As we embark on this WebRTC journey, we realise the potential it has to transform telecommunications especially when it comes to combining web technologies. From Internet of Things to the Medical sector, we’ve seen a massive amount of applications that integrate WebRTC with other web technologies to provide highly-tailored user experience.
Recently, HootSuite announced they will implement voice integration to customer service. Today, I’ll show you a DIY example of how you can achieve this by combining some of the hottest applications and frameworks out there to create an awesome CRM campaign over social media, using WebRTC!
I call it the ‘Customer Support Tweetbot’ and here’s how it goes:
— Voxbone (@voxbone) October 17, 2014
- A big bowl of Node.js
- One ripe MongoDB
- A whole bird from the Twitter farm
- 500gr of Bitly links
- A pinch of automation from a homemade Zapier magic recipe
- Lots of WebRTC! Don’t hesitate to fill the rest of your bowl with it.
How I built the app:
Last time I taught you how to create a Click2Call service you can implement in your email footer using WebRTC. We’re going to use exactly that great app.
The original app was built in Java and the app was parsing the URL (which contained the phone number as a variable like this: http://www.exampleapp.com/?number=3228082181) to retrieve the phone number of callee to call it from the browser. However, it didn’t allow for great customisation so I redesigned the way to build it.
Here’s how: I used our new NodeJS library for WebRTC authentication and created a Node app with MongoDB. What the app does is, when the email recipient clicks on the email footer and is redirected to the app page, the app parses the URL to retrieve the ‘pathname’ as a value for the username. Like this: www.exampleapp.com/snacar – snacar is my user name and the app will retrieve my profile stored in MongoDB. I added some profiles like ’support’ and ‘reception’
If you know a bit of Node, it’s very easy to set up. The sample code can be found here and the app (support profile) resides here for you to take a look at. Now that we have the main app, we will turn it into a super CRM Support bot:
- I created a profile page that calls our support team: here
- Used the link to the page and plugged it in to Bit.ly for a shortened version: http://bit.ly/voxhelp
- Used Twitter advanced search to figure out how which tweets we want to reply to.
- Used a Zapier magic recipe that sends an automatic tweet: ‘Need Help? We want to fix this! Call our support team: http://bit.ly/voxhelp every time someone sends a tweet to the @voxbone handle with #NeedHelp hashtag.
Now we’ve got an awesome tweetbot that lets our customers that are in trouble call us right from their browsers! So far it’s voice only, but as soon as we implement video, we can create a highly-tailored one-to-one help desk platform like Airpair!
— Voxbone (@voxbone) October 17, 2014
The longest part is obviously building the app itself, the rest is so simple to implement and creates an awesome service!
For more info on how to build the WebRTC app that calls the support team , check out our WebRTC SDKs , the documentation, and the sample code!
WebRTC is a very strong tool for communication over the web. Being based on web-standards and open source, you can combine it with any other web API like WebGL, or a vendor’s API. Let’s say you want to click-to-call your LinkedIn contacts, just add LinkedIn’s API to the mix. I’m really excited to be working with this technology and can’t wait to see what you build with it!
If you’ve enjoyed this post, or you want to ask me a question, you can reach me at firstname.lastname@example.org or let us know what you are building in the comments below.
Latest posts by Sacha Nacar (see all)
- Building the Phone of the Future with Voxbone & Cisco Spark - April 17, 2016
- Automating your DID Provisioning with Slack & VoxAPI – Part 2 - January 27, 2016
- Automating your DID Provisioning with Slack & VoxAPI – Part 1 - January 9, 2016