The wireless networking space is bustling with activity as established wireless network players are starting to announce and/or hint at their plans to change to a controller-less, distributed WLAN architecture.
The technical challenge of changing architectures is arduous, but it isn't the only (or hardest) hurdle controller-based WLAN vendors face in making this architectural shift. It will be exceptionally painful – if not impossible in many cases – for these legacy companies to adapt their business models to controller-less.
• Legacy business models depend on controller sales for as much as 40 percent of their revenue.
• The controller-less WLAN lends itself to a pay-as-you go model, dramatically reducing customer's initial cash outlay, which breaks incumbent vendors' expensive sales models.
• Incumbent vendors will alienate their installed base that invested heavily to deploy controllers that will become an unnecessary appendix to their WLAN.
• Software development costs will double because most legacy WLAN vendors will need to develop each feature twice – once for their old controller-based customers and once for their new controller-less customers.
New era, new WLAN requirements
Why has the controller-less Wi-Fi network become so appealing? Time has passed and networks have changed.
The controller-based approach to WLANs was the right architecture for the early 2000s when WLANs were a secondary overlay network, the Wi-Fi security model was broken and access point memory and CPUs were more expensive.
Today, however, Wi-Fi is becoming the primary access layer – a high-performance mission critical network. Controllers have become a bottleneck, a single point of failure and a costly and unnecessary layer of WLAN complexity.
However, as appealing as it might be, it's not easy for incumbent vendors to change course and offer a controller-less WLAN architecture.
One major reason is because incumbent vendors are going to be impaired by their legacy business models. For example, on average 40 percent of the incumbent vendors' revenue comes from these controllers (the other 60 percent comes from APs).
This revenue problem becomes even worse for these vendors because a controller-less architecture allows customers to pay as they go. In contrast, an incumbent vendor's business model forces customers to make a major upfront capital investment by purchasing their controllers at rollout.
Bottom line: Incumbent vendors would take a 40 percent hit to their total revenue by going controller-less. And they take an even bigger hit to their near-term revenue because their customers can spread out their purchases.
This revenue scenario is completely different for controller-less WLAN vendors because the optimal business model and go-to market strategy has been established from the outset.
The pay-as-you-go precedent
There is a familiar precedent where a vendor with a "pay-as-you-go" model came in and changed the landscape of an industry to the detriment of legacy vendors. The company was Salesforce.com (SFDC), and it successfully unseated competitors such as Siebel.
SFDC came on the scene and busted the on-premise software model with dramatically lower cost of entry and accelerated time to value. Siebel, an incumbent vendor with a premises-based business model that required a much larger initial investment by customers, wanted to protect its revenue base and was reluctant to change.
The controller-less pay-as-you-go business model is very similar to SFDC's: Companies can start small and expand to meet their needs because they aren't forced to step up and buy controllers.
Shifting WLAN business models midstream is unrealistic
It is not likely that incumbent vendors will be able to shift direction completely and offer a controller-less WLAN architecture.
It will be incredibly painful and disruptive to their business model, so most of these vendors will do it halfway. If they go full-bore, they are likely to have to completely restructure their sales forces and vendor-pricing, which is huge. Think SFDC: It wasn't so much that SFDC had a big technology breakthrough as much as the SaaS business model changed the up-front cost, and the on-premise software model couldn't make the leap.
It is unbelievably difficult to change to a pay-as-you go model mid-stream. Incumbent WLAN vendors have already established a sales force and a go-to-market strategy that is completely wrong for this model.
Alienating the installed base
If incumbent WLAN vendors decide to take the plunge and transition to a controller-less WLAN architecture, they risk alienating their customers.
Imagine this: Incumbent vendors have legacy customers, they suddenly tell their customers they don't need controllers. That is going to anger the installed base because they are being told their expensive controllers have become paperweights.
Many of these customers will not be willing to accept that they invested in unnecessary technology so they will be reluctant to embrace the new approach. As a result, incumbent vendors will need to support two sides: their installed base and new customers. This strategy can be twice as expensive because they will need double the engineers to implement all of their features in two completely different ways – once with centralized control and again with distributed control.
The future of WLAN will be a mixed bag
Many incumbent vendors will attempt to fully embrace a controller-less architecture, but they will only make it halfway because they will try to appease the ir installed bases while simultaneously capturing new customers.
They will, in effect, be shoe-horning a controller-less WLAN architecture option into their offerings, and that translates to a bunch of compromises because they will be trying to straddle fences. Only an upstart, which has been controller-less from the outset, can go the whole way.
David Flynn is CEO of Aerohive.