Social data platform provider DataSift is teaming up with URL redirection/link tracking service bit.ly to provide deeper analysis of how social conversations and clickstreams affect their brand’s online reputation and performance.
The ‘Real View’ of the Twitterverse
“We’re giving a real view of what’s going on,” DataSift Founder/CEO Nick Halstead stated. “Companies look at volumes of tweets to measure the performance of their brand.”
DataSift serves as a resyndicator of the Twitter firehose, allowing companies to use its API to filter through up to 1 billion daily data items in the firehose to obtain specific tweets based on metadata such as keywords, location, sentiment and influence. DataSift also offers analytics tools to extract actionable information from firehose trends.
However, by partnering with bit.ly, which Halstead calls the “de facto link shortener” for Twitter, DataSift intends to allow companies to do far more than just track who is saying what about them. “Companies will be able to track how a tweet migrates, how often it is pasted into Facebook and how many likes it gets,” said Halstead.
By combining social conversation analysis with clickstream analysis, DataSift seeks to enable activities such as measuring and increasing customer engagement, measuring the specific impact of marketing campaigns and messages on key social influencers (which can include individuals as well as media outlets), and identifying currently trending topics and keywords for real-time content curation.
DataSift also just received a US$ 15 million round of financing from investors including Venture Partners, Northgate Capital and Daher Capital.
Enhancing Marketing with Twitter Analysis
A couple of years ago, Social Media Examiner published three succinct tips for enhancing marketing through Twitter analysis that can provide a lot of insight. These steps are not replacements for more in-depth analysis that can be performed with solutions like DataSift, but are still good practices as part of a more comprehensive strategy.
First, Social Media Examiner advises pasting bit.ly links in a browser and placing a “+” on the end of them, which brings up a statistics page including conversations, clicks and regional analysis. Second, marketers are advised to perform “split testing” of tweet headlines, or running old tweets with new headlines and comparing results. Finally, marketers should use a Twitter scheduling service to send out tweets at different times of the day to see when they are most effective in different geographical areas.
Getting a Grip on the Firehose
As explained by CMSWire a few months back, “firehose” is essentially a nifty name for an expensive tool that allows users to pull in all of the available data from a social network such as Twitter. Massive amounts of information are purchased, scraped and discovered. They are funneled through complicated systems which include futuristic-sounding stuff like "Natural Language Processing" (NLP) which can read live posts, then interpret and analyze them.
However, while companies like DataSift and bit.ly can greatly aid marketers in their quest to actually derive value from the endless streams of social media data constantly flooding in, CMSWire contributor Matthew O’Connell warned readers to remember that “people are funny, robots aren’t” and “no program can accurately gauge human sentiment, particularly when it comes to concepts like irony, sarcasm and just plain false information.”
O’Connell did conclude that despite limitations, firehose tools like DataSift are essential in this age of social media as a dominant consumer communication and engagement channel. Users just need to remember that measuring human interactions still requires a human element.
- Adobe: IBM's Silverpop Deal Could Trigger 'Nightmare'
- Will BlackBerry Once Again be King of Mobility?
- The SharePoint Information Governance Problem
- Why Microsoft's Reign Will Continue
- It's Official: Forrester Says Campaign Marketing Is Dead
- Is Marketo the Next Acquisition Target?
- 3 Ways Social Media is Changing Online Content