. Net income grew in line with the top line at net income of $1.6 billion and adjusted EBITDA of $1.15 billion compared to a net income of $32.3 million and adjusted EBITDA of $61 million for the same period last year.
Retail Monthly transacting users (MTUs) rose to 8.8 million in the recent quarter when compared to 2.8 million at the end of December 2020 and 1.5 million for the 2Q 2020. Verified users were 68 million. Absent a rally in Bitcoin and crypto, these will be tough comps to clear in future quarters this year. The data from July showed that retail MTUs came at 6.3 million. The management expects MTUs and total trading volume to be lower in the third quarter when compared to the second quarter with the company stating, “as volatility and crypto asset prices are highly correlated with trading revenue, the crypto market environment heavily influenced our Q2 financial results” citing declines of 45% in Bitcoin and Ethereum.
Coinbase emphasized they have 9,000+ institutional customers and 160,000 ecosystem partners. Trading volume is primarily driven by institutions at $317 billion compared to $145 billion from retail (about two-thirds) with trading volume at 24% for Bitcoin, 26% Ethereum and 50% other crypto assets. In terms of assets on the platform, the mix is more equal at $88 billion for retail and $92 billion for institutions with 47$ Bitcoin, 24% Ethereum and 25% other crypto assets.
The company derives its major revenues from the transaction revenue. This is correlated with the trading volume. The company plans to reduce the focus on the transaction revenue since it’s volatile and focus on the subscription & services revenue in the long run. The subscription & services revenue include custodial fees, blockchain rewards which includes staking revenue, earn campaign revenue, interest income and other subscription & services revenue. The third classification of revenue is other revenue which includes crypto asset sales revenue and corporate interest income.
Robinhood grew Q2 revenue by 131% YoY to $565.3 million. Revenue growth has been strong yet may have peaked in Q1 as growth has been slowing down. For the full year 2020 it rose by 245% to $959 million and in the 1Q 2021, it was up 309% to $522 million. Transaction-based revenue grew by 141% to $451.2 million, net interest revenue grew by 69% to $67.7 million, and other revenue grew by 177% to $46.5 million. Looking deeper into the transaction-based revenue, options revenue grew by 48% to $165 million while cryptocurrencies revenue increased to $233 million from $5 million in the same period last year. The equities transaction-based revenue dropped 26% to $52 million.
As stated above, Dogecoin accounted for 62% of crypto trading in Q2, which was up compared to 34% in the first quarter. Due to the price decline in this asset, it’s unlikely Q3 will comp well with Q2 partly due to this alt-coin.
Operating expenses increased 169% to $500.7 million. Notably, technology and development expenses increased 248% to $156.3 million and operations expenses rose 232% to $101.1 million. Due to the strong trading activity, the company procured additional cloud infrastructure, which increased technology expenses. Operations expenses were high due to the increase in the headcount of customer support staff.
The company also recorded stock-based compensation in the current quarter and expects to record a charge of $1 billion in stock-based compensation for RSUs related to the IPO in the third quarter. It reported a net loss of $502 million in the recent quarter and adjusted EBITDA of $90 million. In contrast, the company reported a net income of $58 million and adjusted EBITDA of $63 million in the same period last year.
According to the management, due to the seasonal nature of the business, trading activity has been generally strong in the first half of the year. On similar lines, they expect lower trading activity in the third quarter. The sessions and downloads confirm that Q3 will be much softer than Q2. At the end of June 2021, the company had about 22.5 million net cumulative funded accounts compared to 18 million at the end of March 2021 and 12.5 million at the end of December 2020. The customer growth has been strong due to the strong word-of-mouth referrals yet appears to have declined for Q3 (see below).
The company earns the majority of its revenues from payment for order flow (PFOF). This is a method in which the brokerages like Robinhood receive compensation for routing orders to market makers. The transaction-based fees represent 81% of the total revenues of the 1Q 2021 and 80% of the 2Q 2021. Notably, there are conversations going on in Congress about the risks and SEC might consider banning or putting restrictions on payment for order flow.
Similar to its peers, Voyager Digital demonstrated strong revenue growth in the first half of the year. The earnings release for May shows Q3 FY 2021 revenue came in at $60.4 million, up from $3.6 million in the previous quarter fiscal Q2, representing 21,000%+ growth (yes, you read that right). Fee revenue was $53.7 million and interest revenue from custodians was $6.7 million. Operating profit was $30 million for the 3Q FY 2021. There was a press release in July that showed revenue for fiscal Q4 ending in June in the range of $103 million to $107 million, up from less than $1 million in revenue in the year-ago quarter. Sequentially, this represented over 65% growth. Total funded accounts have exceeded 665,000, and total verified users are more than 1.75 million. You can read our full analysis here.
According to Steve Ehrlich, CEO and Co-founder, "Our June quarter reflects continued growth of our platform, with revenues up more than 65% from the March Quarter. Although we have seen a significant decrease in crypto market volume since mid-June we continue to see significant net new funded account growth, net asset inflows, and consistent basis points on spread revenues on our platform continue through today."
The Impending Question is Performance in Q3 and Subsequent Valuations:
Crypto trading apps may be the hardest vertical in tech when it comes to forward guidance as crypto is extraordinarily volatile and trading volumes can greatly fluctuate. The I/O Fund uses app data primarily to see if a company is trending up or down. We do not use app data to predict exact numbers. Please also note, that companies often report various key metrics, such as monthly transacting users (MTU) rather than downloads or sessions. Therefore, Apptopia provides an important glimpse on app activity, however, we are not making predictions on what Q3 earnings will report rather we track the overall trend.
Coinbase’s provided a glimpse for July, stating that retail MTUs were at 6.3 million compared to 8.8 million in the previous quarter. The company stated they believed total trading volume would be lower in Q2 compared to Q3. Apptopia data is showing roughly 5 million downloads with still two weeks to go in the quarter or a decrease of roughly 60% if we factor in the additional two weeks that remain in September.
Here is a glimpse of the sessions which show a similar trend as downloads are indicating more of a 40% decline (roughly speaking if we figure a total of 550,000 with the remaining days). In both cases, Coinbase is doing well year-over-year although the growth has tapered off from the 900% to 1000% range to what may be more in the 100% to 200% range year-over-year. According to analyst consensus, there have been 8 downward revisions with revenue estimated at $1.46 billion in the upcoming quarter down from $2.23 billion in the previous quarter. EPS estimates are currently at $1.4 billion compared to $6.78 billion.
Robinhood did not provide guidance but issued the following statement in August: “For the three months ended September 30, 2021, we expect seasonal headwinds and lower trading activity across the industry to result in lower revenues and considerably fewer new funded accounts than in the prior quarter.” In this case, it’s looking certain that Robinhood will report a steep sequential decline and the company is on track to also report a decline year-over-year unless there is a catalyst in the next week or so.
Sessions look stronger than downloads with Robinhood showing slight year-over-year growth. The number of upward or downward revisions available is not available for this newly public company, yet the revenue estimates are at $427.3 million compared to $565 million last quarter. EPS estimates are at ($0.34) compared to $0.18 last quarter.
Voyager released an update for the quarter ending in June but did not offer forward guidance on the press release. Year-over-year is still extraordinarily strong and Voyager is also the strongest app between the three in terms of downloads sequentially, as well. We discuss the key differences between Voyager and Coinbase in a previous analysis here
The Investors Presentation shows an increase in verified users from 1.75 million reported in fiscal Q4 to 2 million on September 7th. Sessions show similar information as downloads, which is very strong year-over-year activity with a decline sequentially although much less impact compared to its peers. Voyager Digital is listed on the OTC market and does not have analysts covering it at this time.
Conclusion:
We don’t expect crypto trading apps to maintain peak traffic, yet as the market attempts to price these apps with no forward guidance, we could see some volatility. Because we track Bitcoin and other alt-coin holdings very frequently, we performed this research to check on the health of Voyager Digital, another holding at the I/O Fund. Due to its strong retention and rock-bottom valuation, we will continue to hold Voyager. The issue we see with Coinbase and Robinhood at this time are the lockup expirations (or having no lockup for Coinbase) combined with the sequential weakness and the likelihood the two companies can’t provide adequate forward guidance. We are in the early days for crypto so anything could happen, and stories can certainly strengthen. However, absent a strong bitcoin rally, Q3 looks weak for the bigger players yet comparatively strong for the small cap Voyager.
As for Bitcoin, the I/O Fund has been covering this asset for the public markets since 2019 when we added a position alongside other tech stocks to a portfolio when we called for a market cap of $1 trillion. We did not budge on this target even when the asset dropped from the $10-$13K region to the $4K and $7K region. Today, our price target is $120,000 to $160,000. We are not financial advisors, rather we perform deep drive research for our own positions and share our conclusions. Our first published piece on crypto was in 2013 when I published a guest blog from Chris Larsen of Ripple on my early-stage tech blog (now archived). Here is a snapshot of our trading history during the most recent sell-off. These are verified through real-time trade notifications sent to our members at the time of the trade.
In early 2021, we warned our readers that a top was forming in Bitcoin. With our initial downside targets showing a likely bottom in the$37,000-$22,000 region, we cut our position in half, alerting our readers that Bitcoin was in a complex topping process, and to be prepared. However, we also stated, and still believe, this drawdown is part of a much larger uptrend. This last point is key.
As stated in the article, Beth Kindig and I/O Fund currently own shares of Voyager Digital and Bitcoin. This is not financial advice. Please consult with your financial advisor in regards to any stocks you buy.
Royston Roche contributed to this article.
Please note: The I/O Fund does not make earnings calls and we do not "play earnings." There are many things that can be reported to affect a stock beyond downloads or sessions. Gross margins, EPS beat or miss, etc, will not be reflected by downloads or sessions alone. We pull data to reduce risk in positions we already own and share the publicly available information with our readers. Please consult your personal financial advisor before buying any stock. You can read Apptopia’s response to the SEC action with AppAnnie and how the company provides quality information for the public markets.read Apptopia’s response to the SEC action with AppAnnie and how the company provides quality information for the public markets.
Webinar Invitation: September 24th at 1:30 PM PST (US Time zone)
In this market update webinar, we will discuss the price action and setups for ZM, DOCU, SE, BTC, ZI, SHOP, ROKU
When: Sep 24, 2021 01:30 PM Pacific Time (US and Canada)
Topic: Market Update – Broad Market, ZM, DOCU, SE, BTC, ZI, SHOP, ROKU
Please click the link below to join the webinar:
https://us02web.zoom.us/j/84450009933
Passcode: 13245
Or One tap mobile : US: +12532158782,,84450009933# or +13462487799,,84450009933#
Or Telephone: Dial(for higher quality, dial a number based on your current location): US: +1 253 215 8782 or +1 346 248 7799 or +1 669 900 6833 or +1 301 715 8592 or +1 312 626 6799 or +1 929 205 6099
Webinar ID: 844 5000 9933
International numbers available: https://us02web.zoom.us/u/keb315YvaK
Reposting the same post from the forum. This is a quick post – I will follow up with a more detailed blog post later but we wanted to get this stock on your radar.
Aehr Test Systems (AEHR) reported strong results today:
-Sales up 181% YOY
-Bookings up 263% QOQ
-Backlog up 21,500% QoQ
We think the story is just now getting started. Below are some notes on what the company does, what the opportunity is and recent results.
Aehr Test Systems makes silicon carbide testing equipment. This year in July, AEHR announced strong Q4 FY21 results as orders for its silicon carbide testing equipment increased 113% to $5.5 million. Backlog was $1.6m in May and increased to $7m by July 2021. On the call, CEO Gayn Erickson gave more details about the opportunities for silicon carbide devices as he explained that:
“Silicon carbide power semiconductors have emerged as the preferred technology for battery electric vehicle power conversion in on-board and off-board electric vehicle battery chargers and the electric power conversion and control of the electric engines. Our FOX-P family of products are very cost-effective solutions for ensuring the critical quality and reliability of devices in this market, where performance and reliability can not only mean increased battery life, but also whether you have to walk home from a vehicle whose power semiconductor fails in the power train.”
Silicon carbide is more expensive than silicon. This helps explain why it is not used across all industries. However, Silicon carbide excels when it is stressful environment, as it can handle higher temperatures without failure relative to silicon. As a result, silicon carbide has found a niche with EVs and renewables such as solar and wind. Despite being more expensive, silicon carbide is becoming increasingly critical for EVs because if a chip fails in an EV, the vehicle will not work. No body wants to be stuck in the rain.
With demand for silicon carbide chips increasing, there is a sudden and rapid increase in demand for Aehr's silicon carbide testing equipment.
The company reported Q2 FY2022 results today, which came in strong. Sales grew 181% YOY; Bookings increased 263% QoQ from $5.5m in Q1 to $20m in Q2, a record quarter for bookings. Backlog increased from $1.6m in May (Q1) to $36m as of September. The company raised its guide 80% to “at least $50m”.
The company stated that “Our strong bookings include several sizable orders received over the past few months from our lead silicon carbide test and burn-in customer for our FOX-XP™ systems and WaferPak™ Contactors to support testing of silicon carbide power devices for electric vehicles… this Fortune 500 customer is a major automotive semiconductor supplier, and we continue to work closely with them to achieve their test and burn-in requirements and capacity needs. They continue to forecast orders for additional FOX systems and WaferPaks this fiscal year and a significant number of systems and WaferPaks over the next several years driven by electric vehicle semiconductor test and burn-in demand.”
The company also stated it is in the process of adding more customers:
“In addition to this large opportunity with our lead silicon carbide customer, we are currently in detailed discussions with multiple other major silicon carbide suppliers regarding their wafer level test and burn-in needs. We believe we will add several new silicon carbide customers over the next 18 months that will ramp into production on our solutions… We are seeing very strong demand across the industry for wafer level burn-in of silicon carbide devices and continue to ramp our FOX multi-wafer test and burn-in systems and full wafer WaferPaks to meet this silicon carbide market opportunity which we believe is only just beginning.”
The company also shared the following stats:
– The silicon carbide power semiconductor device market is expected to increase over 500% between 2020 and 2026, growing at a compound average growth rate (“CAGR”) of 36% to $4.5 billion, according to Yole Research’s latest forecast
– Deloitte forecasts that the total electric vehicle industry will likely grow at a CAGR of 29% from 2020 to 2025, before reaching 31.1 million vehicles by 2030 and securing approximately 32% of the total market share for new car sales
Silicon carbide is an emerging trend and Aehr is positioned well to benefit from the sudden and sharp rise in demand for silicon carbide. The company is just starting to ramp and the opportunity in front of it is large. To recap recent results:
-Sales up 181% YOY
-Bookings up 263% QOQ
-Backlog up 21,500% QoQ
Disclosure: Bradley Cipriano owns shares in AEHR and he has no plans to change his position within the next 72 hours. The above article expresses the opinions of the author, and the author did not receive compensation from any of the discussed companies.Disclosure: Bradley Cipriano owns shares in AEHR and he has no plans to change his position within the next 72 hours. The above article expresses the opinions of the author, and the author did not receive compensation from any of the discussed companies.
ZoomInfo, formally known as DiscoverOrg, was founded in 2007 and is the premier platform used for highly accurate sales and marketing intelligence. The company is a cloud-based platform that delivers intelligence and analytics to salespeople so they can better target their customers, shorten the sales cycle and increase win rates.
The company is the market leader in business to business (B2B) sales data and has recently reported an acceleration in growth, especially with enterprise customers. The company commands a dominate position, evident by its strong topline growth and cashflow generation. While there are risks, such as privacy concerns and changes to third party cookies, the company is positioned well to benefit from a market undergoing a fundamental shift, as enterprises increasingly modernize their sales processes. In the discussion that follows, I discuss ZoomInfo’s business and the fundamental shift underway in its core market, along with a discussion on its recent financial performance, valuation and key risks.
ZoomInfo’s opportunity
The core of what ZoomInfo does is to help sales professionals know what companies they should be engaging with, who makes the buying decisions and how to contact them.
The company’s B2B sales data is highly accurate as the firm employs a team of 400+ data scientists to train the company’s AI and ML models that constantly source and update ZoomInfo’s 95 million+ company profiles, 500+ million contact methods and 1.6 billion+ daily record events. ZoomInfo is known for having highly accurate information, and the company provides a guarantee that 95% of its data is accurate at any given time. The company provides buying intent data that helps source deals and up to date contact information on decision makers to help close deals.
ZoomInfo operates in a market undergoing a massive fundamental shift, which has only just begun. According to a Forrester report commissioned by the company, only 1.2% of enterprises utilize mature B2B intelligence practices and technology. The report also found that companies that have adopted some B2B intelligence practices and technology generate 35% more leads, resulting in higher revenues and faster growth. CEO-Founder Henry Schuck explained this trend during the Q2 Earnings Call:
“In our conversations with customers, we find companies are still in the early stages of modernizing how they go-to-market. They're just beginning to use data and insights instead of intuition and automated workflows instead of inconsistent one-off sales motion. This is a secular shift that we believe will accelerate.
We estimate that today, the market is only penetrated in the single-digits. And Gartner has indicated that by 2025, 60% of B2B sales organizations will transition from experience and intuition-based selling to data-driven selling, merging their sales processes, sales applications, sales data and sales analytics into a single operational practice.”
The below chart also illustrates how companies are modernizing their sales teams. According to a survey of enterprise CMOs by Gartner, marketing technology has become an increasingly larger part of the enterprise sales budget. Marketing tech has grown from ~22% of an enterprise’s sales budget in 2017 to ~27% of the budget in 2021, taking share from agencies and labor expense. Enterprises are shifting resources away from manual processes and towards tools that improve the efficiencies of sales and marketing teams.
Furthermore, B2B sales and marketing campaigns have evolved into complex projects that cost $100,000+, so having reliable data for highly focused campaigns is paramount. We can directly observe this trend with large enterprise software companies, such as Intuit, Palo Alto Networks and Splunk, each rapidly increasing their S&M expenditures in recent quarters. These companies spend millions on S&M expense per quarter to capture B2B sales.
The below chart illustrates how B2B S&M expenditures has recently accelerated. For instance, the aggregate quarterly S&M expense for the below sample of enterprise software providers increased 28% YOY in Q2 2021, an acceleration from the 11% and 26% growth rates in Q2 2020 and Q2 2019, respectively. The acceleration in B2B enterprise S&M expense adds support that ZoomInfo’s market opportunity is growing at an accelerated rate.
Another trend that supports ZoomInfo’s growth going forward is the rising trend of programmatic advertising, which is highly dependent on accurate data. This is a favorable trend for ZoomInfo, as its robust, high-quality data is critical for efficient programmatic ad-buying. Essentially, as programmatic budgets grow, the demand for accurate third-party data increases. ZoomInfo address this demand by providing targeted audience data and buyer intent data. As the fundamental shift of modernized B2B selling strengths along with a continued rise in B2B S&M expense and programmatic ad-buying, ZoomInfo should be able to continue to grow at an accelerated rate going forward. The company’s recent results also support the narrative that there is still plenty of runway ahead of the firm, which we discuss in greater detail next.
ZoomInfo’s recent results: accelerating enterprise growth and improving cashflows
ZoomInfo recently reported an acceleration in key metrics such as sales and enterprise customer growth, highlighting the firm’s position as the leader in its end market. For instance, Q2 2021 sales increased 57% YOY to $174 million, which beat estimates by $12 million. Q2 sales included a $4 million benefit from acquired companies, and absent this benefit, organic sales increased 54% YOY, which represented an acceleration from the 50% and 53% YOY growth rates in Q1 2021 and Q4 2020, respectively.
The strong topline beat also flowed into guidance, as management increased its FY2021 sales guide by $32 million (5%) to $705 million at the mid-point. The forward guide includes $10 million from newly acquired companies, and absent acquired sales, the guide still came in 3% above the Street’s initial estimate.
Further highlighting the strength in ZoomInfo’s results, enterprise customer growth also accelerated. For instance, customers with annual contract values (ACV) >$100,000 increased 69% YOY to 1,100, which was faster than the company’s 57% YOY topline growth rate. This trend is also evident when viewed on a sequential basis. As shown below, customers with ACV >$100,00) have grown faster than sales on a QoQ basis for the last three quarters.
Generally, enterprise customers are higher value relative to other customer cohorts because they are more likely to expand into new products and can support larger budgets. As a result, the strength in ZoomInfo’s enterprise customer growth improves the quality of recently reported topline growth and also supports a premium valuation.
Another important metric is ZoomInfo’s net retention ratio (NRR), which was static YOY at 108%. ZoomInfo’s NRR is below other tech peers with retention ratios in the 130%+ range. However, the company has made a series of acquisitions and CEO-Founder Henry Schuck explained on the Q2 Earnings Call that he expects these deals to become meaningful to sales in 2022 and 2023 than in 2021. In other words, NRR will likely improve going forward as recent acquisitions are fully integrated onto the platform and cross-selling ramps.
Continuing down the income statement, Q2 adjusted operating profit increased 38% YOY to $76 million, while adjusted operating margin fell YOY from 49% down to 43%. The decline in adjusted operating margin was due to a ramp in hiring, as ZoomInfo’s employee count increased from 1,300 in June 2020 to 2,100 as of August 2021. Adjusted EPS of $0.14 beat estimates of $0.12 by $0.02.
It is also noteworthy that ZoomInfo is well beyond the ‘rule of 40’, as its 57% topline growth rate and 43% operating margin (a proxy for cashflow margin) put it closer to the ‘rule of 100’. In fact, ZoomInfo provided the following slide during its Analyst Day presentation, which showed that the company was in the top quartile for CY21 revenue growth and operating margins.
Further confirming ZoomInfo’s strength is its cashflow performance. For instance, free cash flow conversion was 120% of adjusted operating income, meaning that ZoomInfo collects more cash than it reports as profits. ZoomInfo is able to do this because of its strong market position, as the company collects cash upfront from customers. The upfront collection of cash is a significant advantage for ZoomInfo as it is effectively an interest free loan from customers that helps support ZoomInfo future growth. The upfront collection of cash is also a sign of market dominance, showcasing that ZoomInfo has pricing power over its customers (customers generally want to pay later, and sellers want to be paid upfront). Being paid upfront also supports a premium multiple.
As mentioned above, ZoomInfo is in a dominate market position due to its first mover advantage and large, highly accurate industry specific data. This market dominance is also present in ZoomInfo’s financials, as the company is rapidly growing with enterprise customers and these customers are paying cash upfront. In the next section, we discuss the firm’s valuation and conclude with key risks that investors should be aware of.
Valuation
ZoomInfo claims to have no direct competition, rather it competes with niche operators. Due to the lack of directly comparable peers, it is best to compare ZoomInfo to other fast growing and highly profitable tech firms, such as Zoom Video, Snowflake, Adobe, Veeva and Shopify.
Against this peer set, ZoomInfo’s P/S multiple of 38x was 28% higher than its peers but its most recent growth rate of 57% was also higher than the peer median of 54%. Moreover, ZoomInfo’s forward growth rate of 49% YOY is well above the peer median of 31%. A faster growth rate helps support a premium multiple.
As discussed above, ZoomInfo also reported an acceleration in enterprise customer growth. Since enterprise customers have larger budgets and can pay more and expand into more products, they are higher value and support a premium multiple. Furthermore, the company has pricing power as its customers pay cash upfront, which helps support future growth and also supports a premium multiple.
Key risks and conclusion
Data protection is a major theme globally. The FTC is increasingly enforcing data privacy in the U.S, European Union enacted the General Data Protection Regulation (GDPR) in 2018, the U.K. has a Brexit-amended GDPR that went into effect in 2021 and California Consumer Privacy Act went into effect in 2020. These laws have added tremendous complexity and impose certain restrictions and obligations on companies such as ZoomInfo.
However, this data compliance complexity can actually work in ZoomInfo’s favor, as it provides a barrier to entry. New entrants must try and compete with ZoomInfo’s robust data and also comply with complicated compliance burdens, which could make the endeavor cost prohibitive.
ZoomInfo also takes privacy and compliance seriously and has a dedicated team to processing requests for deletion of contact information and also announced an expansion to its privacy team. The company’s goal is to build trust and the company has implemented a program for providing direct notifications to individuals that are in its databases.
There are also risks associated with third-party tracking cookies and the IDFA changes announced by Apple. These changes impact the way that data is collected by third-parties, and could limit ZoomInfo’s buyer intent data. However, ZoomInfo rolled out ‘privacy clusters’ in 2020 “which allow ZoomInfo to deliver B2B intent in a privacy-first way without the reliance on cookies or other Identifier For Advertisers (IDFA) or Personal Identifiable Information (PII) based tracking”. So while ZoomInfo will likely be impacted by the change in third-party tracking, its focus on B2B company data, rather than individual level data, should limit the impact on the firm going forward.
In conclusion, ZoomInfo is positioned well to benefit from a fundamental shift happening with B2B selling. Enterprises are looking for ways to scale their sales and marketing programs with repeatable processes, and ZoomInfo has the data platform to facilitate this trend. The company reported an acceleration in growth as well as an acceleration in enterprise customer growth, which supports a premium valuation. Moreover, the company gets paid upfront in cash, evident by its strong cashflows, which further highlights the company’s market dominance and also supports a premium valuation. While there are risks, such as privacy concerns and changes to third-party cookies, these trends may actually work out in ZoomInfo’s favor by increasing the barrier to entry. Looking forward, ZoomInfo can be expected to continue to grow at a robust rate as B2B sales increasingly modernize.
Technical Setup
By Knox Ridley
Zoom Info’s recent earnings report attracted a swarm of new buyer. This can be seen with the large spikes in volume, which propelled ZI to new highs. After breaking out of its IPO high at $64.40 (green on the chart), ZI has been consolidating above this breakout, which is historically a bullish sign. It has further formed a minor cup and handle pattern just below the $67.50 resistance zone (blue on the chart), which it is currently attempting to breakout from. If confirmed, we expect to see a nice move within our momentum portfolio.
Disclosure: The I/O Fund owns shares in ZoomInfo and does not have plans to change its position within the next 72 hours. You can access the I/O Fund’s positions herehere. The above article expresses the opinions of the author, and the author did not receive compensation from any of the discussed companies
UiPath’s core product is UI-based automation and they’ve recently acquired Cloud Elements to add API-based automation. The software robots are able to work across programs in the background, build applications, send emails and interact with chatbots. Although making AI/ML actionable is UiPath’s sweet spot, the trend towards low code/no code is also a tailwind for UiPath. We know the AI-bellwether Nvidia has led the market this year, which helps us gauge where we are in the AI cycle as semis must move first. Therefore, Nvidia helps provide evidence that we are early to UiPath (unlikely automation moves exactly in sync with GPUs) and the goal will be to remain invested through the ups/downs as the story unfolds.
I had posted on the forum my thoughts on the decelerating revenue and why the unique business model and change in billing terms is causing the top line to look weaker than the company actually is. Management emphasizes to focus on ARR and we break down why that’s important. Although a change in billing terms can be seen as a weakness, we don’t think this is the case as the change in terms could open up the company to more customers who aren’t willing or able to pay up front. We had outlined in our original analysis that UiPath is an expensive product with a customer concentration at the enterprise level. Customer growth above $1 million was up 100%, therefore, UiPath is showing strength in its core customer base. However, UiPath is now ready to invest at the citizen developer level in the Studio X product.
Earnings Overview
This section posted on the forum Thursday, September 16th:This section posted on the forum Thursday, September 16th:posted on the forum Thursday, September 16th:
UiPath beat and raised guidance, so there was no issue here. Rather the issue is the company’s growth is decelerating and this is raising some question marks as to how long the deceleration will continue. The market is forward looking, and in this case, the market is pricing in lower growth. The guidance does not fully illustrate the deceleration in licensing and deferred revenue. We break this down for you below.
First, I want to talk about UiPath’s unique business model which is to license software robots and then charge for support and maintenance. Here is the difference as outlined in the S-1 filing:
Licenses:
We sell term licenses which provide customers the right to use software for a specified period of time. From time to time, we also sell perpetual licenses that provide customers the right to use software for an indefinite period of time. For each respective type of license, revenue is recognized at the point-in-time when the customer is able to use and benefit from the software, which is generally upon delivery to the customer or upon the commencement of the renewal term.
Maintenance and Support:
We generate maintenance and support revenue through technical support and the provision of unspecified updates and upgrades on a when-and-if-available basis for both term and perpetual license arrangements. Maintenance and support for perpetual licenses is renewable, generally on an annual basis, at the option of the customer. Maintenance and support represents stand-ready obligations for which revenue is recognized ratably over the term of the arrangement.
The reason the company emphasizes ARR as the key metric to focus on is because it accounts for UiPath’s upgrade-heavy business model and it shows the strength from retention. Expansion revenue is essentially what drives UiPath rather than yearly subscriptions alone (revenue). Once a company licenses software robots from UiPath, they are more likely to license more software robots over time and to need more maintenance and support. This upgrade cycle is unique from other cloud companies that have only specific cohorts they can upgrade and are more focused on yearly subscriptions (i.e., Pro Plan to Enterprise Plan, etc.).
Management at UiPath is communicating that the ARR forecast is more important than the revenue forecast as it accounts for the upgrades they are expecting. The dollar based net retention rate for the company is very high at 144%. Evidence of the expansion revenue model is also seen in the company’s lifetime value which is 233X for the top 25 customers and 62X for the top 100 customers. UiPath’s customers are enterprise customers with large budgets, which is why we saw the $1 million+ ARR cohort up 100% this quarter and those accounting for $100,000+ ARR up 59% this quarter.
Why UiPath Sold Off Despite a Beat:
Given what analysts know about Q3 guidance, the current projections for fiscal year 2022 is at 44%, which is down from 81% in the last fiscal year. We’ve included the analyst projections for the following year, which right now are at 34% growth to $1.17 billion.
However, the ARR hints towards stronger revenue growth in the future. For fiscal year 2021, revenue of $608,000 exceeded ARR of $580,400. This year, we are seeing ARR slightly outpace revenue if we based revenue projections on analyst consensus. While revenue is forecast to grow 44% this fiscal year, ARR is forecast to grow at 51%. Historically, UiPath’s revenue exceeded ARR.
Another explanation for why we are seeing the lapse between ARR and revenue is that the CFO mentioned a change in terms of how the company bills. In the past, the company billed multi-year deals all at once, and instead, they are shifting towards billing annually. The management stated the reason is that it allows more upgrades as their customers’ needs change and it results in less up front from their customers. In the example provided, instead of buying 10,000 robots at once, they would buy 1,000 and then 5,000 and then 10,000 with the new billing structure with the revenue realized across three years rather than realized in one year.
Here is the quote about the difference between revenue guidance and ARR: “So when I talk about an annual ramping contract, one of the things that is really positive for us is digital transformation is a long-term trend. And what has happened with the strength of UiPath’s platform is automation is a staple for the long-term requirements for customers to transform the way they work in digital transformation. So what that means is instead of buying simple annual contracts, what we see a larger demand for is getting larger term commitments from some of our customers. But the way they look at that is instead of buying 10,000 robots today, they may buy 1,000 robots today, 5,000 next year, 10,000 in year three. And those – the license deliveries would happen into those years as we go down.
And so, one of the things that we look at is that we like that because that is better ROI for our customers. And I – when we think about the impact on financial metrics, two things. One is remember, we – and I repeat this, we really drive our company to ARR. From an ARR perspective, there is no impact that is there. Based on the way the contracts are structured, if license delivery happens in the out years, then that does have – that creates variability in revenue because we only can recognize the revenue upon delivery of the license. And so that is kind of the way that I think about it from a modeling perspective on revenue. But again, I stress ARR really, there is no impact, and that’s how we drive the business.”Based on the way the contracts are structured, if license delivery happens in the out years, then that does have – that creates variability in revenue because we only can recognize the revenue upon delivery of the license. And so that is kind of the way that I think about it from a modeling perspective on revenue. But again, I stress ARR really, there is no impact, and that’s how we drive the business.”
This change in how customers are billed likely led to lower licensing revenue in the current quarter of 20% growth compared to 72% growth from the fiscal year (not apples-to-apples comparing Quarterly to Fiscal Year but helps provide color). This is down from 57% last fiscal quarter ending in April. We see evidence that those licenses will be recognized in future quarters not only in ARR but also in support and maintenance growth, which is recognized ratably and grew 74% year-over-year compared to 79% in the quarter ending in April. In other words, customers aren’t falling off or downgrading rather they are paying for licenses across many years rather than pre-paying.
Expanding on Earnings:
UiPath reported revenue of $195.5 million compared to the consensus for $184.3 million. This represents 40% growth year-over-year compared to 65% growth in the prior quarter. EPS also beat at $0.01 compared to a consensus of ($0.05) EPS. UiPath has a total of 9,100 customers with 600 added in the recent quarter. The company also has 4,700 partners after adding 400 in the most recent quarter. The Partner Network is part of our thesis on UiPath and we think this number carries significance in terms of a defensible position. This quarter, the company highlighted its partnerships with Alteryx and Smartsheet.
The company’s adjusted gross margins are at 86% in the most recent quarter with adjusted free cash flow at a loss of $3.5 million. As stated, licensing revenue slowed down with the revenue mix being $95.5 million in licenses compared to $79.5 million in the year-ago quarter. Maintenance and other Support was at $90.3 million compared to $51.9 million in the year-ago quarter.
We believe the management is correct in focusing on ARR. As stated above, revenue typically exceeds ARR. Therefore, if we draw conclusions based on the historic performance of the company, the revenue growth would be above 60% in this quarter and above 51% in future quarters. In addition, net new ARR was up 33%. We will need to see how long it takes before the company absorbs the change in billing terms, but due to where we are in the AI cycle, we prefer to be patient. If this was cloud software, which is moving towards consolidation, we would be more concerned. Hopefully, the above section explains why we are not concerned at this time.
Guidance came in higher than expected at revenue of $208 million at the midpoint compared to consensus of $206 million. The guide in ARR was at 54% next quarter and at 51% growth for the fiscal year. When asked if the raised guidance was from a demand signal, the company pointed towards their Net Retention Rate, which remained robust at 144%, indicating that demand for their products has remained strong.
From my perspective, UiPath has very few comparables on the market but we can lump the company in with cloud software with the understanding that cloud software’s growth reflects a mature market while UiPath’s does not. Forward fiscal year P/S is at 32 based on $874 million consensus estimates for this year and 1-year forward is at 24 if based on consensus estimates of 1.17 billion. We think this is a reasonable valuation for a company that is at the forefront of automation, which is one of the least-hyped yet most practical commercial uses for AI and ML. However, the full lockup is next month and we had stated in our initial coverage on UiPath and reiterated across other analysis that even the most quality companies come under pressure from insiders and early investors needing an exit.
Of the companies pictured above, ZS, TEAM and NET have similar forward revenue growth and we can see they are fetching much higher valuations. It will be interesting to see how this unfolds if UiPath’s revenue does indeed catch up to ARR and enterprise customer growth.
Product Catalysts:
The focus of this update has been primarily on the financials as we’ve written a deep dive on the product, which you can find here.
UiPath has a few catalysts this fall, including releasing attended robots for the Linux operating system and also for Mac users. Mac OS users make up roughly 17% compared to Windows, there is a higher concentration of Mac users among citizen developers and also in enterprises in the tech industry. Linux makes up a very tiny portion of overall operating system market share at less than 2% yet dominates cloud infrastructure at 90% penetration. In 2021, 100% of the world’s top 500 supercomputers ran Linux. Here are the releases the company has planned including a web-based version of StudioX:
Now, we continue to invest in StudioX as a major tool to foster the community of citizen developers. We are going to extend it also to be available multi-platform. So, you will expect quite a bit of investment from us in the coming few quarters. It’s – right now, we are doing really a major advance into multi-cloud and multi-platform. And we are launching Linux-based robots. We just announced yesterday the public review available. We are launching Mac support. We are going to launch early next year, the web-based StudioX that will make it even easier to adopt. But overall, we really believe that it’s important to have a suite of tools that cater to a large array of options from professional developers to citizen developers and to all business users.
Conclusions:
The lockup for UiPath happened in tranches after the first fiscal quarter results and also following the second fiscal quarter results, per the S-1 filing, with some restrictions. The full lockup with no restrictions occurs on October 18th. This is likely weighing on shares more than anything in the current ER as some shares were up for lockup expiration at the time of fiscal Q2. Participating in IPOs rarely works out in the first few months of the listing unless you’re actively trading, which we stated clearly in the original write-up on UiPath. However, we decided to add this to LTBH to show our seriousness in building AI positions.
With that said, we will likely hold off on Confluent until post-lockup. You’ll get this analysis soon to set the stage for why we are building MDB, ESTC and eventually CFLT as a package as we think there’s an important trend bubbling beneath the surface. Keep an eye for this deep dive end of next week.
The I/O Fund believes venture capitalists are some of the world’s best tech growth investors so we were thrilled to join Jason Calacanis on This Week in Startups. Jason is a successful angel investor with a portfolio that includes Uber, Robinhood, Wealthfront and Calm. In the fast-moving podcast, we covered topics including:
· SPACs, which ones you should own and which ones you should avoid
· The difference between bubbles and bear markets
· How Roku has been able to stave off Big Tech
· Why I/O Fund is ultra-bullish on Nvidia
· Whether China is worth the risk
· If hybrid work-from-home is an investable trend
Timestamps:
1:19 SPACs
5:25 Interesting SPAC story
16:43 Roku
27:00 Will tech spin-offs create more value?
30:45 Nvidia
36:16 Robinhood
43:21 Chinese stocks
48:43 Post-pandemic world
49:52 Hybrid work-from-home
59:26 Crypto
1:11:32 Gemini
SPACS
SPACS gives a chance to investors to invest in early-stage tech. It has got its pros and cons. Some of the merits include that individual investors can get a better price and higher returns. It would be prudent for investors to have the right skill sets to choose the best stocks since the early growth company’s financials could be limited compared to well-established companies. Investors also should be careful about companies banking on a lot of forward growth because if they miss estimates and the stock sell’s off immediately.
At the same time, we cannot ignore the SPAC market as we could miss some of the best growth opportunities. We will need to have some risk management and a good exit strategy; this is how I/O Fund approaches the SPAC market. We found stories that we liked, played some momentum, and we had an exit strategy.
Video clip: From 0:25 till 5:22 minutes.
Macro Trends (Is it a bear market now?)
The current market situation is not like the 1999 or 2008 bear markets. Many investors think that the sell-off of the last year was a bear market. It was just a bubble because of irrational, momentum-driven, and human psychology. We have seen many such bubbles in the past few years. So, bubbles don’t mean bear markets. It means that there’s an excess of liquidity. What leads to bear markets and recessions is typically Central Bank policy, which is incredibly loose right now. The U.S Federal Reserve is expected to continue this policy to stimulate growth a little longer.
Video clip: 9:22 to 11:05
Can Roku fight big tech?
Roku has the first-mover advantage in advertising-based video on demand (AVOD) versus subscription video on demand (SVOD) specifically not only by the cord cutters but also from the brand advertisers. They have designed an operating system that is cheap and flawless and fits well with smart TVs. All that is great because they have the hardware and the operating system. But ultimately, Roku’s path to Wall Street gains is that they are an ad platform.
We are moving into a world where advertisers on linear pay-TV are moving on to the cord-cutter companies like Roku. For the first time in 37 years, Budweiser did not advertise in the super bowl this year. So, the addressable market is huge for companies like Roku.
Video clip: 16:55 to 21:05
Bullish on Nvidia
Nvidia has yet to tap its real market, which is Artificial Intelligence (AI). A lot of people talk about gaming. But I firmly believe that this is a data center and an AI accelerator play. Most of the industries in the future will run on Nvidia. This is a big statement from me since I have researched very well and been writing about it for about three years. The stock is up 340% since my initial coverage. The main differentiator for Nvidia is GPU chips which are very good for AI.
Video clip: 30:45 to 33:08
Our exposure in Chinese stocks
I/O fund has exposure in Chinese stocks through an Electric Vehicle (EV) stock and a cloud infrastructure stock. We stayed in those markets because China has about 1.3 billion people and a lot of them are making decent money. So, when you think of EV companies selling to this huge population subsidized by the Chinese government we strategically played in that market.
There are cities in China that are of the size of Los Angeles that no one even knows the name of these multiple cities. The investment thesis is phenomenal but political risk is high. However, the market’s short-sightedness can make you miss the rally. For example, the French stocks, who was going to invest in French stocks last year? However, after the country announced the second lockdown last year, the stock market was up 25% in about a month. So, we think that inevitably this might happen with China and we are seeing the evidence of that in one of the position’s we are holding which is showing technically strong signs that it is going up.
Video clip: 43:21 to 47:19
Hybrid work-from-home
Hybrid work-from-home stocks and why the I/O Fund thinks it’s an investable trend
49:52
Free Market Redistribution of Wealth
You can see the exodus of people moving from one state to another. People are actually able to make more money from the tools the tech is providing. They also have more time and on their own terms. It’s more like a free market national redistribution of wealth and it’s happening in a positive manner.
Video clip: 55:55 to 56.18
Crypto Investment
In the past we have mentioned that the debate on crypto has been distracting investors from the opportunity that these new technologies can deliver. We have invested in Crypto from a tech perspective. We think Bitcoin has done the unthinkable. An actual store of value on par with gold and with the dollar. We are comfortable since it has some real-world disruptive value and we run technical on it and ride the trend.
Video clip: 59:26 to 1:00:35
The I/O Fund currently owns shares of ROKU, NVDA, STEM, Bitcoin, Ethereum, alt-coins, and crypto brokerages such as Voyager Digital. This is not financial advice. Please consult with your financial advisor in regards to any stocks you buy.
Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock or crypto/asset in the companies mentioned in this podcast.
