What’s the point in buying a videoconferencing system?
Zeus Kerravala, analyst, Yankee Group: I’ve long been been a video skeptic. A lot of people think of it as the solution to no problem. If you’ve ever been on a bad video call, it makes the experience worse (than a phone call). However, if you’ve ever been on a really high-quality video call, you can start to understand the impact it can have. It creates a much more personal experience. People do tend to retain information better when it’s visual instead of just audio. It can actually add a lot to the collaborative nature of what you do.
Until corporations start to mandate that people use it, they won’t use it. But once they do, it’s got a lot of stickiness to it.
I think you can tell a lot from people’s body language. It can make a lot of sense in negotiations. It creates a lot more of a personalized experience. When you truly want to collaborate with somebody, get some feedback, that’s when it makes sense.
Philippe Giraux, director of consulting services, CGI Group Inc.: The types of organizations that are looking into these types of solutions are quite varied and more and more diversified. It could be increased productivity for their management staff, to decrease travel time. Any organization that’s decentralized will usually benefit from videoconferencing solutions. You could also look at hospitals that are looking to videoconferencing from a diagnostic perspective as well as for consulting.
How much is it going to cost me?
Kerravala: I’ve talked to some companies that have brought in the high-end teleprescence solutions and the way they’ve paid for it is they’ve just slashed the travel budget. They’ve used it instead of a corporate jet. But that’s a very small set of people. I’ve talked to other firms. I think it depends on your level of use. How much you spend on it depends on the potential impact. If you’re simply going to use it for briefings, I wouldn’t spend a lot on it because the potential impact is going to be small. But I talked to one pharmaceutical company and they said it’s cut their development time in half because they can get people together much quicker. They were prepared to spend a lot on a teleconferencing solution. With drug companies, you can’t put a dollar value on development time.
William Stephanos, director of consulting services, CGI Group Inc.: There are several alternatives. You can purchase the equipment and run the service yourself, or you can rent the service from a third-party, in other words outsource it.
If you’re going to purchase the equipment – there are different components. You have the end points, the bridging system and the management scheduling interfaces. For bridging, you would be looking at prices ranging from $2,000 to $5,000 per port. If you were going to outsource it would be 60 cents to 75 cents per minute for an unassisted conference. For an assisted conference with an operator, it would be 80 cents to a dollar per minute.
There are many different ranges of devices with different numbers of ports. So depending if you’re buying a 12-port or 24-port device, then you’d just multiply that by the price per port. That’s only for the bridge or the multipoint control unit. Then you have to add the end points, the software controller and the scheduler. That could add up very fast to a minimum of $50,000 to $100,000. You can do the math If you’re going to outsource the service.
Whom should I buy from?
Scott Hollows, video-mediated communications analyst, Ontario Institute for Studies in Education of the University of Toronto, Knowledge Innovation and Technology Lab: There’s an advantage to every system out there and instead of picking one that’s the absolute best way to do it, there is no one best tool. That’s one of the reasons (our lab) requires an operator to handle the wine list of opinions.
Kerravala: As far as technology and leadership goes, companies leapfrog each other. I think right now, Cisco has a lead over HP’s Halo, but it’s a new solution. When HP announces Halo version 2, they’ll have a technology lead. You should never buy technology based on who’s got leadership at any particular time, because it always changes. What I would look it is total cost of ownership. How much does it cost? How easy is it to integrate into your network? How much does it cost to maintain it? How much bandwidth does it use up? What does the vendor offer in terms of support services? What does your room look like? How is the lighting and acoustics? All of these things impact the experience.
What’s the ROI?
Giraux: Typically your justification for buying a system would be reduced cost of traveling, increased efficiency. These would be kind of soft costs so they’re difficult to put down in a hard figure fashion. It’s also a question of culture. Some organizations have a very face-to-face culture, whereas others don’t mind being a little more high-tech. The more low-tech the business they’re in, the less likely they are to deploy these types of solutions.
Some companies are taking an environment stance and trying to gain an image as an environmentally-friendly company. By deploying video-conferencing and reducing the travel which produces emissions, it would help with that stance.
Hollows: I think our first system paid for itself in the first year just based on saved air fares. Our first system was when SARS hit Toronto, so nobody was interested in coming to Toronto. Overnight, everybody was interested in giving this videoconferencing thing a try. I think with the combining of goals and interests with limited budgets, more and more people are coming together regardless of geography. You’re getting committees and groups getting together regardless of where they are. You’re find that more and more with universities – they’re developing things together instead of in competition.
Can I expect bandwidth issues?
Kerravala: If you’re running it as a shared service on your network with other data applications and maybe even voice, then you’ve got to make sure you have enough bandwidth. You have to have a running QofS and make sure that traffic is separated from others. What I’ve seen some companies do is they run video on its own separate network. In fact, that’s the way Cisco’s Telepresence works, for the most part. While it defeats the object of convergence and it costs more, if you get a lot of impact out of it, then the cost of the network itself becomes nominal. If you’re using in a teaching hospital, you run it on its own network. If you use it for a trading desk or something, you can get away with running it on a shared network.
The worst thing you can do is bring this thing in, have it perform poorly then have it impact the performance of all of your other applications, because then no one will use it.
Stephanos: The recommended bandwidth per channel, let’s say if you’re using an H2 64 compressor, for a room-based video-conferencing solution, it would be at least 192 kbps for audio and for video. And at least 384 kbps if you need audio, video and data collaboration sessions. Then at least 1 mbps for high definition. We’re talking about high bandwidth unless you’re talking about PC or desktop videoconferencing with a one-to-one connection where the bandwidth is probably a little smaller.
The old technology used to use ISDN connections . . . which would offer 64 kbps per channel then you would bundle the channels to provide adequate bandwidth for your specific applications. That’s the old legacy system. If you’re going to use IP networks over packet network, like IP VPNs,(virtual private networks) definitely the technology that would be recommended would be MPLS (multi protocol label switching). All the different carriers offer that today. Of course, you’ll have to pay for that service, but they will be able to provide the necessary bandwidth as well as taking of delay and jitter and all of those kind of things.
You can also use Layer 2 technologies which are Ethernet based.
How much training is involved for the user?
Kerravala: One of the largest things to overcome is people thinking, ‘Oh no, people will see me a home working in my pajamas.’ For the most part, once you start using it, especially in a corporate setting, you don’t think about it anymore. It’s just like meeting someone (face to face).
I think the social barrier might be the biggest one to overcome, but once people start using it, it’s not so much of an issue. People first think, ‘Well, I don’t want my boss to know what I’m doing,’ but that used to be the way it was with instant messaging and cell phones. They didn’t want to be available 24×7 but now they want the technology.
I think the systems have become a lot of easier to use.
Hollows: It can depend. You get the whole range of user experience, knowledge and comfort. Going along with that is the level of user interest in gaining knowledge. There are some people who are happy to have me pulling the levers and operating things for some and they’re happy to calling the shots, but there are others who want to touch the controls and they know they’re doing.
Some people need video-conferencing etiquette: understanding that there can be a lag; if everybody talks at the same time, it can garbage things up.
How much support do they require?
Stephanos: If you’re going to look at conferencing room systems, especially the legacy systems that we’ve seen, they require quite a bit of effort to support, in terms of configuring them, making them operational . . . sometimes they don’t work, sometimes you have to put some options in to make them work properly. The typical person on a conference call wouldn’t be able to handle these configurations, so you would need someone who’s knowledgeable in the organization. Sometimes it does require human intervention.
Hollows: (Our lab) is one of a kind in Canada, so when you have a problem, it can be difficult. I don’t want it to sound like the support networks aren’t there, because we just had an issue and they were fantastic, but the initial question was, OK we have an issue and we have to figure out how we can support this. It’s not something that seeing instant and everyday service from technicians, but when we needed it, they were there. It takes a chunk of time to support it.
The frustrating effect it that when it goes down it has a domino effect because you the space without anybody in it to fix it. If you have two or three more events you have to come up with a shoestring workaround to bridge the gap to when to you get near the equipment again. It’s the balance of uptime and maintenance time I guess. Anything you do preventative-wise is good. But it’s the same with any Microsoft problem: you get a patch and it fixes one problem but creates two more.