Saturday, April 14, 2012

Data Mining andPredictive Analytics: Dilbert, Database marketing and spam

Dilbert, Database marketing and spam

Ruben's comment that referred to spam reminded me of an old Dilbert comic which conveys the misconception about database marketing (e-marketing) and...

Source: http://goo.gl/mag/YpqL7

Shared via Google Currents



Sent from Samsung tablet

Friday, December 04, 2009

Twitterlytics for Customer and Market Intelligence

Twitter generates a very large amount of user generated content. This content could include, apart from a lot of garbage (like 'I started Tweeting ' and 'My doggy is scratching his back'), valuable opionions on products, comparison of closely competing products, likes and dislikes of individuals, wishes and feature lists, demographic hints, the geographic location details, timelines, complaints and grievances, recall requests, hate and libel, stock out complaints, product improvement suggestions, subtle hints of market saturation, competitor moves, technology and fashion trends in the space, etc. This list is by no means exhaustive but tries to give a flavor of the kind of consumer feedback and market intelligence one can get through the social media and especially Twitter.

I like Twitter (and then other social media like Facebook) more than the lengthier blogs because of the precise, succinct and targeted feedback that the former can generate. While a blog would generate more detailed analyses and intellectual reasons behind the author's story, the character limit in micro-blogging forces a user to be more precise and focussed. Not that current technology cannot parse the longer blogs, the shorter version makes it all the more convenient for more people to participate. And hence many more tweets are generated daily than the number of blogs posted.

Besides pulling information from Twitter and making the right sense of it, the use of Twitter for pushing to prospective or current Customers/followers is too well known and need not be discussed.

There are already numerous services like Twittercounter and others that generate aggregates and statistics regarding the tweets, topics, views, etc. Most of these services do not get into the semantics and mine the data for the vast treasure that it hides.

There is also another important factor making the Twitterlytics important. The stream of content generated provides an almost real time input for the Analytics. With multiple options to post your views like the mail, desktop and mobile applications, text messages and the twitter web interface the time lag between the actual event and a post on it is reducing. Though this is not a highly essential requirement from the analytics perspective for corporates to gather Customer feedback, this does provide opportunities for improving the Customer experience by having your software focus on individual Customer complaints. The company in question can then take actions specifics to resolving these individual complaints. Also, the near real-time information flow help detect aberrations and prevent problems before they have blown into a major Customer Service fiasco. Twitter stream mining could still prove to be a valuable asset to the company without having to deploy personnel to monitor these social media.

Besides this, possibility of selective profiling of followers with their consent could open up many more interesting possibilities for the corporates making it a really interesting Customer and Market intelligence tool. Opportunities are limited only by imagination.

Cellphones and Cancer

Though I love using cellphones and am a fan of couple of brands and models, I was never too sure of the effects the radiation caused on the human body and health. This article in Times of India was a relief to see. Please visit article from Times of India (TOI) A study tracking the impact for 30 years should give reasonably reliable results.

However, this seems more of a population study and not a controlled clinical trial (It would have been very difficult and expensive to do such a study for a long time though), any further confirmations by future studies are welcome.

Test post via email

This is a test post. Please ignore.
Sent from BlackBerry® on Airtel

Market Intelligence: A Business Imperative or an IT Function?

I would argue how much of this Business Imperative can be met through It Support.