Do no evil: Reject WLAN feature licenses

Feature licensing gets right up under my skin. When introduced to my company for the first time (before I started working here), finding out that they didn’t support a feature-licensed sales model was refreshing. Since moving from CWNP to a vendor 15 months ago, I’ve seen more and more of this feature licensing non-sense creeping into the market. It’s so bad at times that it makes me want to cry.

Aruba has been the reigning king of evil feature licensing for a number of years, but their enemy-for-life, Meru Networks, has recently decided that, “if they can’t beat them, join them.” Obviously Meru has seen how much money Aruba takes from customers with this unique form of evil, so now that they’re public, they figure they can do it too without any repercussions.

Let’s dig into the details of this evil licensing, shall we? There are lots of different kinds of licensing in Wi-Fi systems, and I hope that I’m not missing any in this list.

Management licenses: This is how a Wireless Network Management System (WNMS) is priced. It’s based on the number of APs under management. This is industry standard (read: normal) because you have to price management in some equitable manner. This seems as fair as any alternative I can think of.

  • Caveat: Some controller vendors will hit customers twice with this type of license by selling an AP management license for both the controller and the WNMS. This is evil.

Feature licenses: Enterprise-class Wi-Fi systems are, in general, fairly feature-rich … some more than others of course. Vendors single out important features and charge licensing fees to “unlock” these individual features in their system. It could be WIPS, VPN, Remote AP, N+1, video conversion, or any number of things. This is evil.

  • Caveat: Some features are licensed “per controller”, some features are licensed “per AP”, and some features are licensed “per client session.” Vendors call this “license consumption”, and it is very evil. There’s 100% chance that the sales representative for a company selling feature licenses knows more about their feature licensing scheme than the customer does. Who do you think is going to get screwed if there is any confusion?

PHY licenses: The customer buys 802.11n hardware, but it’s ‘licensed’ for 802.11a/g. This is marketed under the guise of ‘upgrade to a faster network when you’re ready while spending less money now’, but it’s really a Trojan Horse that screws the customer twice. First, the network design for 802.11a/g and 802.11n are completely different. Second, the customer isn’t in control of 802.11n pricing later. Just evil.

  • Caveat: Customers should be ready to redesign their network and likely move a bunch of APs around when they ‘upgrade’. Some vendors don’t enforce licensing…until they do. Some vendors enforce licenses per-AP and others enforce licenses per-radio. This is spectacularly evil.

Redundancy licenses: There are two parts to this type of license. Some vendors make you buy a “backup controller license” which isn’t as painful (read: expensive) as licensing everything that is licensed on the primary controller. Some vendors make you pay an N+1 license. Both are evil. Some vendors make you fully license everything on every controller, regardless of whether the controller is acting in a hot-spare capacity or as primary. This is so evil that it deserves an award.

I think that’s all of the license types, but I might’ve missed one or two. If you know of any I missed, please chime in. The more, the merrier.

Now that you know the types of licenses, let’s talk about how they’re sold. This is the only thing more evil than the licenses themselves. Since management licenses are common to all vendors, and there’s no real alternative, we’ll skip this one as “industry standard”, but we’ll definitely focus on the others.

Specifically I want to talk about “split CAPEX”, which is a method of selling the customer “a little in the beginning” and “a bunch later.” Neither are optional purchases.

Evil License Selling Strategy #1: time-bombed licenses

I’ve seen this “split CAPEX” strategy happen over and over as we’re competing against vendors who sell feature licenses. It basically amounts to a Trojan Horse: “just get in the door now so that you can rape and pillage later.” There’s a massive price differential between permanent licenses and less-than-permanent licenses (which were recently introduced solely to “get in the door” when competing against more attractive solutions). After the expiration period, the customer gets KILLED on re-licensing all of those features, and now that the equipment is deployed in the customer’s network, the vendor is unlikely to give the customer any discounts. Why should they? So…the customer pays for the time-bombed license up front and then pays list price for the permanent license on the back end. Ouch. That’s seriously evil.

Evil License Selling Strategy #2: “Accidentally” leaving out necessary licenses

I recently wrote an internal blog (that our marketing folks won’t let me post publicly because it’s too …um … blunt) called simply, “Wi?” It’s about vendors selling half of a Wi-Fi solution. In order to be competitive against controller-less solutions, controller vendors must leave out half of their network. When you look at their quote, many, if not MOST of the licenses that make them “feature rich” aren’t listed – even when the customer’s requirements specifically require those licenses. They’re hoping that the customer will trust their sales guy enough to not ask questions until they have the initial PO safely in-hand. That “feature rich” controller that the vendor told the customer about doesn’t actually exist until the customer buys lots of feature and PHY licenses. Evil … just plain evil.

Evil License Selling Strategy #3: “Accidentally” leaving out necessary system components

This one is a biggy. A friend once told me, “one database isn’t a database.” He meant that without redundancy, the chance of failure was high and that you’d soon be facing ‘no database’ because it had been lost. The same is true in Wi-Fi. Wi-Fi without redundancy isn’t Wi-Fi. Redundancy should be inherent, but when the architecture lacks inherent redundancy, there are band-aids (such as hot standby, N+1, or clustering) available that allow for the addition of the needed redundancy – at a price of course. If there’s any mission criticality at all to your Wi-Fi infrastructure, you should either go controller-less or buy one of these band-aids…seriously!

The evil comes in when you see that controller vendors quote their system with zero redundancy to lower initial CAPEX. They KNOW that the customer will eventually add this, and “so what if they don’t have any redundancy for a while.” That’s just wrong. Their quotes are missing a redundant controller and all of its feature licenses. Their quotes are missing the WNMS because a single controller can act as a WNMS in some instances. By leaving out the second controller, and thus the WNMS, they don’t have to charge you for WNMS/AP management licenses until later either. It’s all evil trickery to lower initial CAPEX.

Evil License Selling Strategy #4: Selling support for licenses

Dude, you gotta be kidding. Selling a support contract on a license? I have in my hand a copy of a competitor’s quote where they sell license support based on a period of time (1 year in this case). What in the world is this for?…perhaps in case a license breaks and need replacing? Evil, evil, and more evil. Where’s the love?

Evil License Selling Strategy #5: Renaming a license and selling it again to the same customer

This is just wrong…and silly. Mr. Customer buys a firewall license from Company-A and says, “self, this works pretty good…I’m happy.” Then, he finds out that his firewall has been deemed “last generation” and in order to have his firewall “upgraded” he needs to buy the same license again because it’s been renamed, “Firewall Next Generation.” Now he says, “self, this is FU, and I’m pissed. I want my upgraded firewall for free because I’ve already paid for it AND I’m paying that stupid annual support fee on my license also!”

When will it end? Methinks when customers get fed-up with it and say, “NO MORE!”

This public service announcement has been brought to you by your friendly neighborhood Wi-Fi Watchdog: the Devinator. Do no evil. Reject feature licenses.

SHARETweet about this on TwitterShare on LinkedInShare on FacebookShare on Google+Pin on PinterestDigg thisShare on RedditShare on TumblrShare on StumbleUponEmail this to someone

Devin Akin is Chief Wi-Fi Architect at Aerohive. Devin has over 10 years in the wireless LAN market and over 15 years in information technology. Devin's background includes working as a network design engineer for EarthLink, AT&T/BellSouth, Foundry Networks, and Sentinel Technologies as well as working as an RF engineer in the US Army. He has authored and edited several books with Wiley-Sybex and McGraw-Hill and holds some of the industry's most esteemed certifications, including CWNE, MCNE, MCSE, CCNP, CCDP, CCSP, and INFOSEC. He is considered an authority on Wi-Fi.