ThoughtBuzz launches today – Is it yet another social media monitoring tool?
I got a chance to hang out with Anshul Jain and Ashok Patro from ThoughtBuzz a couple of days back. They were one of the selected startups that was chosen to be part of unConference 2009 but due to unforseen circumstances, they couldn’t join us for the day. However, it’s great to see that they’ve been progressing really well since our last meet up in April. Daniel Goh from YoungUpstarts did a really good feature story of them back in July covering their background and their journey. I’ll be discussing more on the product and a case study on Palm Pre that they shared with me.
Real Time Analytics and Buzz
We all know the potential benefits that are being touted by real time search and buzz generating by consumers. While having comments and information being spewed real time is cool, tracking it is a whole different ball game. Understanding what caused certain peaks and troughs in user buzz can make or break a campaign. For example, the buzz tracker here provides a graph that tracks the overall volume of articles over period of time. Nothing special here in terms of the feature, but definitely something useful for understanding the way consumers are reacting to your campaigns.
ThoughtBuzz also takes into account a specified number of the latest mentions of your product/brand/service to give an instant analysis of online conversations. If a user wants more analysis, ThoughtBuzz can combine historical and real time data to provide a big picture overview.
Machine Learning, Social Media Style
Can systems really tell you if the buzz you’re generating is positive? Negative? Or not even relevant to your business? And who’s saying this anyway? Is it some loud mouthed online gaming 14-year old punk who’s got nothing better to do on a Saturday afternoon then to go hunting for trolls and shooting out back comments for products he/she has no clue about. ThoughtBuzz aims to help by using what they call sentiment analysis and topics of discussion. By carefully analyzing information in an article and mapping that information to a comprehensive database of keywords relating to positive, negative and neutral feeling, ThoughtBuzz is able to “calculate” if an article or comment has a positive/negative/neutral effect on your product or brand. They are well aware of the fact that it’s not always right, hence the machine learning part. Users will be able to fine tune this by selecting their own sentiments, which the machine takes into account and learns overtime.
Purely Technology or the Hybrid Approach
While we were discussing the features and technology, the discussion inevitably came towards systems like Brandtology that use a hybrid (software + people) approach versus a purely technology approach. Anshul was very quick to point out that there isn’t one right approach for customers and I totally agree. Sentiment analysis will be a difficult concept for clients to lap up initially, mostly due to the untested nature of the algorithm. Companies would still prefer that the platform be managed with a human touch, but with rapid increases in technology and some acceptance or errors, this expectation will slowly decrease. Furthermore, for smaller companies looking into understanding the impact of their brands in social media, it wont be financially feasible to pay for a hybrid approach. I honestly think that the real value is in providing services to clients after they have used the product and understood the impacts of social media on their brands. Customers would want to know how to improve sentiments or rectify problems rather than just know if they’re brand/product is well received or not. It’s like analytics software. Knowing how much traffic you’re getting and where it’s coming from is good, but what you really want to know is how to improve the traffic numbers and how make the leads more relevant. In my opinion, that will always be a problem with social media monitoring tools and only time can tell which company makes it.
Here is a short pitch about the product from the ThoughtBuzz guys!
tagged
