There is a lot of hype over SD-WAN at the moment, with movements by telcos and OEM suppliers investing in some of the new startups. And folks that have been in the “hybrid WAN” space are focusing on adding SD-WAN routing. So if you’re looking at changing your network the moment, what are the choices for SD-WAN currently?
As I see it, it currently boils down to three main choices, depending on who you are and what your stance is. But will you win from that choice, or are the SD-WAN suppliers going to gain at your expense?
So what are the choices for SD-WAN currently?
Small to medium enterprise market space
In this space the use of SD-WAN is going to be a no-brainer. I’m looking for the ability to control my network costs, take advantage of cloud apps (I’ve never known different if I’m one of the many startups), and keep my users happy. It wasn’t that long ago that the CTO was also in charge of choosing those Macs and iPhones for the two staff, so a hands on approach with minimal effort wins here, especially of it builds on something that I already know.
So likes of cloud managed wifi providers win here. Their extension of the Wifi access point management platform and their security gateways to provide VPN services is just what the business needs.. And the business doesn’t really care about the need to use other connections, but having that facility is great for when it grows a little more (having an SLA for those Skype for Business calls across the network would be good).
Don’t get me wrong, there are lots of players in this space, which is where the bulk of SD-WAN business will occur. But do they already have a footprint in a network, or are they something else which an overstretched person needs to learn?
As a market leader in the space that allowed me to get my foot in the door, I gain market share overall, and most importantly in a space which is fiercely competitive, that of SD-WAN.
As a customer, I get a solution that works, integrates with other things I have, and won’t be to much of a bother, unless a cloud update needs me to visit all my sites to reset and reload kit.
A mostly captive audience of a Telco provider
The next their of users are those that are coming to the end of their telco contract for a traditional MPLS network, and have heard from organisations using Gartner’s figures of up to 70% cost reduction by using the Internet in their network. They see this as a stick to beat the provider down on cost.. But in reality the cost and pain of changing a provider who is mostly OK, but doesn’t react to changes quickly or provide much visibility of what is going on in the network, is way to high to face.
These users are more concerned that there is a service, and someone to complain to when things break. If the telco was really annoying them, they would bite the bullet and move. But it’s easier to moan and get service credits than to face the disruption.
Here the Telco is going to roll-over and offer managed SD-WAN, if there is one in their portfolio (or get creative on MPLS pricing, which may still offer higher margin for a year or so).. This protects their infrastructure investment and customer base. It allows them to reuse the current circuits, and to charge new provisioning fees for the Internet element. What’s already in the ground stays in the ground, and they make it even more painful to move away as more circuits are involved. The margin here may drop overall, as there is considerably less on Internet bandwidth than MPLS, but they’ve stopped a customer churning, and given themselves 6 months to a year of service breathing space as implementation and “bedding in” occur.
If the telco has chosen their SD-WAN partner correctly, and have done automation to the hilt, then they may even save on their own cost of change, increasing margin here.
As a new SD-WAN prospect here, push on the prices as much as you can, and fix them where possible, because you’ll be with the provider longer than you expect.
The hard core do-it-yourselfer
The hard core do-it-yourselfer is the kind of customer who wants their specific requirements met, and has the teams to be able to support this now.
So they setup bake-offs, proof of concepts, and go through rigorous testing to make sure they get the best of breed environment that both their server and network teams agree (or at least don’t disagree too much) on. The suppliers assure them that as they platforms are programmable, they can do anything their APIs will allow from code, and possibly several other things beside.
Their network team insist they can code, and are set ambitious targets for them to achieve integration of the platforms, but with no DevOps experience they’ll rapidly fall behind, become disheartened and return to the old ways of CLI, which at least gets the job done. The two orchestration tools still stubbornly refuse to cooperate in the future.
And here the suppliers win again, as multiple suppliers get orders (although not as big as a single supplier solution).
And the customer? They get increased cost of deployment, longer payback times, integrationa and service challenges, and it still won’t deliver 100% of what they originally wanted.
And what of the future?
In the future the industry will continue to acquire elements of new markets, both to capture market share (expect a lot of unhappy users as their orphaned platform dies during the end of life process), and to acquire new market segments and technology. The big boys will get bigger, and the smaller ones will have a harder time finding news ways to be innovative as the market place stabilises.. Or they become niche, focused on one thing they do so well that others find it very difficult to displace them in the one specific use case they have substantial benefit.
If your looking at SD-WAN make sure you get the benefits, not just the supplier. Make your choices for SD-WAN currently work for you.