If your organization is considering implementing real-time video conferencing, one of your first questions should be: “Do we buy an off-the-shelf system or do we build one ourselves?”
According to Tsahi Levent-Levi, an independent analyst and creator of “What to Look For in a Live Video Platform,” a 10-part buyers’ guide web series about video conferencing technology, there are distinct differences between the two that you need to be aware of.
“When you build, there’s a lot more work you need to do and a lot more knowledge you need to have about the intricacies of video processing. You still need some understanding of the technology when you buy an off-the-shelf-system, but a lot less,” Levent-Levi said. “When you buy, you aren’t buying just the ability to run your service today. You’re buying the ability to run it in the future as well, with all the necessary upgrades, scaling, compatibility, and maintenance work that’s required on a day-to-day basis.”
Levent-Levi noted that some companies want to build their own system because they believe they can do it better. To succeed, he said, the company needs developers with a deep understanding of the technology and experience working with it, along with patience and time. Even if a company decides to build a system itself, it will be working with vendors for different components and needs to be wary of some of the red flags that indicate a poor choice. These include a lack of experience with Voice over Internet Protocol (VoIP) or video processing.
“Some vendors just lack the understanding, and the learning curve isn’t an easy one,” Levent-Levi said. “That should be the first telling sign, though somehow many miss it. I think that’s because many of the consumer video services are free to use, so the assumption is that implementation is trivial.”
Developers are the most expensive resource when building a system independently. “Building your own is much more expensive in the short term. It takes time and manual labor. Building also increases time to market, something many would like to avoid. Usually, the estimates won’t even include the things that need to come later, on an ongoing basis, to maintain the service,” he said. “Buying is considered to be more expensive to operate and almost free to start using. I’m not sure that’s correct. In most cases, for most companies, buying is probably the better approach.”
Levent-Levi’s first response when a client wants to build its own video system is to try to understand the reasoning behind the decision. If the client’s main incentive is money, he’ll challenge the decision and attempt to change the client’s mind.
He has unfortunately seen clients try to build a system independently and fail. “Sometimes they have a very good reason for building on their own, but with any development project there are risks,” Levent-Levi said. “My role in these projects is to reduce the risk — I can’t eliminate it. Sometimes, they try again. Other times, they close the project, and some companies have even gone out of business because they were built around that project.”
So you can see how critical the decision is between buying and building a video conferencing system — the success of your company may depend on it. If you’re at the decision-making moment in your video technology plan, you’ll find more valuable information in Levent-Levi’s web series, “What to Look For in a Live Video Platform” as well as in his previous blog post, Mini-Course Shows How to Choose The Right Video Technology Platform.
Tsahi Levent-Levi is known for his popular website, BlogGeek.me. He’s a 25-year industry veteran who has closely followed WebRTC technology and, for the past five years, has been a WebRTC consultant. He is cofounder and CEO for testRTC, a company offering testing and monitoring services for WebRTC applications. Tsahi has licensed software development kits (SDKs) to developers on which they build communication products. He worked at Radvision, which was a pioneer in video communications, and he’s been a developer himself as well as a product manager and chief technology officer.