Outsource WebRTC? Rent, don’t own

So you’re building a WebRTC application, in or outsource it, you ask? An what pieces would you outsource? What risks are you taking and what are the advantages?

We all know the great success of Amazon Web Services, over $3 billion in revenue for 2013, but starting AWS from scratch would have been really tough (even if you had the cash). How do you fund a scalable infrastructure for a customer that doesn’t yet exist. Lucky for AWS they already had a huge anchor tenant in Amazon. A unique situation.

WebRTC is moving fast and already there are a slew of service providers offering “by the drink” services specific for WebRTC. A non-exhaustive list includes: Addlive, Crocodile, Plivo, Pubnub, Rebtel, Tokbox, Tropo, Twilio, Vline, Weemo and Xirsys.  My friends at webrtchacks maintain a nice list that worth tumbling thru. I admit I sometimes struggle to figure out what each of these service providers offer which is more a testament to fluffy marketing than anything else.

If I could (and dared) to give you a single piece of advice having lived thru the debacle of VoIP. Save as much money as you can for marketing and adopt the Mars bar attitude, if the user doesn’t eat it, don’t worry about it. I lived thru many VoIP providers who spent buckets of money on back office technology and custom development, only to arrive on the market to find it tougher than they imagined and with only shekels left in their pockets to fund the important customer acquisition phase. Save the intellectual debates for the IETF.

The onus on these service providers is significant they have to take on the tough stuff, otherwise why would you bother using them. Need to transcode 500 video streams right now, they have to be ready for that as well. A phone number in Des Moines or an SMS message to Nigeria need to be handled with equal ease. Each month you should expect an email announced a reduction in pricing and new functionality you’d been dying for. Don’t under estimated the grueling task that a back office provider has, the more for less axiom always grinding away on them.

All of this power for your measly monthly spend of $12.85 (stop complaining about their rates, it’s either insignificant or you don’t really have a business).

For such an early technology, WebRTC is already offering a wide assortment of third party engine providers. Their trying to create their platforms without much volume to base it upon, thus they’re building on hope. We’re off to a great start in 2014 with the providers clearly ramping to add more functionality to ease your way into the market.