By Surajit Agarwal
Messaging apps are the new battlefield and over the last couple of months Google, Facebook and Microsoft have shown serious intent at grabbing a share of the pie. Facebook has a head start here – the social networking giant has over 1.9 billion users on its two messaging platforms—WhatsApp and Facebook Messenger and will fiercely fight to protect its territory.
In contrast Google only recently announced Allo. The next generation Messenger App from Google is expected to launch soon. Microsoft displayed its intent by purchasing little known Wand Labs, a company that says its mission is to “to tear down app walls, integrate your services in chat, and make them work together so you can do more with less taps.” Facebook, on its part, announced Deep Text which it describes as a “deep learning-based text understanding engine that can understand with near-human accuracy the textual content of several thousand posts per second, spanning more than 20 languages.”
Essentially the big three are building real time contextual understanding of human conversations so they can serve up relevant information, or if you like, ads while you are messaging. For example if you’re chatting with a friend and say you’re planning to leave by taxi, a link to Uber, Ola or any other cab service will pop up in the messenger app. But if you message your friend saying you’ve arrived in a taxi, the link will not appear.
Google Allo uses the example of two friends chatting about lunch. They decide on an Italian restaurant and the Allo messenger picks out one near their location. The ability to serve up relevant information in real time is the pot of gold at the end of the messenger rainbow these companies are chasing.
Unmish Parthasarathi, Global Head of Digital Sales at the International Cricket Council, points to China’s popular messaging service WeChat as the “go to model for next generation messaging platforms.” The China-based chat service has a large, integrated partner network, with the result that users can do pretty much everything from within the messenger interface…from shopping, to booking a table at a restaurant to buying movie tickets or hailing a cab. They even have the ability to book a doctor’s appointment from within WeChat.
Unmish says, “The evolution of chat from social networks can be compared to what happened in the TV broadcast world. First we had free to air TV, which is how I would describe a social network like Facebook. Now people want smaller, well defined and personal niches. So something like Netflix which provides a better viewing experience more choice and more personalization. As the niches get better defined the quality of the audience improves significantly. That means better ROI for marketers.”
It’s easy to understand why the big three are so keen to build a stake in the next generation of messenger. According to Business Insider, more people use messenger apps than they do social networking platforms. The shift away from social networks is real and therefore the need to quickly monetize the next big digital platform. The messenger platforms are key to build revenue in largely mobile first markets like India. We already have over 200 million users using messenger apps via their mobile phones. This number is expected to jump to over 700 million by 2020…just four short years.
A significant part of this number is expected to be mobile first internet users, thereby giving advertisers the opportunity to reach out digitally to an audience that has been out of reach so far. The challenge in a chat app is to be able to provide relevant and contextual information in real time. That is what Allo claims to do and that is what Facebook says Deep Text will allow it to do. Not everyone, however, is convinced the experience is going to be as seamless as the companies will have us believe.
Sujit Janardanan, Vice President Global Marketing and Corporate Communications at Aranca, says “the ability to monitor intelligent conversations and serve up relevant information sounds interesting, but I’ll have to see it to believe it. We have to see how well machine learning works. From my own experience with chat bots I know that as long as it is monotonous and repetitive they work. But once it gets complex, human intelligence is very important.”
That level of complexity increases when you consider much of the new smart phone user base will be chatting away in a local language. Facebook does say Deeptext can mine data across twenty languages…a number that is bound to increase, but for now the best model for reaching digital newbies could be to follow the WeChat model. If it works in Mandarin and Cantonese, it should work in Hindi, Telugu, Tamil, Marathi, Gujarati!
The author blogs about digital marketing.
Publish date: June 22, 2016 2:26 pm| Modified date: June 30, 2016 3:45 pm