DB Digital, the digital arm of Dainik Bhaskar Group, has become the second largest Digital Group in the country in less than 10 years of its inception. With more than 101 million unique users and an expanded portfolio from News to Business, Entertainment, Fashion & Lifestyle, Sports, Food and Gadgets, it is utilising analytics in a big way. The flagship website DainikBhaskar.com is the world’s largest Hindi news site and the second largest news site in India, while the other properties by DB Digital like Divya Bhaskar, Money Bhaskar, Bollywood Bhaskar and Divya Marathi are leaders in their respective genres.
Heading the DB Digital arm of Dainik Bhaskar Group is Gyan Gupta, who is the Chief Executive Officer and comes with 18 years of exhaustive experience in media, digital content and technology; handling multiple profiles including business development, marketing & pre-sales, delivery & operations.
Post their win at the recently held Asia Pacific Awards- BIGGIES, where the most innovative media companies were honoured for their various initiatives on data and artificial intelligence, Analytics India Magazine (AIM) interacted with Gyan Gupta, who has been instrumental in conceptualising multiple unique features and properties and introducing many path-breaking initiatives for the brand.
DainikBhaskar.com won three awards– first place in excellence in data driven business process efficiency for their analytics platform- Wisdomnxt, and, first & second place for excellence in data driven product development and artificial intelligence for recommendation. In a conversation over phone, Gupta shared his insight on their analytics platform, adoption of artificial intelligence and analytics to stay ahead of the competition, and much more.
AIM: DB Group was recently honoured for being the most innovative media company for implementing data and AI. When did you first adopt AI and how have you been implementing it in your daily functions?
Gyan Gupta: For us the data journey has been a long journey, in terms of adopting data and then applying a layer on it. We started venturing out in that direction about two years ago. At that point of time, which was largely big data, we can say that we started with big data and ways to leverage that. The business problem that we were trying to solve then was recommendation– how can a person coming to the site can consume more, how can the attention span be increased, and more.
This was the start of our journey as we started creating recommendation engine. As we ventured into that, we realised that it was a lot more complex than what we actually thought it was, and we started building on top of NLP and related technologies for creating recommendation engine. As these helped in creating recommendation engine, we are now finally trying to get into personalisation. We want to see how we can address personalisation for a user who is visiting our website and feed them with content that they want to see. We are trying to make it simpler and easy for a user to pick out those 10-15 articles that they might want to read rather than letting them explore huge sea of articles. To equip the user with what he needs, is our primary focus.
AIM: Would you like to talk about the AI based App? How is it being used?
Gyan Gupta: The company is heavily invested in developing recommendations engine, which form the bedrock for predicting the interest of an individual based on interest and preferences of similar users. Our app works on this collaborative filtering, enabling content delivery by capturing user click stream data and identifying the reading patterns of the user to process and eventually serve the user what’s relevant and specific to him or her. Keyword-tagging, Story-stitch, and Association-Mining are just some of the many elements that the team is working to get the App to deliver likeable and relevant content and videos individually.
AIM: What have been the benefits of adopting artificial intelligence (so far)?
Gyan Gupta: I think, largely we have been able to improve the depth of our website. As we can now have a clear metrics in terms of the time spent on the site, we have been trying to use it to get important insights. One of the other things we are using AI is to serve personalised ads. Our performance has grown up by 2-3 times, which is quite significant. We are also working on improving notification so that there is a higher click through rate. The aim to improve the traffic and number of people coming back to the app and we have been using AI and data to improve extensively on that. Our in-house platform, Wisdomnxt, helps editorial to see and monitor the traffic and focus on stories that they want to do.
AIM: Would you like to talk about the analytics platform Wisdomnxt and other tools used in analytics, highlighting few use cases of how it has benefitted DB Digital?
Gyan Gupta: Wisdomnxt is our proprietary/ in-house platform, a tool rather, that helps editorial team to decide on the stories to publish. It lets decide what’s working and what’s not working for the website in terms of content. It is built essentially on the top of big data, and there are a couple of thing that it does, a) it tells the team about what’s working and what’s not, b) it gives a kind of snapshot or storyline that they can work on, c) the way it is operating across multiple platforms, on our own site or any other distribution platform, d) it also helps to predict the reach of the story. There’s a prediction engine that we have built on top of it. It also allows us to see what others are doing on a similar content.
So these are basically the use cases– increasing the depth, increasing the retention of users, increasing the consumption of the content and more. The largest use case has been on the advertising side, where it helps in personalised ads, improving the CTR and client responses. These are the typical stuffs that we are working on right now.
AIM: Targeted ads and content for media houses is always a very competitive thing. How do you stay ahead of competition?
Gyan Gupta: There is nothing much about staying ahead of the competition, as the idea we believe is about creating value for the user. If we create a value for even one user, that should be good enough.
AIM: What is the size of the analytics and AI team? What are the kind of skill sets you look for while hiring members in your team.
Gyan Gupta: While hiring we look for 2-3 major skill sets. We look for data scientists largely who can analyse a problem, study the data and create trends out of it. On the technical side, we look for people who can work on Hadoop, Spark, Python, R.
AIM: What are the challenges you face being in the disruptive technological space?
Gyan Gupta: The most significant challenge is that this is fairly a new space and there are really not a lot of use cases that actually exist out there. You have to find a lot of things yourself, inspect the content, especially it is difficult with the content space around Hindi. We are trying to overcome these challenges and define the next level.
Try deep learning using MATLAB