It took a long time in the making, but Broadcast Audience Research Council (BARC) finally started releasing data from April. It now rivals Nielsen-Kantar's TAM Media Research. BARC chief executive officer (CEO) Partho Dasgupta spoke at length with Vanita Kohli-Khandekar on the conflicts, technology, financials, and other challenges of setting up the world's largest television audience measurement service. Edited excerpts :
What challenges you faced while creating BARC?
The challenge has been carrying the various stakeholders, with their varying interests, together. (BARC is a venture of the Indian Broadcasting Foundation, Indian Society of Advertisers, and Advertising Agencies Association of India). When I was being interviewed for the job (in 2013), I asked Punit (Goenka, Zee CEO and BARC chairman), whether BARC was an association of people or companies? He replied if it were an association, they would have hired a secretary. As CEO, I had to figure out everything, starting from funding.
The biggest achievement in these two years has been getting the funding going without any member shelling out cash. We are a Section 8, not-for-profit firm. We can make and use profits but cannot give dividends to shareholders. So, venture capital and private equity were out (Nielsen is largely owned by institutions). So, we sounded out two to three banks and those were happy to be on board. We have managed to raise Rs 180 crore from banks and got shareholders to stand as guarantors. This ensures everyone has an interest in getting the system going and they stand to lose if it fails. There is an additional Rs 100 crore that broadcasters have spent on watermarking technology for being on the BARC system.
How is BARC different?
We are a technology-enabled research firm. About 76 per cent of our investment has gone into technology. Globally, joint industry bodies give the mandate to two to four vendors, which usually include Nielsen and Kantar (TAM's parents).
India needed a scalable yet value-for-money rating system, given that our ad-to- gross-domestic-product ratio is very low. These are conflicting imperatives. Then there is the integrity of the system. It is key. This is what led us to 12 processes, 30-plus vendors, all best of breed. There are 51 software applications running across seven locations and more than 1.1 billion viewer records generated every year.
All of this comes together with technology. No one part of the process knows what the other is doing. The whole thing is being done for the first time - so, there is no context. We did the meter in Rs 25,000 each. (TAM meters cost a reported Rs 1.5 lakh each). We want to get it down to Rs 10,000. We have meters in Kashmir, in the northeast. But there are challenges - two of the meters in Bihar were stolen, there have been attempts in Mumbai to bribe people.
What are the checks and balances to protect data integrity?
We have multiple ways of identifying outliers in the data. If you don't watch Hindi GEC, general entertainment channels, (going by your viewing habits, peer, or control group data) and you are seeing a Hindi GEC for four hours at a stretch suddenly, we investigate. It maybe that there was IIFA (International Indian Film Academy) awards on Colors on a Sunday and you watched that, so that is logical. But if there are other reasons not logical, we quarantine that home. Then we investigate, and if there is evidence, we can act against a channel for tampering with data, by not including it in the ratings data for six months. (This hits the channel's ability to get advertising).
The shareholders of BARC are bodies with different interests. How does one handle that?
There was a lot of friction in the past two years. On watermarking, there was a strong opposition from some large players. At every stage, people have been for or against something. It has meant lots of debate in the boardroom and the techcom meetings. But there were no votes. All decisions were unanimous and everybody stuck to the final decision.
There is a perception in the industry that big broadcasters control BARC. How true is that?
The Indian Broadcasting Foundation (IBF) has a 60 per cent stake in BARC. The Indian Society of Advertisers, and Advertising Agencies Association of India have 20 per cent each. If broadcasters are dominating, they would dominate IBF, because it is between them that they need to work. Advertisers are very vocal members on the (BARC) board.
Have you deployed any fair-use guidelines to stop cherry-picking of data?
It doesn't work till there are punitive powers. We are debating it, though it is a subject for the IBF. We have built some stuff into the software. If you want to look at Andheri (a suburb in Mumbai), the software won't allow it, because the sample is less than 200. In some genres, we can make the software work differently, so that it will give data only over four weeks. In many countries in the European Union, thematic (niche) genre data are aggregated for four weeks. It is more a question of maturity of the market.
Please outline the BARC sample?
We should be reporting data for 22,000 homes in two months (12,000 now). This includes 100,000-plus and less than 100,000 towns, and rural homes. By March 2016, we are hoping to be in 30,000 homes. (According to the BARC website, this figure should go to 50,000 homes in four years).