Why I went with Plivo, not Twilio
Plivo has been my favorite phone API platform for awhile. I built my first Plivo project about a year ago because at the time they were the only option to directly integrate with SIP which was great because it allowed us to integrate their service with our existing carrier infrastructure on a very low protocol level. Twilio can sort of do this today, but their integration still feels a little hacked together.
From Plivo’s site:
Plivo’s story goes back to 2011, when the founders, Mike and Venky, accidentally exchanged messages on Github. At the time, they were both independently seeking a robust communications platform to develop and integrate Voice and SMS capabilities into their web apps.
They couldn’t find any platforms that allowed the flexibility to write applications using HTTP APIs without having to pay high carrier markup fees, so they decided to roll out their own solution: Plivo Open Source, which soon turned into a full-featured cloud platform called Plivo Cloud.
This to me is where it gets interesting for most people. The open source Plivo project was great, but still required a somewhat working knowledge of Freeswitch and could take awhile to get working just right, even for a seasoned engineer. Plivo Cloud changed all that.
Plivo cloud is essentially a hosted version that lets you integrate their API functionality really quickly but still allows deep SIP integration with an existing infrastructure if you need, just create a SIP endpoint and you’re off and running.
Pricing was also a factor. The goal I had while putting together Building Phone Applications was to show how to build these amazing business applications using voice and SMS as affordably as possible. Because Plivo’s pricing is lower and offers native SIP support there is much more value in our applications by using Plivo.
Second I really couldn’t find much documentation beyond the very basic specs and examples. Twilio, being the leader by market share simply has more projects out there, despite the fact that Plivo is more powerful. Another benefit when looking at the Twilio comparison is that TwiML is pretty similar to Plivo XML but just different enough to be frustrating if you’re trying to port a project or example over to use Plivo so I saw a great opportunity to add to the body of knowledge out there.
Finally I wanted to focus the examples in Building Phone Applications around a 100% improvement from the status quo business phone platform. In order to do that I needed to be able to leverage device registration, which doesn’t work without a lot of hacks and/or middleware with Twilio. Plivo’s API let’s me create a SIP endpoint, generate the registration and secret and register a phone or app directly to their SBC. Game Over.
I continue to watch Twilio and if they add the functionality I may release a second edition, but until then Plivo’s still my top choice.