Man in the red-hot seat - Partho Dasgupta, CEO, BARC India

afaqs! caught up with Partho Dasgupta, CEO, BARC India, to understand how the new system was developed and the way in which it works.

Touted as the next big thing in the broadcast industry, the Broadcast Audience Research Council (BARC) is expected to launch a TV audience measurement currency soon. What could come as a big relief to the industry, will also pose a threat to the existing measurement system, TAM (a JV between AC Nielsen and Kantar Media Research), which might have few takers.

Launched in 2012, BARC is an industry body in which the IBF (Indian Broadcasting Foundation) has a 60 per cent stake, while the Indian Society of Advertisers (ISA) and Advertising Agencies Association of India (AAAI) hold 20 per cent each. Besides the Board that is chaired by Punit Goenka, managing director and CEO, Zee Entertainment Enterprises, BARC has a technical committee and a commercial committee, with representation from the three stakeholders.

With a paid up equity of Rs 5 crore (according to the RoC), BARC has raised a debt of Rs 100-150 crore from banks. In the new world of BARC, meters will be called 'barometers' (TAM's meters are called 'peoplemeters'). With an increase in the number of households covered, the new system is expected to give more accurate and credible viewership numbers.

The credibility comes from the watermarking technology. An indelible audio code is inserted - with the channel ID and the time it was played out – as soon as the channel is beamed out to the satellite. The meters in the households decode this info when the channel is tuned in to.

afaqs! caught up with Partho Dasgupta - in the backdrop of the decision to postpone the original roll-out date of October 1 - to understand the making of BARC. Excerpts:

Some broadcasters say they haven't received any communication from BARC. Why is that?

It is the BARC's and the shareholder bodies' responsibility to inform the industry. We keep our shareholders' updated regularly. They sit on our board and are in our technical committee. We meet them to give updates. The ones who are complaining are those with whom we are seeking meetings but they have not been able to make time for it.

We not only give presentations to our board and committees but also to the respective executive committees of ISA, IBF and AAAI. So, the right people are informed. We had a six-city road show in March to tell broadcasters about BARC and our technology. Two hundred fifty channels have already placed orders for the watermarking embedders, of which half are installed. We have also started a monthly newsletter for all our stakeholders.

The deadline for the roll-out (October 1st) has been pushed further. Comment.

Uday Shankar (Star India CEO) said in one of our meetings: "We are creating this for a generation, not for a few years - so do it right at the right cost, even if it takes a bit of time." I strongly believe that this should not be done in a hurry. The meters are already being sent to the ground. The backend is ready and we will start seeing the data. We want to test the data and validate it before using it commercially. It will take a few more months.

Globally, when there has been a changeover of systems (in our case it's a green field), it has taken three years and that too for a panel size of 5-7,000. At 20,000, BARC will be the largest TV audience measurement system in the world.

What were the concerns of the different stakeholders of BARC when BARC was formed?

The stakeholders were not happy with the existing data. That's why they combined to form BARC. Even the government played an important role in pushing the whole system to work towards its formation.

The biggest concern of the three was transparency. The issue of low sample size also was worrisome. Thirdly, the investment in technology was not up to the mark. The fact that data was not reliable and that people were not there to explain what was happening got them to react. The industry not being taken into confidence was a strong issue.

How does BARC's system address those concerns?

Firstly, the sample size we are taking into account is roughly three times that of the present system. They cover 7,000-8,000 households while we will be present in 20,000. Secondly, the current system has 9,000-10,000 meters while we are installing 25,000. Also, we will be covering urban and rural towns while TAM is just urban.

Thirdly, the whole backbone of this system is technology. We are called a research company, but BARC is a technology-enabled company - 76 per cent of the investment in the first couple of years goes into technology. So, we are putting a huge emphasis on technology.

The fourth is the way we have designed this system for integrity. For example, the data is seen by the people or technology team but they do not know where it is coming from. The design and quality control people know the kind of home it is coming from but don't know the addresses. The people who know the addresses don't know the weights of the homes. Even in the panel agency, no one knows the whole set of addresses, they only know the households at the local level.

The biggest of all is transparency. Starting from where the sample would be to what kind of meters would be drawn to all the other rules we are putting in - everything is being done by the industry.

Once the problems of the existing system became apparent, what were the choices available to you?

We could have jolly well gone with one single vendor who could have done everything, and we would have just published the ratings as the owners of that data. The route we have taken is a tougher and more complicated one. We wanted to make this system much more robust and not be dependent on one or two agencies.

The other big thing was the choice of technology. We had two or three modern choices, apart from the ancient choices, which is there in the current system. Again, we went with the one that would not just be cutting edge today, it would also be what India would need for tomorrow.

Will BARC be tapping into other screens?

Yes. Digital is the next big thing. We will soon see 4G and the market will push us to start monitoring these devices. The watermarking technology can do that too. The broadcaster would be able to monetise its beam even if the consumer is watching content on any other screen. But we are not releasing this as there is still some work that needs to be done.

By when are you looking at reaching 50,000 households? For a country as diverse as India, is even 50,000 adequate?

As per the government notification, we should be reaching the number in four years, with the addition of 10,000 panel homes every year.

Answering the second part, it's a question of economics. The market has to support more meters and more households being sampled. Now when I say that, it is not just that it has to rise proportionately on a 'x-y curve' (linearly), we have to look at crashing the cost of meters further, so that we can put out more of them.

Can you elaborate on your funding and potential revenue?

To start with, we raised debt finance from banks to set up the system. Ideally, this money should have come from these three bodies but that would have been equity. It was decided to go the debt route to simplify the whole process. Also, the intention over a period of time is to make BARC self-funded.

Besides, around Rs 100 crore more has been invested by the broadcasters to buy embedders, but that is separate.

Going by the 2013 and 2012 numbers, TAM will probably get subscription revenues of Rs 140-150 crore by March 2015. How much is BARC expecting?

BARC is targeting similar kind of revenue or lesser.

Will there be a big outlay if the cost of the barometers goes up further? Who will bear it?

Right now, we have got the meters at a price 'x' but the backend remains 'y' only. We will have to scale that up but it doesn't require too much of an investment. The main cost will be the extra meters that will require more investment. Can these meters be made cheaper? We will have to work on that and are looking at it.

Can you compare the BARC measurement system to any other large measurement system in the world?

Similar systems exist in the UK, France and South Africa. There is one system for the whole nation and that is controlled by a joint industry body. In terms of scale, nothing comes close. All these countries have 5,000-7,000 meters. In terms of money, it's 4-5 times more, even for a smaller sample size. In terms of technology, France has been using something similar for the past seven years, and the US has moved to it in last one-and-a half years.

BARC's is a much larger and complex measurement system compared to any other in the world. Its barometer costs just a fifth or sixth of TAM's peoplemeters. The kind of money we are putting for 20,000 households would be a third of what one would spend for 7,000 meters in markets like the UK, France and South Africa.

How many vendors have you partnered with? How will having multiple vendors help?

BARC has partnered with 26 vendors and the idea is to have control over costs and maintain integrity. The problem of working with one company is that you get 'married' and can't get out. Here, if I don't like one vendor, I can replace him.

In India, our efficiency, the way we operate and the number of hours we put at work is different, apart from the fact that we have much better technology. The people who are working with us are fantastic. There are just three foreign vendors - Civolution (which developed the technology), Mediameterie (which handles the back-end) and Mark Data (for front-end solutions). The rest are Indian vendors like Magic9 Media & Consumer Knowledge (for research), Hansa (for meter installation) or vendors like Intel, who have a presence in India.

Once BARC sets up its audience measurement system, what are the challenges in the next 3-4 years?

The moment you go to a very large sample size, you apply much superior technology and look at things more systematically from a statistical point of view. There is bound to be a lot of change. For instance, there would be expansion of content consumption on small screens, increased geo-targeting by broadcasters and marketers. I think managing these changes would be a big task.

Which bit has been the toughest so far?

Every day, you have a new problem, whether it is getting the right people, vendor partners, choice of technology, choice of breaking it up in the large process system or carrying everybody along in all the decisions.

How much work is left?

There are four different streams coming together. The first is Playout Monitoring. Prime Focus will be downlinking all channels to capture the schedule (programmes, promos, ads) of all channels, which will be fused with timeslot data to get the final viewership data. Testing of data has started. Second is the embedder installation at the broadcaster level. We are looking at 350 + channels to put these embedders in the first go.

Third is the installation of meters and the setting up of infrastructure. The test meters are already on ground and testing is in progress. Fourth is the sample design and installation. Once we finalise the final sample homes, we will install the meters.

What will be the subscription model? How often will the sample homes of BARC change?

We still have to decide on the packages for subscription but not all data may be available to everyone. For the first time, we will put the whole rate card on the internet. The software will be the same for all subscribers, but what is enabled could be different. Like the current measurement system, we will also keep a track of how many terminals the software is running on. Twenty five per cent of sample panel homes will be replaced every year.

Any interesting method that you would be adopting apart from the ones mentioned?

There is one on the GUI software, which we are working on. A subscriber can watch 4-5 different channels on one screen. Visually, you can see the ratings change, how people move from one channel to the other, every second. That's the module we want to introduce once ready.

To download the PDF version of the article, click here.