Here’s my presentation entitled, “Video CDN Trends: Latest Pricing, Customer Challenges and Growth Opportunities” from the Mile High Video workshop event this month. Happy to answer any questions in the comments section or you can email me directly at dan@danrayburn.com. Apologies for the rough voice, had a cold. #mhv2020
Sponsored by
Recently Cloudflare announced that their researchers have been working to improve internet privacy and security through a DNS protocol called Oblivious DNS over HTTPS, or ODoH. While the announcement suggests that it will improve internet privacy, their proposal can actually lead to significant privacy issues, if it’s adopted. In addition, as it is written today, ODoH is likely to seriously impact internet performance or force providers to invest in software and process changes because it strips information that ISPs and CDNs require to do effective and efficient mapping to ensure performance for their users.
The IETF is an internet standards body made up of an international community of network designers, operators, vendors, and researchers concerned with the evolution of the Internet architecture and the smooth operation of the internet. Community members bring proposals for internet standards to the IETF through published “drafts”, mailing lists, and typically through in-person meetings. Proposals are submitted, and require working groups to pick them up in order to progress them into standards.
At this point, ODoH is an early-stage draft proposal which has been discussed in the “dprive” (DNS Privacy) working group but not yet adopted there, meaning that it is in a formative stage. Common practices to encourage the IETF to adopt work in a working group include deploying production implementations of proposed technologies, and driving interest and support from vendors and businesses, which is what Cloudflare is doing with their announcement.
So why is there contention around the proposal? Cloudflare’s head of research Nick Sullivan has stated that ”sweeping technical changes to the internet will inevitably also impact the technical community. Adopting these new protocols may have legal and policy implications.” Some of these legal and policy implications are detailed in a blog authored by Akamai Fellow Erik Nygren, back in 2018 on what encrypted DNS means for the Internet as a whole.
ODoH is an extension of DoH, so let’s start with that. The protocol exists because DNS queries are sent in cleartext. This means anyone on the network path between a user’s device and the DNS resolver can see both the query that contains the website the user wants to visit, as well as the IP address that identifies their device. Both protocols are designed to increase user privacy by preventing queries from being intercepted, redirected, or modified between the client and resolver – something known as a middle (MiTM) attack. DoH encrypts communications from the client to its resolver, and ODoH takes this a step further to obscure the client from the resolver.
DoH itself is just a protocol for doing DNS lookups over HTTPS. Most of the contention comes from ways in which DoH resolvers might be discovered and configured. For example, when DoH is used at the application level, it can bypass name servers configured at the OS level. So a web browser can come with a list of DoH compatible resolvers already configured, and traffic from that client would then use those DNS settings.
In 2019 Mozilla turned on DoH by default for all Firefox users, using Cloudflare as the server. This meant that the browser would prefer DoH via Cloudflare. This has been heavily criticized as an anti-privacy move, since Mozilla is essentially handing off all DNS resolutions to a single for-profit corporation. ISPs expressed concerns over their ability to perform lawful interception and content filtering (for example, legal requirements or parental controls), and many feel that since Cloudflare is an American company, this could not only be centralizing a large portion of the internet but also making it subject to law enforcement from a single government.
To explain this, switching from using an ISP’s local in-country DNS resolver to a DNS resolver that is out-of-country could make both privacy and performance worse rather than better, regardless of what communications transport is being used for the DNS. Consolidating DNS lookups to a few services also introduces new risks for enabling the correlation of user activity, and these services potentially become highly attractive targets for subpoenas and extra-legal attacks. This is all made substantially more challenging as many users lack a way to judge the level of trust they have with various DNS service providers, making it hard for them to make an informed choice.
So Cloudflare’s ODoH announcement is their way of asserting that they will provide users with an option not to send IP information to their DNS resolvers, which Cloudflare claims will ensure privacy. But this introduces another challenge: that ODoH will impact performance because it introduces significant latency, and also strips information that is required to do effective and efficient CDN mapping. Cloudflare’s research paper about ODoH provides testing conditions that are not relevant to the real-world, and the performance impact numbers mask latency introduced by load times for real web applications.
There are a wide range of other options being explored by the IETF. Given the wide variety of use cases, there is not likely to be a single solution. Many of the leading proposals involve a mixture of sources for secure DNS resolver configuration, device policy, associated and designated resolvers, and user choice. Something like ODoH may fit into this for resolving names where performance may be traded off for possibly improving privacy, but at this point, the concerns and potential pitfalls seem to outweigh the benefits.
Here’s a rundown of some interesting news over the past few days with links to the stories on LinkedIn where discussions are taking place:
1. You have to offer something better than others in the market or solve a problem. In other words, you must have a “competitive advantage” in some way.
2. Good execs hire people smarter than themselves. Quibi hired a lot of very smart people, on many fronts, but then didn’t listen to them.
3. Collecting the right consumer data will tell you how consumers want to engage with OTT services. We have plenty of this data in the video world but Quibi ignored it.
4. The power of any mobile content offering is sharing. By not allowing viewers to act as the marketing vehicle for Quibi’s content, Quibi threw away the biggest value of being on mobile.
5. Calling the service Quibi, a name many could not spell or pronounce, results in higher customer acquisition costs due to the education that is required.
6. Marketing the service, instead of specific shows is a mistake. Consumers don’t care about the “channel”, but rather the content. Don’t highlight the “length” of the show, highlight the content. We all know content is king.
7. You have to set the proper expectations with advertisers. 22 brands bought $150M in advertising, #Quibi sold them on the metric of “reach”, but didn’t have it.
Math doesn’t lie and no matter how you ran the numbers, from a P&L standpoint, it would not lead to a profitable outcome. The content costs were too high and the fee they could charge consumers would always be too low. 100% of their revenue was generated from one service, without any revenue diversification like other OTT services have. When you have a plan that is setting yourself up for failure from day one, with a business model that doesn’t work, why should they be congratulated for “trying”? You can’t spend well over $1B to get a service off the ground and then “learn on the way”.
Quibi wasn’t a “startup” with a few dozen people and they ignored their own employees feedback, many of whom were the exact demographic they were targeting. Whitman has said the role of Quibi was for use “waiting for a doctor’s appointment or standing in line at the bank.” In those instances, consumers were not lacking video viewing options. TikTok, Instagram, Facebook, sports highlights, YouTube or the pause button, on any streaming service, works just fine. Qubi was looking to solve a problem that does not exist in the real world.
In all industries change happens and you need to be ready to adjust, even if you don’t know what the change will be. Some employees say there was no “what if this doesn’t work scenario” discussed internally. It can’t be an “all or nothing” approach. Lack of communication kills companies and you have to put in place a strategy to pivot, when the time comes, hopefully proactively instead of reactively.
With Black Friday almost here, I’ve compiled a list of the best deals when it comes to streaming media devices. I’ve not included pricing for the new Xbox Series X/S and PS5 gaming consoles since inventory is extremely limited and pricing varies based on all the bundles offered.
Roku
– Roku SE for $17, Walmart exclusive
– Roku Streaming Stick+ for $30 ($20 discount), direct from Roku and other retailers
– Roku Ultra for $70 ($30 off), direct from Roku and other retailers
* Aside from the Walmart exclusive, all other deals start 11/20 and ends 11/30 or while supplies last
Amazon Fire TV Stick/Cube
– Fire TV Stick Lite for $18, ($12 discount), direct from Amazon (Nov 20-27)
– Fire TV Stick for $28, ($12 discount), direct from Amazon (Nov 20-27)
– Fire TV Stick 4K for $30, ($20 discount), direct from Amazon (Nov 20-27)
– Fire TV Cube for $80, ($33 discount), direct from Amazon (Nov 20-27)
Google Chromecast
– Chromecast (not 4K) for $19 at Walmart
– Chromecast with Google TV for $40, ($10 discount), Google Store, Best Buy, Walmart and others
– Chromecast with Google TV for $90 (comes with 6-months of Netflix), direct from Google Store
– YouTube TV is offering a free Chromecast with Google TV, once you make your first payment for YouTube TV of $65 (Note the offer is only good for first time subscribers and ends December 31st, 2020)
Apple TV
– Apple TV 4K 32GB for $169, ($10 discount), via Walmart
– Apple TV 32GB for $144, ($5 discount), via Walmart and B&H
TiVo Stream 4K
– Retails for $50, no details yet on any discount being offered by any retailer (Amazon reduced the price by $3 on Amazon Prime Day)
During Akamai’s Q3 earnings call (transcript), the company referenced doing away with overage pricing and how that was allowing their customers to have a more predictable spend with Akamai. Due to the holidays and in particular with retail customers, it’s common for customers to see some big peaks in their business, with regards to traffic. While Akamai discussed how this helps better adjust traffic and spend for certain customers, the biggest advantage for Akamai in doing away with overage charges is really a competitive one.
Zero Overage Fixed Fee (ZOFF) pricing, as Akamai calls it, provides a construct where as long as their customer does not exceed their traffic commit by 2x or more for multiple consecutive months, they will not pay any overage charges. So, if there are marketing sites that don’t have massive bandwidth requirements, non-core apps that are not susceptible to bursting, or APIs that are being distributed and protected by Akamai, they can be included in a ZOFF contracting structure without their own separate traffic commits and bills. This effectively allows customers to add additional delivery services to their existing commits with Akamai, at no additional cost or commitment.
Akamai has been quietly changing their pricing strategies to be better aligned to the company’s customer base, with flat fee pricing for large media streaming customers based on subscriber/download volume, and increasingly a zero overage model for businesses that primarily monetize via websites, apps, and APIs. While Akamai says this strategy was designed to de-risk adding any internet-facing application to the platform for fear of incurring significant traffic overage charges, it actually serves an even more important benefit from a competitive standpoint.
Doing away with overage fees helps Akamai keep more traffic on their network from customers who are growing, which in many cases, may have previously been offloaded to a competitor. CDNs would specifically target Akamai customers and tell them to send traffic to their network instead of Akamai’s, the moment the customer hit their bandwidth cap with Akamai. The selling point being that the customer would not have to pay any overage fee to Akamai and would get a lower price point from the competitor as they grow their traffic. By Akamai doing away with overage pricing, that selling proposition by competitors disappears and makes it harder for them to get their foot in the door to get a slice of Akamai’s business.
With this year being marked by unpredictable traffic patterns and pressure for businesses to find savings with their IT vendors, Akamai says they have seen the strongest penetration in commerce, financial services and healthcare verticals, in adopting the no overage pricing strategy. Some may wonder why the media vertical is not called due to OTT video consumption and the reason is because most media contracts for the delivery of video and software downloads haven’t had an overage pricing component tied to them, across the industry, for many years now.
This isn’t to say that all overage type fees have disappeared completely from the CDN industry. There are cases where a customer can be hit with an additional fee if a certain percentage of their overall traffic volume, based on a specific country or region of the world, falls short of what they committed to. But that’s really a fee tied to a specific region, as opposed to overage fees that in the past were simply tied to the growth of a service. Akamai’s smart to have done away with overage pricing, more from a competitive standpoint than anything else. Some might argue Akamai is losing revenue from overages, but that’s short-sighted thinking since there is a greater opportunity to generate more revenue over time, from keeping new traffic on their network.
Here’s a quick recap of what you need to know from Q3 earnings from AT&T, Comcast, Microsoft, Akamai, Fastly, and Limelight. I’ll post another roundup from Apple, Amazon, Facebook, Google and Twitter shortly.
AT&T Q3 Earnings: Lost 590,000 premium TV subs, (37,000 of which were AT&T TV NOW subs). Grew HBO Max to 8.6M subs. A record high 357,000 AT&T Fiber net adds and 158,000 total broadband net adds.
Comcast Q3 Earnings: Lost 273,000 pay TV subscribers (down from 477,000 in Q2); 22 million sign ups for Peacock TV (up from 10M in Q2); Revenue of $25.5B (up from $23.7B in Q2).
Microsoft Q3 Earnings: Revenue of $37.2B, up 12% y/o/y; Intelligent Cloud revenue of $13B, up 20% y/o/y. Office Consumer products and cloud services revenue increased 13% and Microsoft 365 Consumer subscribers increased to 45.3M.
Akamai Q3 Earnings: Total revenue $793 M, up 12% y/o/y; Web Division revenue $418 M, up 8% y/o/y; Media and Carrier Division revenue $375 M, up 16% y/o/y; Cloud Security Solutions revenue $266 M, up 23% y/o/y. Full year revenue guidance of $3.164B to $3.189B for 2020.
Fastly Q3 Earnings: Revenue of $71M, up 42% y/o/y; GAPP operating loss of $23M; Capital expenditures of $14M, or 20% of revenue; Adjusted full-year 2020 guidance to $288.2M-$292.2M ($8M from Signal Sciences acquisition).
Limelight Q3 Earnings: Revenue of $59.2M, up 15% year-over-year and 1.1% quarter-over-quarter. Reported a GAAP net loss of $4.0M. Leaving full-year 2020 guidance unchanged at $230M-$240M. 20 customers accounted for approximately 79% of their revenue so far this year.