It's practically become cliché to put social media analytics and big data in the same sentence. Deciphering the social media firehose was one of the first use cases when big data analytics made the jump from internet companies like Google, Facebook, or Yahoo to the enterprise. Technically, it was a short leap from going to clickstream to the variably structured data that populates social media streams.
Founded in 2010, DataSift was one of the first commercial software providers to offer a platform for aggregating live data from social media. Its original claim to fame was as one of Twitter's two live data partners, with its own license for reselling and repackaging the entire Twitter data feed. Given that Twitter has been known to be a fickle partner at best, it's not surprising that DataSift's semi-exclusive arrangement didn't last. After Twitter bought Gnip, DataSift refocused toward becoming an all-purpose aggregator. You can still get the Twitter firehose live through DataSift, but now you have to buy the license directly from Twitter.
Today, of course, social media monitoring is one of the most commoditized areas of consumer marketing analytics. There are numerous free and paid-for marketing apps that gauge mention of your brand. Providers like IBM integrate social media monitoring with predictive analytics tools. There are a number of dashboards and dedicated tools (primarily available in the cloud) for tapping into live social media feeds, pushing content onto them -- not to mention an almost endless array of APIs for developers to write their own apps to ingest social media data.
DataSift offers those integrations to the usual suspects of social networks and other channels such as blog sites, YouTube, Wikipedia, Instagram, and so on. But those integrations are part of a broader entity extraction and indexing engine that can be used for discovering the desired marketing insights.
It also peeks into what it terms "non-public" data sets, such as Facebook, which have stringent privacy restrictions for ensuring that data extracted is anonymized. Facebook of course provides its own analytics; the benefit of a third-party platform like DataSift is that incidence of certain terms or topics can be correlated across other social network channels.
But the latest addition of DataSift's non-public data offerings is LinkedIn, a social media platform that provides a more B2B professional channel compared to most social networks. That is very much in line with the goals of its last major round of funding in 2013 to expand its realm to business data.
Today, you can use LinkedIn's own analytics to drill down primarily on followers and page views, although it has recently added some content marketing metrics. DataSift goes beyond that to provide more freeform analysis of anonymized data across the full corpus of LinkedIn data within monthly time buckets, based on a multi-faceted view of 130 dimensions. It does so with an analytics engine that uniquely sits inside LinkedIn's firewall.
Clearly, DataSift's differentiator is combining natural language processing with a highly granular slicing and dicing of LinkedIn content. At this point, you can correlate it with analytics from other channels such as Facebook using the same DataSift query language. But that just prompts the question: Why not add an integration tier so you could selectively combine the results from LinkedIn with Facebook or other social channels to which DataSift already links?