Buzzword Overload: Getting Social with Big Data
When someone suggests integrating Twitter with your website, what are they offering? Maybe they’ll pull in a list of your recent tweets and showing them on the homepage or let your users share articles on your site.
Not bad. But what if we could do more?
What if instead of using social platforms as either a decorative component on your site or a method of simply sharing content, we could pull content in from the platform itself and create something unique that would draw visitors to your site?
Take Twitter for example: on average they’ll see 500 million tweets posted on average, every day. That’s millions of people tweeting about everything from current events, which TV shows they like, or don’t like, to simply ‘random thoughts’. That’s impressive and all, but how can all that random information draw people to your site? We need two ingredients: an idea, and an API.
Enter the API
We developers love API’s (in case you’re interested, API stands for “Application Programming Interface”). It not only allows us to get at ‘raw’ data stored behind a service, but generally allows us to filter down the (generally masses of) data to give us a slice we might be interested in. Now is probably a good time for an example…
Say you’re the Exec of a TV station (and also a developer - multi-talented). You’ve just launched pilot episodes for some new shows in the UK and you need to make a decision about which show to keep on. Tricky.
You know people tweet about TV shows all the time, so you decide to use the Twitter API to collect tweets relating to your new shows. The Twitter API allows you to set a geographical filter on your search (the UK), and allows you to restrict the results returned to only the terms you’re interested in (in this case, #show1, #show2, #show3 etc).
Great, now you have your tweet data… what now?! Well, maybe use the data to plot a few graphs to see when people are talking about your shows (“hmm, #show1 saw a lot of chat when it was on, but the #show2 discussion just keeps on going!”). You might even decide take this feed and turn it into a feature on your network website - the show that gets the most Tweets wins!
Big Data, Big Ideas
Pulling in a load of raw Twitter, Facebook or Instagram data on its own isn’t going to generate a lot of interest. We need to shape and present this data in a form that’s new and interesting. This is where the ‘idea’ part comes in.
Depending on the end goal, the idea may centre around adding functionality to an existing section on a site, or may even be the main feature the site is built around.
One idea we’ve been working on is the concept of using ‘tagged content’ on a site (generally as part of a campaign). For example, we might set an app up to periodically query Instagram looking for photos tagged with a specific product’s hashtag, pulling in any matching images into our system for display on a relevant page (not forgetting that all-important moderation step!). This means users don’t have to do anything more to get their photos on your site other than tag their photos using Instagram.
So what about making the idea the focal part of the site? Handily, here’s an example we recently launched for AXA: Twitchoo! (http://eqtr.axappphealthcare.co.uk/twitchoo). By collecting data from Twitter over the course of a few weeks, we were able to determine flu ‘trends’ by looking for some key words people were tweeting about:
“Wife has given me her cold only now it's Flu!”
Now this tool isn’t recommended for any official pandemic planning activities, but what it does do is get people talking and sharing about the client and health in general - which is ideal!
Seeing the Wood for the Trees
There’s a lot of information out there. Maybe too much. We’re only just scratching the surface of what’s possible with social data; it doesn’t stop at share buttons or showing a list of your recent Tweets.
Once you start looking beyond what’s available on the surface, there’s a wealth of freely available information that social networks have made available, just waiting to be tapped into.
Useless noise, or a goldmine?
Written by Paul Aikman, Web Application Developer