How India's BJP used data analytics to swing voters


The Bharatiya Janata Party (BJP) won a landslide victory earlier this year on the back of a carefully scripted communication and public relations strategy. While it is no secret that it used social media effectively to communicate its message very little is known what went behind the scenes. Arvind Gupta the national technology head of India's ruling party finally reveals how his team used analytics to influence voters in the 2014 General Elections

Arvind Gupta, BJP's national technology head speaks to social media experts in Mumbai earlier this month

"BJP’s election campaign was a very good mix of all technologies. It wasn’t just social media but all digital platforms, including mobile and mobile Internet. Combined with advanced use of analytics to decide how, where and when to communicate, the campaign targeted 810 million voters in 543 constituencies across 11.36 lakh (1.13 million) polling booths."  That is how Arvind Gupta, BJP’s national technology head, summed up the party’s digital communication strategy at a recent international gathering of social media analytics experts in Mumbai.

The seminar - ‘social intelligence for business success: a case of three cultures’ - was organised by Adfactors PR. Gupta told delegates how during the one-and-a-half years between December 2012 (when the BJP won the state elections in Gujarat for the fourth time in a row with Narendra Modi as Chief Minister) and February 2014, the party did not engage in any traditional publicity campaign.

Using data analytics

"During this period, our primary campaign was on digital and social media. We were involved in setting the narrative, changing the tone, and reacting to social media posts," he said.

"Technology was our primary toolkit over these 18 months, when we engaged, converted and kept engaging with our target group. Apart from our own set of volunteers, we specifically focused on converting the undecided voter or what we call ‘the floating voter’ who had not made up their mind on which candidate to vote for."

The digital campaign, Gupta said, required a lot of planning. This included 3-4 years of meticulous data collection, collating information in a scientific manner, and making timely actionable analysis.

"We had data on each of the 543 constituencies. We knew how many mobile and Internet users were present in each constituency. The same holds true for social media users. Alongside, we used analytics to understand which polling booths had voted for the BJP in the previous elections," he said.

For each polling booth data analytics was used to segregate voters into blocks to determine who were pro, undecided or against the BJP. 

"We first used deep analytics to understand group communication behavior and then used appropriate technology to communicate with them," he said.

A BJP supporter posing against a poster depicting Narendra Modi during the election campaign

So while in some cases the BJP used the Internet in others it decided to use the mobile. But some voters had to be reached through good old fashioned direct human contact. The BJP used a combination of analytics and technology to organise all volunteers online. For the communities positively inclined towards voting for the BJP traditional mass communication media was used.

"However, to reach out to the other two segments – undecided and negative – we used micro-communication using Internet, mobile, social media and even BJP volunteers on the street.  All these tools were deployed during the last one and a half years of the campaign."

Social media chatter

The party’s Information Technology cell monitored the social media round the clock for all conversation in which the BJP was being discussed and volunteers used personalised messaging to outline the party’s policies, manifesto, and the overall vision.

"People may not have necessarily used a particular hashtag or tagged someone particular on social media, but as long as they were discussing specific issues, we could reach out to those conversations, understand what they were really saying and respond to them in real time," he told delegates.

Gupta illustrated the effective use of data analytics when a statement made by the party leader (and now Prime Minister) Narendra Modi during the course of the campaign was threatening to snowball into a controversy. On October 2, 2013, almost seven months before the May 2014 elections, Modi expressed his policy to prioritise "toilets before temples".

Seizing opportunity

"We ran a very quick analytic exercise on what the chatter was about within India on social media. We figured out that more than 45 percent people agreed with this. And, these were mostly floating voters. They appreciated that a political party was actually talking about a vision for India. So, we found a new constituency agreeing with us. However, a section among the traditional BJP supporters had probably misinterpreted his statement as having to make a choice between temples and toilets. Mr Modi had, of course, never meant it that way and was making a point about hygiene, cleanliness and sanitation," said Gupta.

"The (communications) team immediately went to work, set the agenda that sanitation was a serious subject, and pushed to convert this entire argument of ‘toilets over temples’ to ‘sanitation for all’. We created much chatter around the subject, and there was so much debate around it that public sentiment turned decidedly positive," Gupta told delegates. "Analysing the online sentiment using social media analytics, we found that the initial support-base of 45 percent went up to 68 percent by the end of the next day."

The ‘sanitation for all’ policy agenda is now centre stage for the government of Prime Minister Modi. He has set an ambitious target for every school in the country to have separate functioning toilets for boys and girls within one year.

The seminar brought together communications specialists from Europe, China and India to discuss how businesses can leverage analytics and social media more effectively towards increased competitiveness. These included Thomas Stoeckle, Group Director, Analytics, W2O Group, Sam Flemming, CEO Kantar Media, CIC and Santosh Desai, CEO of Future Brands.

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins

Already registered?
Sign in

Recommended for you

Recommended for you

Explore further