FSLY explained

https://softwarestackinvesting.com/fastly-edge-compute-expla…

FSLY has grown into my largest position (20+%) this article is the best explanation of why I am not trimming my position. The author likens the DNA at ZM to that of FSLY. Fair warning it is a very technical and fairly long article.

30 Likes

Thanks! A very timely article for me. I’m a late comer to Fastly. This week I was reviewing my portfolio to see what improvements I could make. This year’s top performer by a large margin is Zoom (ZM) and also my largest position. Next is Teladoc (TDOC) which is outpacing most SaaS stocks. Third in my portfolio is Everbridge (EVBG) which is doing nicely but well below Teladoc, it just does not seem to be as mission critical as other services.

One has to be very careful not to chase performance which typically leads to buy high and sell low, the reason most mutual fund investors underperform these funds! I’m looking for a better performer than Everbridge but there has to be a business reason for the change, not just a difference in recent performance. After an initial preselection of ten candidates I reduced the list to four:

BILL - cloud based back-office financial operations
DOCU - cloud e-signature
FSLY - edge computing platform
LVGO - healthcare realtime data capture

BILL reminds me of an excellent Peter Lynch pick from way back when, ADP which was a back office favorite. DOCU is one I really like but I bought it and sold it a while back because it did not seem mission critical and differentiated enough from the competition. LVGO reminded me of an old position of mine, BEAT which is not performing as I thought it should. In healthcare I’m sticking with Teladoc. That left Fastly which seemed mission critical to an old IT guy but I really did not understand why it should be a big winner. The linked article is a huge help. Thanks!

An important takeaway is this cautionary note:

Investor Take-aways

While Fastly is benefiting from investor excitement and possibly some hype, there are no guarantees around the success of Fastly’s Compute@Edge solution and the eventual size of the market for it. Cutting edge technology innovation doesn’t always translate into market penetration and competitive advantage. My purpose here is to explain why I think Fastly’s fundamental approach to addressing long-standing challenges in CDN and distributed serverless infrastructure is unique. The investment thesis is that like Zoom, Fastly’s intense focus on a limited set of difficult problems with the goal to create noticeably better outcomes for developers and DevOps teams, will ultimately drive investment returns over the long term. Personally, I think Fastly is fundamentally bringing a better architecture design to distributed, serverless compute outside of the data center. While nascent, similar to Shopify’s planned usage, I can think of many applications for this technology.

The parable of the better mousetrap attributed to Ralph Waldo Emerson is not true in the high tech world, the market seems to pick the winner at random. How then does one decide which competitor to buy. When there is a clear market leader like Zoom, the choice has already been made and the problem solved. When there is no clear leader one has to depend on subtle clues. The article notes Fastly’s most prominent clients which read like a list of the top Who is Who! Microsoft, Google, Amazon, Shopify… These people know who is who!

The other very telling clue comes from a website that rates CDNs, CDNPlanet.com. They track 26 CDNs but in the “Compare CDNs side-by-side” panel they have preselected the two top contenders, Cloudflare (NET) and Fastly (FLSY).

https://www.cdnplanet.com

Denny Schlesinger

Build a better mousetrap, and the world will beat a path to your door

https://en.wikipedia.org/wiki/Build_a_better_mousetrap,and

43 Likes

Denny,
I probably should read Poffringa article first. I am quite bullish on Fsly but as he says Compute@Edge is not a guaranteed homerun. Also, let us not forget Shopify uses NET as well. I have not checked but it is quite possible the others like Google, Microsoft are using other CDNs like Akamai as well. It is highly unlikely they use only Fsly. Could it be that they have chosen Fsly for solving some specific use cases as opposed to all use cases which are commoditized? This is important to understand because the current euphoria about Fsly seems like it is set to displace Akamai and others.

2 Likes

Also, let us not forget Shopify uses NET as well.

It’s pretty typical for large companies to use multiple CDNs. For instance, Fox Sports used multiple CDNs for streaming the SuperBowl earlier this year. However, Fastly was the sole back-end for all of them.

The Shopify case is a bit perplexing. On one hand Shopify itself says:
CDN stands for content delivery network. Shopify provides merchants a world class CDN run by Fastly. Using a CDN means that your online store will load quickly around the globe, despite the Shopify servers being in North America. You can read more about Fastly on the Fastly website.
https://shopify.dev/tutorials/customize-theme-troubleshootin…

And Fastly says:
Shopify felt constrained by their former CDN provider when they needed to experiment with new features or adjust CDN settings for better performance. Any feature change, A/B test, bug fix, or tweak required assistance from a support engineer, which sometimes took up to two days and slowed them down considerably. Since Fastly is built on Varnish, Shopify is now able to manage customizations on their own time by deploying instant Varnish Control Language (VCL) configurations.
https://www.fastly.com/customers/shopify

But as late as April 2020, Cloudflare is apparently still being used for at least part of Shopify:
https://www.searchenginejournal.com/4-new-edge-seo-capabilit…

At any rate, it should be expected that large diverse websites would use different CDNs for different functionality in an attempt to get the best of breed for different applications. Hosting images is different than streaming video is different than security is different than SEO is different than digital payments is different than - well, you get the picture. And, of course, different companies have different presences in different countries, so depending on the location of the target users different CDNs could easily be utilized.

And, finally, as with the SuperBowl, using multiple CDNs helps isolate you from a failure in any one. It’s been just over a year since Cloudflare’s embarrassing failure that took down a large number of popular sites for about 30 minutes: https://slate.com/technology/2019/07/cloudflare-internet-out…

17 Likes

Also, let us not forget Shopify uses NET as well. I have not checked but it is quite possible the others like Google, Microsoft are using other CDNs like Akamai as well. It is highly unlikely they use only Fsly. Could it be that they have chosen Fsly for solving some specific use cases as opposed to all use cases which are commoditized? This is important to understand because the current euphoria about Fsly seems like it is set to displace Akamai and others.

Please note that I said "When there is no clear leader one has to depend on subtle clues." If Sherlock Holmes were alive today, he would be the world’s best computer program debugger because he relied on logic informed by clues. It’s quite amazing how similar the methods of crime solving and debugging are, look for clues, look for the evidence, and come to conclusions. But the clue is not the solution. Smorgasbord1 explains beautifully how the various players use the available alternatives. What the clues said to me is that Fastly is most certainly in the run and not a long shot. That’s all it did, it gave me the confidence to invest and watch like a hawk because no one can know the future. The clues just told me the odds are very good.

Denny Schlesinger

16 Likes

Gary Alexander, an author that writes for www.seekingalpha.com, wrote a free article that hit the wire yesterday afternoon, suggesting FASTLY may be overvalued. Link to article below

https://seekingalpha.com/article/4357373-fastly-stock-is-run…

I don’t know how good his insight is but I find it interesting that FSLY is up today about 8% (as of 12p ET). Some items that find interesting:

• FSLY’s YoY revenue growth rate (38%) and gross margin (58%) are among the lowest of the companies I follow (CRWD, LVGO, ZM, Alteryx, NET, DDOG, DOCU, OKTA, and COUP), yet FSLY investors are continuing to value Fastly. I find the comparison to NET (CloudFlare) interesting. NET is a company that competes for the same customers (more on that below). I find it interesting because NET has better revenue growth and gross margin numbers, as of last quarter: 48% and 78% respectively . The premise is that FSLY’s gross margins will improve in the future as Fastly’s costs to deliver internet traffic will be reduced, as a percentage of revenue over time as its scale improves.

• As mentioned above, FSLY and NET compete for the same customers, but FSLY targets the developers of those companies (those that are actually doing the work) whereas NET targets the CIOs/CISOs, who often approve the purchase. Peter Ofringa (www.softwarestackinvesting.com) suggests a developer focus is an advantage in that market (and I agree given my experience at Intel). Typically, good CIO/CISO’s often ask their developers & engineers for their SaaS recommendations and approve the purchase, not (necessarily) the other way around. That said, Muji is preparing his earning reports for FSLY and NET and will compare the two. I’m looking forward to his report.

For fun, I calculated EVSO ratio (using today’s prices as of 1215p ET, rounded to the nearest dollar) using Ron’s Oomph factor and Saul’s Oomph factor, that data below. Note I don’t put much value in the absolute number, rather, the order of the data is more interesting to me.


Company	EV/Sales	EV/S/Ron's Oomph factor	     EV/S/Saul's Oomph factor
CRWD	 $16.5 	               6.77 	                     5.75 
LVGO	 $34.3 	               10.03 	                     8.50 
ZM	 $57.3 	               11.47 	                     9.56 
AYX	 $30.0 	               16.13 	                     13.67 
NET	 $31.1 	               18.22 	                     15.18 
DDOG	 $54.9 	               19.63 	                     16.92 
DOCU	 $31.7 	               20.78 	                     17.46 
OKTA	 $39.3 	               23.66 	                     19.88 
COUP	 $43.8 	               28.98 	                     24.60 
FSLY	 $39.8 	               36.03 	                     30.03 

The fact that FSLY is popping again today suggests these type of analyses are more “fun” than material or statistically significant.

I hope you find this data useful.
Gary

5 Likes

The fact that FSLY is popping again today suggests these type of analyses are more “fun” than material or statistically significant.

These analyses are interesting but I’m hard pressed to discern any information which might help decision making. It curious to me that the weighting of companies in my portfolio approximate the order in the table. NET is an exception.

In Jan or Feb I compared the relative metrics for NET and FSLY and considering that their customer base overlapped I swapped FSLY for NET. Several deep reviews of FSLY have since been published which afford a better perspective on both companies. So I concluded that I had made a mistake, BUT before I could remedy matters FSLY announced earnings and moved up by more than 45%. I switched at that point and kept buying FSLY on the way up so I had more than a double. It has kept moving.

I think now FSLY has gotten ahead of itself so I cut back the holding by about 40%. Its now 4.5% of the portfolio. I drew this conclusion based on cautions embedded in the recent reports on FSLY by Poffringer
as well in writeups by Bert and by Beth Kindig. All point out that the "compute @ edge " technology, as well as FSLY’s low latency strategy has not been fully proven which leaves room for a future slowdown. We shall see.

I made the further error of concluding that NET and FSLY were interchangeable. They are not, but I’m not the best person to explicate the differences. I decided to leave the matter rest and to put investible cash in the big 3 Zoom,CRWD and DDOG. We shall see.

Cheers

Thanks Draj, sounds like you agree with Mr. Alexander’s thesis. BTW, I should have put the links to the posts related to the derivation of the “Oomph” factor; they are listed below. I agree (and most on this board agree) that they are more for entertainment and relative comparisons, but some use these type of assessments to help inform where new money could be invested in between quarters.

rdutt - created Oomph factor & EV/S/Oomph
https://discussion.fool.com/the-oomph-factor-34182263.aspx
Saul - on Oomph factor
https://discussion.fool.com/here39s-why-i-like-the-oomph-factor-…
https://discussion.fool.com/here8217s-how-i-calculate-the-oomph-…

2 Likes

The advantages of FSLY are explained well on the board here: https://discussion.fool.com/i-would-like-to-compliment-smorgasbo…

Offringa’s very detailed breakdown has been referenced above. Interesting that while he says there are uncertainties, FSLY is the largest position in his personal portfolio.

It also seems that FSLY is held by more people on this board than NET. Which given the metrics shared above I’m not clear on why??

However, I’ve put more into FSLY as it continues to perform well.