/var/blog
Showing results for 
Search instead for 
Do you mean 

Flash Markets: Why SD-WAN might not be the market maker many hope

by Trusted Contributor on ‎04-25-2017 07:00 AM

In 2011, it was SDN. A couple of years later, it was NFV. We saw it with overlays. To some extent, we are seeing it now with DevOps. And for anyone watching the networking space intently, we are now seeing it with SD-WAN.
 
SD-WAN is the next big market maker for networking. Or is it?
 
SD-WAN in a paragraph
 
For those of you who might be new to SD-WAN, I will describe it very briefly. Essentially, it’s WAN gateways doing some DPI and then making policy-base routing decisions with the intent of supporting hybrid WAN. Enterprises can send important traffic over expensive MPLS links and less sensitive traffic over normal broadband links. The premise is that this will reduce connectivity costs for enterprises with distributed sites (could be campuses, small or remote offices, or even kiosks with a point-of-sale device). Sometimes, companies add WAN optimization to it. They frequently include secure transport (DMVPN). And the whole thing is managed from the cloud.
 
I have taken some shortcuts here, but this spiritually gets you up to speed. 
 
A net-new market?
 
Depending on who you believe, the SD-WAN market will grow to be huge in the next couple of years:
  • IDC claims $6B by 2020
  • Markets and Markets says $9B by 2021
  • Research and Markets says $7.5B by 2021
  • IHS puts up a more modest $1.3B by 2020
 
Even a cursory search will yield a bunch of results. The point is that this thing is going to be HUGE. 
 
First, you have to understand that this is basically the WAN gateway (or branch router, or ISR, or whatever you know the market to be called) market. It’s not like people will continue to buy into their existing solutions and then layer SD-WAN on top. Rather, they will simply substitute the SD-WAN solution for whatever they previously used. 
 
If you believe that this is a net-new market, the first question would be: how will all this TCO be saved if buyers have to support the old and procure the new? It’s not new.
 
In fact, not only is it not new, but it will be flat spend at best. If the procurement costs are more, then the hybrid WAN connectivity savings has to be enough to justify both the added spend and the risk. While the connectivity costs might go down, it’s difficult to imagine that the Cisco sales rep isn’t going to just offer a discount to keep the account captive when she hears that another vendor is making inroads. 
 
So no, SD-WAN is not a new market, nor does it represent a significant expansion of spend.
 
What is a flash market?
 
You could argue that SD-WAN will be a premium service. If it is a direct moneymaker for the carriers, then the suppliers can reap the rewards on the back end, right?
 
The challenge here is less about SD-WAN and more about how providers sell services. Let me explain a bit.
 
Imagine that the market is dominated by three carriers, A, B, and C. All three of these carriers buy equipment and software from the same vendors. They offer more or less the same services. And historically, they have really only been able to monetize connectivity services.
 
Of course, connectivity has become a commodity game over time. Aside from very local issues (like cell tower locations, or access infrastructure), most people have exactly the same service regardless of which carrier they use. 
 
At steady state, users are pretty locked into their provider. The only thing that makes them move is a major marketing play (unlimited data), until that play is matched by the competitor. Maybe one carrier gets exclusive rights to sell a handset (why AT&T with the iPhone was a big deal back in the day). Or maybe a new service is introduced (like SD-WAN).
 
When the new service comes out, it’s the only thing that gives users a reason to leave their provider. This is why new services are so important. So for 6 months, carrier A can pull from carriers B and C. Of course, B realizes this and immediately starts its own effort. Six months later, B has the ability to stem the flow to A, and they can participate in the continued pillaging of C’s subscriber base. At this point, C is playing defense. The new service is not going to attract any net-new subscribers, but not having it certainly puts them at a disadvantage.
 
So even if the new service persists forever, it’s really only of value until the major competitors that form the subscriber pool are at parity. In this case, the market only really exists for a short period of time. This is a flash market. 
 
Heavy first-mover advantage
 
If subscribers are really only mobile between carriers while the service is new and not yet fully adopted across the industry, there is a heavy first-mover advantage.
 
In fact, because subscribers tend to only leap at specific points, once a carrier has a subscriber, they retain that subscriber for awhile. This means that the right way to value a service that attracts new users is not purely based on that service revenue but on the lifetime value of the customers it attracts. Put more clearly, the value of my SD-WAN service is however much I get you to pay for it, plus whatever else you spend with me over your account lifetime. 
 
Given this, it is arguably more important that services come fast than they be cheap. I would happily lose money on a service if it meant I could sell you other services. This means that carriers can rush a service to market, and then spend the next however many years trying to make that service less expensive.
 
SD-WAN a premium service?
 
So this leads to an interesting question: is SD-WAN a premium service? In the early days, that is absolutely the case. But when the whole market has the service, will that remain?
 
It seems more likely that once the service is ubiquitous, it likely becomes a commodity add-on to broader connectivity services. If this is true, there is a real question about how money will flow through the ecosystem in support of the service. Indeed, while the service is directly attached to revenue streams, it can be monetized directly and aggressively. But when those revenue streams become smaller, what happens to a carrier’s willingness to pay?
 
I suspect the answer here is that it goes down. Which means that the euphoria about SD-WAN that people are experiencing today is likely to subside over time and relegate SD-WAN to yet another service that once promised lofty new markets.
 
SD-WAN will survive
 
Don’t take any of this as a suggestion that the technology will not survive. I think SD-WAN is real. I believe it will continue to play a role in distributed enterprise architectures. But it is unlikely that it will remain a premium service driving net-new growth in the industry. 
 
An interesting side effect, though, is that if the profit pools get drained over a relatively short time horizon as part of the natural dynamics of flash markets, what will happen to all the companies who base their entire strategy on the service remaining a high-margin one?
 
SD-WAN will create a lot of corpses
 
Ultimately, SD-WAN has created a lot of buzz. A lot of companies are pursuing the promise. And most of them will end up dying as the directly monetizable service slides into commodity. Only the ones who are able to build for the next flash market, and the one after that, and after that again will be able to build a sustainable business.
 
The fact that the primary route to market is likely to be through service provider only makes this harder. If it takes some time to get the volumes up, these companies might not see volumes until after the service is already commodity. That has the doubly bad property of delaying the sales volumes until the prices have already dropped—a doomsday scenario for sure.
 
If this leads to the untimely demise of a bunch of SD-WAN companies, the question for carriers is one of planning. If the SD-WAN supplier of choice is gone, then that creates a vacuum in the service portfolio, which goes back to SD-WAN as a table stakes connectivity service. In the absence of the service, it is a liability. 
 
This means that providers will need to open up the catalog to multiple suppliers to mitigate risk (and to create supplier leverage, of course). This will put even more competitive pressure on the suppliers, which will only accelerate the carnage. 
 
The bottom line
 
While SD-WAN is hot, it is not really a new market. As the broader WAN gateway market evolves to include SD-WAN functionality, it’s going to create short-lived battlegrounds for the carriers. And those battlegrounds will become short-lived battlegrounds for suppliers. But ultimately, when everything is settled, SD-WAN will be just a part of networking, probably having lost its fancy name as the functionality gets subsumed into what is simply normal operations. 
 
Understanding the dynamics here is critical as people evaluate how they will respond. Far too many people get caught up in the hype and ignore the likely repercussions. And that can lead to the worst of all worlds: shoddy planning without any chance of recovery as the rest of your competitors pass you by. 
 
The strategic play here is to bet on platforms. Treat the services as add-ons. And make agility the new TCO.
 

Announcements
Juniper TechCafe Ask the Author
Labels
About the Author
  • Mike is currently acting as a Senior Director of Strategic Marketing at Juniper Networks. Mike spent 12 years at Juniper in a previous tour of duty, running product management, strategy, and marketing for Junos Software. In that role, he was responsible for driving Juniper's automation ambitions and incubating efforts across emerging technology spaces (notably SDN, NFV, virtualization, portable network OS, and DevOps). After the first Juniper stint, Mike joined datacenter switching startup Plexxi as the head of marketing. In that role, he was named a top social media personality for SDN. Most recently, Mike was responsible for Brocade's datacenter business as VP of Datacenter Routing and Switching, and then Brocade's software business as VP of Product Management, Software Networking.
About /var/blog

Subscribe to /var/blog  RSS Icon

Follow Mike on Twitter