Unified communications implementations are difficult to effectively monitor and manage simply because there are so many components, and they’re not necessarily all from the same provider. Even if they are, at least some components were likely gained through acquisitions, so integration may not be quite as neat and clean as customer would like.
With this topic in mind I talked with Darc Rasmussen, CEO and Managing Director of IR, which makes proactive performance management tools for various areas of IT, including UC.
UC presents a particularly vexing monitoring and management challenge because of all the capabilities it integrates, Rasmussen says. The very phone call over which I spoke to him presented an example: he was calling in via his PC, I was on Skype and another participant was connected by a mobile phone.
“All this technology means if something goes wrong, you’re not trying to find a needle in a haystack, you’re walking into a field full of haystacks,” he says. “First you have to find which haystack it’s in, then you have to find the needle.”
That’s is what IR Prognosis is intended to do. It is an integrated management and monitoring platform that can deal with UC solutions from Avaya, Cisco, Microsoft and more. “It finds the needle and helps companies be proactive in avoiding problems, optimizing the ecosystem and getting aligned with the business about delivering better services,” he says.
Performance Monitoring Tool Eliminates Downtime, Saves Black Friday
Rasmussen provided two examples of how actual customers used Prognosis in their UC environments, the first being a large financial services organization. It operates a series of contact centers for its credit card payment business that serves retailers as well as consumers.
In October of last year, a few weeks before Black Friday, the busiest shopping day of the year, one of the contact centers went down. A week later, another went down. Being a large company, the firm’s systems are highly resilient, with warm backup facilities, so there should have been an orderly shutdown and transfer of systems to the backup without interruption.
But that didn’t happen. Systems went down and had to be restarted, which took up to an hour. Not good when outages cost $1 million to $3 million per hour, Rasmussen says.
Using Prognosis, the customer determined the cause was a bug in one of its software engines, provided by “a major supplier in the UC market,” he says. The vendor had a fix that required implementing a new version of its software.
Implementing new versions of anything wasn’t going to fly with Black Friday coming up. Instead, the customer used Prognosis to identify the leading indicators for when the bug would rear its head again. The system triggered an alert that gave the technical teams about 45 minutes to activate redundant systems, institute an orderly shutdown of the affected systems, then restart them.
This happened about a half dozen times during the busy holiday season but in each case the contact center was unaffected Rasmussen says. Pretty cool.
Using Performance Management Tools to Enforce SLAs
His other example showed how a performance monitoring tool can pay for itself by holding vendors accountable to their service level agreements and invoices.
This was a Fortune 500 company that implemented Prognosis to help it understand what kind of call quality it was getting from its UC voice and video solutions.
“It’s not just a question of whether there’s dial tone when I pick up phone, but what is quality of the interaction?” he says. “When I open a Lync session to IM, does it work the first time? When I have folks on a Skype conference call, is it working? What’s the quality of the voice and video?”
He notes that Skype for Business users have different expectations from users of the free consumer-grade Skype service, with zero tolerance for poor voice quality. A survey IR did of some 200 companies in 44 countries showed the top concern for implementing Skype for Business was managing the performance in a multivendor environment. “Number two was delivering an excellent user experience,” Rasmussen says.
This Fortune 500 organization was able to determine its SLAs were not being met by its UC service provider. They also gained clarity into who was using which services, and how they were being used. For example, were they really getting a return on the investment in videoconferencing to reduce travel?
“But where the investment really paid off was in reconciliation with the service provider on what was actually being used,” Rasmussen says. Few organizations know with certainty exactly what they’re being invoiced for by service providers, he notes.
“This very large organization suddenly had complete clarity around their users, technology and end points,” he says. “When they reconciled back to what they were being invoiced, there was a clear discrepancy. The discrepancy was so big it immediately paid for the investment performance monitoring.”
Yes, the stories come from a biased source, but I thought they were still viable examples of how performance monitoring and management tools can do a world of good for UC environments. And we’re always up for hearing more stories, so if you’ve got a good one, just let us know in the comments below.