WebRTC is fast becoming an effective way for businesses to interact with clients through video conferencing. We recently received an enquiry from a company about the effectiveness of WebRTC for single video transactions and our discussion brought up some interesting points that we’d like to share as part of our WebRTC blog.

How Often does WebRTC Fail

When it comes to one-off video sessions, WebRTC is definitely a viable solution; it’s frictionless and it can be cost-effective to have the project as your preferred choice for video communication. We are stressing the “one-off” because that’s the kind of videos with the most problems. If your business model has a “repeat call” operation, then the failure rates tend to dramatically fall as your clients are more educated and / or better equipped. So in this article we discuss about the failure rate of your random client that will only need to use your service once and only once, let’s roll the dice…

How often does WebRTC Fail?

And by “fail” we mean that a connection between two clients is not able to be established. The expected call failure rate for a non optimized WebRTC infrastructure is about 10-15%. Anything above 20% would be worrisome.

WebRTC can have up to 15% failed calls on non optimized platforms

By “non optimized” we mean that you are a startup that has just finished building your WebRTC infrastructure on your own, Turn & Stun servers, client code, signaling mechanism and the business logic that comes with it. With no previous experience on the field you are definitely bound for some surprises, and this is one of the pain points Netscan was built for.

So that’s 15% that’s directly cut-off from your business revenue, are you prepared for that hit?

If you use one of the established PaaS or SaaS WebRTC services like xirsys and tokbox you can expect those failure margins to drop down to 5-10%, this can only happen out of raw experience and a well maintained, robust and always available infrastructure.

However, no matter the experience and robustness of your technical infrastructure there is one core business related thing that you need to consider:

Your Clients’ Locations

Every business has a different client base and depending on the nature of your industry, WebRTC may not always be the most effective solution.

The first question to ask is where your clients are likely to be situated? Will they primarily be office-based or will they be mobile? For example, if your clients are likely to be using video communication from hotels or other public places, WebRTC connectivity could be an issue. Some public Wi-Fi connections have very strict firewall rules, which often make it difficult to establish a WebRTC connection.

So, cost-wise, depending on how much business you might lose out on due to client location, you might want to consider some alternative fallbacks.

Fallback Solutions

If connectivity problems are beyond your acceptable margins, then you should look into a more enterprise solution on the client’s part, like having them install a browser plugin or a stand-alone application for windows and mac os x. Ideally, you should look for a service that guarantees it can penetrate firewalls and reliably establish a connection.

Of course, the effort of maintaining two real-time communication solutions can be impractical, however, the principal factor that you should consider here, is user experience.

WebRTC is seamless, requires no installations and ‘works’ from the get-go. If you find that WebRTC works for your company in most situations, then you shouldn’t abandon it with all your clients, especially if the alternative is to install custom software that requires additional ongoing maintenance and support.

Do you think WebRTC is a viable solution for one-off video sessions? What other factors do you think businesses need to consider? We’d be keen to hear your thoughts.