Cloud-based UC's Double Lag
A look at three vendors' approach to the market reveals why cloud communications services still trail their on-premises counterparts.
Hosted UC services are now available from all major telcos, most minor ones, and from an increasingly diverse set of systems integrators, systems vendors, consultancies, resellers, and value-added distributors. However, the UC feature set available via many of these services has lagged behind that of UC solutions sold to businesses to deploy on premise. In the not-so-distant past this was because the platforms on which the hosted UC services were built were noticeably less advanced than on-prem solutions. This is still the case, but to less an extent than before. Let's take a brief tour of three hosted platforms to see how they've been modified to reach (or at least approach) parity with on-prem solutions and why there's still a gap between cloud- and premise-based UC.
Cisco has been on a tear, updating its cloud-based UC platform, Hosted Collaboration Solution. Version 9.0, introduced last year, brought it to feature parity with UC Manager, something that Cisco had long wanted to deliver. It also added contact center capabilities previously lacking in HCS, as well as various mobility features, such as number reach and clientless network-based FMC (fixed mobile convergence).
Version 9.1, released earlier this year, added integration with Cisco's CTX video end points, IBM's Sametime and Notes, and IMS-integrated mobile devices. Version 9.1 also rounded out the contact center feature set with outbound IVR (previously it supported inbound only) and new contact center management software, based on the VOSS software that Zeus recently wrote about. And version 9.2, also introduced this year, added multichannel capabilities to its contact center software (previously it was voice only) and expanded contact center agent scalability to 4,000 (previously it was 500-1,000 agents).
BroadSoft introduced BroadCloud to provide a fuller set of UC functionality (instant messaging, web conferencing, desktop video) lacking from the telephony-centric BroadWorks platform on which telcos have long built their hosted voice services. At first, BroadCloud was available as a BroadSoft-offered service, where telcos that deployed the BroadWorks platform in their networks could offer the wider set of UC features but only by reselling and/or white-labeling the vendor's BroadCloud service.
There were pros and cons to this approach. On the plus side, telcos offering a BroadSoft-based hosted telephony service could now also sell a richer set of UC features without having to actually upgrade their infrastructure. On the minus side, it made for a rather complex delivery model, with part of the overall hosted UC service based on infrastructure in the telco's own networks and another part based on infrastructure in BroadSoft data centers. This has now changed, however. Recent updates to the BroadSoft solution set now let telcos deploy the BroadCloud server in their own networks to offer the full range of telephony and UC services themselves. Or carriers can keep reselling the BroadSoft service. The choice is theirs.
Microsoft also has a rather convoluted story when it comes to hosted UC. Lync Online is Microsoft's own cloud-based UC service that can be adopted either with or without Office 365 and that service providers can resell. Some, like Telefónica, do resell, while others see Lync Online as direct competition and won't touch it with a ten-foot pole.
Then there's Lync Multitenant Pack for Partner Hosting, which I touched on in a recent blog. This is the platform that telcos are supposed to deploy to offer their own hosted Lync services. Version 1 supported IM and presence, but not Enterprise Voice for common PBX capabilities like call park, emergency services, and IP desk phone support. Version 2 corrects some of these weaknesses, but it still doesn't support voice features (call admission control, branch office resiliency, malicious call trace) that have long been in the iteration of Lync for deployment within the enterprise rather than for hosted services. And in some cases (as with call park, emergency services, response groups, and analog device support) integration with a third-party solution is required rather than the feature being inherent to the Lync Multitenant Pack software itself, as it is with the standard Lync software.
In part because of the feature set issues, in part because the multitenant solution is comparatively immature, and in part because they're concerned that Microsoft is becoming as much competitor as partner, telcos typically stick to building their hosted Lync services around the standard enterprise solution. But Microsoft's three-pronged approach to hosted services makes a certain degree of sense. Sure, it's sloppy, with different solutions offering different capabilities, at least for now. But services based on each solution have a different target market: Lync Online for businesses opting for the Office 365 approach to Microsoft Office as a service; Multitenant Pack-based services for telcos targeting small businesses and SMBs needing little to no customization; and Lync 2010/2013-based hosted services for large enterprises needing a customized UC service that scales to thousands of users.
What worries me, however, is the snail's pace at which carriers deploy and then upgrade their hosted UC services. I've been speaking with a number of the 40 or so telcos offering Cisco HCS-based services, Many are still running version 8.6 of the software and are in no hurry to upgrade to 9.x despite its enhanced feature set. They're not getting much in the way of pressure from their customers to deliver better mobility and customer support and other features that are central to the latest HCS software rev, so why make the investment right away?
As for BroadSoft, the company says there are 500 service providers offering telephony-centric services based on BroadWorks, but declines to break out how many of these are also offering BroadCloud-based services based on servers running in either the BroadSoft cloud or in the telco's own network. I've asked a handful of telcos about their plans to make instant messaging, web conferencing, and video capabilities available to customers subscribing to their BroadWorks-based services, and each time have been told, "No, we're just going to stick with the telephony-centric services. No need for BroadCloud now thankyouverymuch."
And in the case of Microsoft, I have yet to hear of a telco offering services based on Lync Multitenant Pack. (If you know of any please shoot me a line.) There's at least one telco offering a multitenanted Lync service that it sells to SMBs. But this is something they've built from scratch on their own and, at least for now, have no intention of replacing with the multitenant platform that has Microsoft's seal of approval.
The result of all this is a double lag between the UC solutions that developers sell businesses to deploy on premise and the cloud-based services that telcos sell businesses walking the hosted UC path.
The first lag comes from the time it takes the vendor to bring the hosted UC platform to something resembling parity with its on-premise corollary. The second comes from the telco not upgrading its hosted platform when the developer makes important new releases available.
I fear that this will result in hosted UC services being perpetually behind the UC solutions sold to enterprises for on-prem deployment. And it could stymie the adoption of cloud-based UC at a time when its adoption has the potential of hockey-sticking.
But I tend to geek out over features and parity issues. Businesses making UC service buying decisions are likely to be more focused on other issues, such as overall cost, op-ex purchases, ease of deployment, and adopting UC services from a telco, VAR, or other partner with which they already have a trusted business relationship. So the lag may be just that--a delay, but not necessarily a deterrence from businesses' continued adoption of UC on a hosted basis.