FSLY NET Buy Alerts
Today We Are ALL Living On the Edge In More Ways than One--BUY Fastly FSLY under $95 Target $155 & Cloudflare NET under $42 target $85
Hey Subscriber,
For the foreseeable future, we are going to add new and some old favorite "beat, beat, raise guidance" digital platform plays in our favorite post-COVID 2021 secular transformation sectors when they DROP back to their 50-day moving averages.
Why? Because after running the data on over 250 secular growth stocks in our Digital Platformity universe of stocks, the data is 100% clear--the 50-day price moving average is where the high conviction human beings are adding to the greatest long-duration secular growth sectors in the world after the >20-day moving average momentum robotic algorithmic trading buyers bailout on the break of the 20-day trading support.
If you don't believe me, just pull up the chart of ANY of our favorite digital platform stocks--NVDA AMD SHOP CRM NOW WDAY SPLK AAPL ADBE W INTU CHGG BABA GOOGL MSFT ETSY SQ PYPL FB etc--they ALL look the same on a chart.
Key Point: The risk in leading digital platforms ("DP) stocks is waking up one morning and your position is down 30%+ as some minor hiccup in their earnings has the algo buyers selling pre-market open. But, if you understand the serious advantages that leading 21st-century bits & bytes DP companies have over their 20th-century atom-based industries (with a few exceptions of course in the post-pandemic world to come), the data now says those puke up days were DP buying opportunities.
So we WILL buy the pullbacks to the 50-day moving averages because it is working (until it stops working of course.) Let's start with the second derivative plays on eCommerce and digital workflows i.e., the prime technological beneficiaries of the New Normal enterprise.
Come with Me to The Edge: We Are Betting On the Two Leaders FSLY and NET
So let' talk "Edge Computing" and why the pandemic so accelerated the digital platform networks into 1) data center computing and 2) network edge computing. The explanation may get a little nerdy, but no one's gonna get hurt I promise.
Part of our edge computer network investment thesis goes like this: 111,680: That’s how many applications were filed to start new businesses the week of August 10, a 69% increase over the same week last year, according to the U.S. Census Bureau. It’s not just a one-time spike, either. The previous week saw 113,860 applications for new business tax IDs (up 75% year-over-year). And more than 100,000 new business applications have been filed every week for the last eight weeks, surpassing the number of applications filed in the whole second quarter of 2020.
And that is just in the US--the same digital eCommerce startup explosion is occurring worldwide.
Key point: One of the beneficial aspects that we said would come out of the pandemic would be "creative destruction" and massive entepreneurial reinvention. Right on cue, this surge in digital entrepreneurship has led to a historic surge in e-commerce sales worldwide over the course of the pandemic. With cloud-based eCommerce digital platforms like Shopify, Big Commerce, Etsy, Square, and Paypal making it easy for anyone to start a business to sell things online, it’s likely that a significant chunk of those new businesses are being started by people at home on their laptops.
Second wave--the 3-5 year acceleration of the flexible workspace and workflow enterprises compressed into just 3-4 months--from companies with a handful of employees to 5 million+ employees. Think of cloud-based applications like Slack, Zoom, Teams, and shared vertical workflow applications like ServiceNow, Workday, Paycom, Docusign, TR Digital Platformity companies all.
But where do Shopify, Etsy, Square, Slack, DocuSign and so many other cloud software-as-a-service SaaS get the digital horsepower to run all these stores/apps at the speed of light so buyers don’t have time to change their minds, users don't get frozen screens, and at the same time, keep these stores and apps secure from nefarious hackers??
Answer? The network edge cloud and edge computing networks.
With the acceleration of cloud-based computing has come a spectacular parabolic rise in data and need for the network edge cloud. You know this reality: "Today we are creating as much data in a day than we did from the dawn of civilization through 2003" says ex-Google CEO Eric Schmidt and others.
With that data, a new computing possibility arose — we could now create digital inference models and algorithms using artificial intelligence (AI) and deep learning that are incredibly data hungry but meaningfully and rapidly move the ball forward in the evolution of humankind (I'm ready for the AI vehicle to take me and friends out to dinner and home WITHOUT the risk of DUI--trust me!)
AI takes a ton of hardware and requires billions of lines of coded software. Our old friends Cloudera (CLDR) and long time holdings Nvidia (NVDA) and AMD are central enablers in AI/ML (and we are patiently waiting to repurchase them on new entry points--especially NVDA).
But in all of this, another transformation has emerged; because as the world of technology has moved to embrace the centralization of data, so the unimaginable load on the cloud has increased enough to create a new problem — digital latency.
Latency – The Need for Speed
Latency can mean many things, but in technology, it simply means the amount of time between a request and a reply from that request. For example, the time it takes from pressing a key on your keyboard to the time it shows up on your screen. That latency is extremely short because it happens locally.
Another type of local latency is on your smartphone. A latency on our phones that we can tangibly feel, for you iPhone face recognition users, is the time you hold the phone up to see your face and the time it takes for the phone to open as it recognizes your face.
Training versus Inference
But there are more pressing issues to latency than waiting for your iPhone or iPad to open (and let me tell you know--APPLE is going on our buy list soon with the next meltdown--with long term options). Imagine a self-driving car communicating with the cloud, when it suddenly faces a dangerous situation — let’s call it life or death — turn left, right, or stop. In this case, one second would be devastating and would virtually eliminate the possibility of self-driving.
Key Point: While self-driving features are trained in the cloud, where the heavy lifting is done, but the inference is done locally, with a computer in the car. Training involves the development of an algorithm’s ability to understand a data set, whereas inference is when the computer acts on a new data sample to infer an answer to a query.
With all of these new demands coming into the technology sphere, names have changed. What was once called the cloud, is now called the central cloud. And with that comes a new idea — the space in between the massive data collector and training engines in the central cloud, and the spot that does the immediate actions with inference. This space is called the edge cloud.
And it is the edge cloud that is exploding 5X faster now than the data center central cloud.
The Edge – It’s the Speed of Light and Application Flexibility & Function at a New Level
In the iPhone example, the edge computing node is the phone. In our self-driving example, edge computing is done in the car itself. But the edge cloud sits in between the hardware (a phone or car) and the gigantic central cloud.
This edge cloud idea new enormous shift coming to our world. Here is one estimate edge computing market size worldwide in 2018 and 2025 from Grand View research (that we subscribe to BTW.)
Key Point: The emergence of the edge cloud is a massive disrupting force today and locked into 35-45% annual enterprise revenue growth for the next 5 years!.
Why? Because we have the new 5G mobile network coming forward like a data creating freight train. With 5G will come unheard-of speeds from simply a mobile connection. 5G is expected to support 1 million connected devices per square kilometer and deliver up to 20 gigabits per second of peak data download rate for each cell site (20 gigabits are 20,000 megabits).
Who cares? At that bandwidth rates, we download a full-blown high definition movie in about 3 seconds on our cell phones. A single music file could be downloaded in under one second — in fact, 500 songs would take one second.
And that’s not Wi-Fi; it's just your standard mobile network speed.
Now obviously as the edge cloud moves forward and 5G wireless networks are lit up, there are of course new hardware needs. Here is the forecast size of 5G and high-performance edge infrastructure hardware market worldwide from 2017 to 2022.
IN their investment presentation, Fastly does a nice job of illustrating the edge cloud versus legacy central cloud (OK a little nerdy but buck up buttercup!)
The pink color shows where developers have control versus that blue color where IT has control. Again, both Fastly and Cloudflare are built for application developers — that’s what the edge is for--to make the really cool stuff the app developers come up with actually work without experience destroying latency (as well as provide a data security level against hackers as well--that story in a minute).
Both Fastly (FSLY) and Cloudflare (NET) operate a stand-alone edge cloud platform outside the central data center for processing, serving, and securing its customer’s applications. Fastly defines the edge as the point in the network just before you lose control of the data. But an edge cloud platform moves data and applications closer to users, at the network edge — so you can meet rising customer expectations, deliver secure, competitive online experiences, and bring innovative ideas to life, faster.
By leveraging the right infrastructure, enterprises can scale smarter, improve security, and optimize the user experience. But the move from the central cloud to the edge cloud introduces another major cybersecurity vector and pain point. About 25-35% of the revenue both players is edge network security because security and delivery cannot be disentangled. If you have a cloud, even an edge cloud, you need security. Period.
Business Nuts and Bolts
Fastly and Cloudfare servers (the edge cloud) do not sit in the centralized pubic clouds (although they are in the same building), rather they sit in a point-of-presence (PoP) room. A PoP typically houses servers, routers, network switches, multiplexers, and other network interface equipment, and is typically located in a data center.
ISPs (think Cox/ATT/Verison/Spectrum) typically have multiple PoPs. PoPs are often located at Internet exchange points and colocation centers.
An easy way to understand a POP is "it's the room where the Internet world gets connected." Fastly/Cloudflare are both in all major metro areas, 60+ to date, across all geographies, both with plans to get to 100+ locations in the longer-term. In the Pop, both deliver custom BGP (Border Gateway Protocol), custom load balancing, custom (cyber) security, and SDN (Software-defined networking) enabled management.
For those keeping score at home of tech terms, Border Gateway Protocol is the routing protocol used to route traffic across the internet. It’s oftentimes referred to as “the routing protocol that makes the internet work.”
For more scorekeeping, SDN is an approach to using open protocols, to apply globally aware software control at the edges of the network to access network switches and routers that typically would use closed and proprietary firmware.
FSLY has 55+ terabytes of capacity in all of its geographies combined and has issued 30 patents. NET is catching up to Fastly but has more security IP.
And today, Fastly announced the acquisition of Signal Sciences for $200M in cash and $575M in stock. The deal is expected to close this year. Fastly says the web application protection company will build out its security portfolio "at a time when security at the edge has never been more critical. Signal Sciences' tech will be integrated into Fastly's upcoming Secure@Edge offering, which offers "a modern, unified web application and API protection solution that will power and protect companies looking to further or begin their digital transformation." Fastly expects the deal to be accretive to its growth and gross margin.
Customers
With tech companies like FSLY and NET so new to the public domain, we we have to look at customers and customer growth and retention rates, not financials. But each are valued at 20-24 X revenues which, at 35-40% annual growth rates and 135% revenue retention rates (means they ADD new capacity to existing clients), both are valued significantly less as a multiple of revenues versus many well know digital platformity stocks.
This is Fastly list; -Cloudflare says it is trusted by "approximately 25,000,000 Internet properties in basically every industry and sector possible.
Further, both NET/FSLY have all six cloud titans as partners — that is, data can be sitting in any of those centralized clouds at MSFT, GOOGL, Alibaba, Amazon AWS, and Oracle and still use Fastly/Cloudflare on the edge. This means the enterprise can have data stored with any of the cloud titans, and the software layer they use for Fastly or NET is unchanged; developers do their work once, and that’s it.
Why does this matter? Because Forrester Consulting found in 2018 that 86% of enterprises have adopted a multi-cloud strategy. Further, IDC predicts that 90% of companies will be multi-cloud by 2021.
It’s a trend that builds a wide moat for both companies and lessens any future competitive strikes from the cloud titans when (it’s a when not an if) they decide to get into the edge cloud realm.
Key point: If Microsoft/Walmart buy the US/Canada/Australia business of Tik-Tok, my money is on them buying Fastly who provided them their edge cloud. And this is why fears of "Fastly losing 12% of its enterprise revenues from the sale of TikTok" is completely bogus--MSFT and Walmart do NOT HAVE the edge cloud capacity--and it would take them years to build their own.
Finally, it's the “land and expand” strategy known to most software companies that gives Fastly and Net the strategic approach to new customers and growth from them moving forward:
Here are the 36% CAGR and "Land & Expand" charts for Fastly that is almost perfectly matched by Cloudflare. FYI in 2020, annualized growth for BOTH companies is estimated to have exploded to 55-60%+ which will continue above trend for the foreseeable future.
So there you have it--all the above is why we must invest in the "new cloud" that is the new fast secular grower in the world 2020-2025.
PS--we had a typo on our USOI upgrade--buy under $5, not $15 (apologies). But the range on USOI is 4.65 to $4.85--I like to put limit buy orders in at $4.65 to add shares. AND another way to play the huge dividend is to wait for the dividend announcement and ex-dividend date. You have to buy the shares 2 trading days BEFORE the ex-dividend date to get that months dividend. But AFTER the date of record is passed, you can trade out of the USOI shares because unlike most dividend stocks, these ETN shares do NOT trade down on the day the dividend is paid. This week they traded UP on the Aug 25 payment date!
Despite regulatory risks, their technological advantages and network effects (value of service dependent on the number of users) give them greater protection against competitors.
All 5 companies are sitting on huge cash piles, as their cash holdings exceed debt.
High price-to-earnings ratio indicate that investors expect higher earnings.