Panel-based measurement is better: LV Krishnan

21 Oct,2015


Last week, the Kantar Media-Nielsen jv TAM Media and Kantar-owned IMRB announced the TeleWeb ratings for an integrated study of viewership on television screens and the internet – desktop/laptops as well as on devices. We caught up TAM CEO LV Krishnan to understand the finer details of the offering…


You had spoken about the multiscreen measurement earlier. But do you think it could’ve come in a year or two earlier?

Well, it required a lot of work to put it together because one is a non-linear medium and the is a linear one. To bring together both into the same database and be able to tell an advertiser that using a combined medium maybe better than using one medium etc requires work. We did a lot of groundwork. If only publishing numbers in terms of saying which sites were getting hits and which sites were getting what kind of profile of audiences. it could’ve been easy. It could’ve happened much before than now. Probably about 4-5 months back. But to integrate both, this is the earliest we could’ve got to.


Given that there is an MoU with BARC on meters the merger move is work in progress, what happens after the merger happens?

We’ll see it at that time.


BARC is also doing something similar, it is also looking at multiple screen measurement.

From our perspective we are looking at the present moment of time. We’ve done almost a year-and-a-half of work into this exercise. Obviously, since we are still running the television panel, the idea was to launch it and help the industry use it. What will happen a couple of months down the road is something we haven’t thought about at all. From our perspective, the work on TAM BARC integration is going well at a high speed level. That’ll continue.


Since IMRB’s digital measurement exercise has been on for some five years now, couldn’t this have happened earlier?

Two important things. Before we launched something like the integrated data we also needed to ensure the fact that the two panels are stable. TV was completely stable. The online one was a new one launched around 2011-2012. So it took time for that panel to settle down and get used on a regular basis because at that time if you look at it the advertising online was hardly around 2-3% of the advertising that was going over there. So, even if we had launched at that particular moment of time, the integrated quotient, the usage could’ve been very low. The demands coming in from the advertisers that we need do it the integrated way. Only in the last one year, in the beginning of 2014, when we were interacting with advertisers that there was a huge push going towards digital. Probably also because of the fact that most of the advertisers were MNCs were also being pushed by the global directors saying that let’s get starting investing on digital because globally digital was picking up strongly.


What we previewed at the press conference last week was about English channels and the Hindi news channels. But I’m sure the difference is significant with English GECs.

English GECs are basically international sites. So, we need to see what you want to track within that and how much of it is with the Indian audience largely. So, probably with time we’ll look at English entertainment also as a next step.


Do you see a lot of viewing happening online in the case of all GECs?

There is, Hotstar, Sonyliv, Ditto, all of them are coming in the Top 50 sites. Platforms like Hotstar are among the Top 5.


Now a channel like Colors has ‘Comedy Nights With Kapil’ which is among the most watched as per your numbers, but a channel like Colors which targets urban audiences is not in top channels list in tele-web, vis-à-vis Zee which is on the list?

That’s because of Ditto TV. Ditto is the push for Zee. It has certainly got some audience in the metro markets. Colors is yet to have a platform that could be visible and for which audiences are walking in.


So, will you say OTT is playing a big role?

Yeah, certainly.


What is the kind of response that you’ve seen from the advertising community, from the advertising agencies and media agencies?

We don’t know yet the response because we just launched today but we’ve been working with the advertisers on this for almost a year now. We connected with them and we realised what they are looking for. So, therefore from that perspective this is exactly matching up to what their expectation was from a feasibility perspective of trying to alleviate a campaign on a multimedia level.


There is no joint industry body to contend with the way you work. There is no industry association as such…

No, there is no industry association involved in this. We are on our own and like any other research we are marketing it on our own for users to take their independent decisions whether they want to buy it or not or use it or not.


There have been views on the entire element of panels. In the case of television, the cost of set-top boxes ensured that the sample size of panel homes couldn’t grow. But no such thing exists for the internet, so why only 6-8000 sample size? Do you think it can be browser-driven so that anyone can download and install?

There are two different kinds of measurements that happen across all media. One is measurement based on universe and second one is measurement based on panels and specific sub-groups of profiles. The problem comes in when the universe study is when you don’t know who the individual is, who is responding to a particular viewing. So, most of the dynamics related to the universe measurement is already available with the publishers themselves like a Google or a Facebook knows about their customers largely and to that extent Facebook I could say knows it even better because age and gender is something they capture apart from the geographic location which is universally available to them. So, therefore, from a universe perspective, limitation is a fact that while you’ve a larger base you don’t know who that base is, who that individual is in terms of profile. So when you are targeting it, you are targeting more based on the behaviour pattern of the respondent rather than on the basis of demographics. In a panel0based exercise you are able to measure demographics as well as in terms of behaviour patterns together. So you know individually what the profile of the guy is, what profile of home he comes from, what are the numbers in the home, what are the platforms they have in the home that can be access point or a competitive perspective, a competitive touchpoint. So, all this kind of information is available on a panel-based exercise. Therefore, we are able to integrate between two mediums very effectively in this kind of a panel-based exercise. In a universe level it’ll be very difficult to integrate two measurement systems. So, largely speaking, for an understanding of a consumer behaviour and trying to monitor it on the basis of his behaviour, a panel makes sense actually. But in the longer run, we could actually do it the way it happens in television where the set top box data and the panel data work together. To therefore use a panel data is to optimize the set top box data. Similarly here too the panel-based data can be worked on to the universe data. You could mine better data at smaller discreet levels. These things will happen in future.


How much of the TAM sample are you looking here?

Six metros put together will be about 10,000 individuals actually, which is close to 2500 paneled homes.


On a lighter note, had the news on NDTV scoring so high on TeleWeb come in earlier, things would’ve been much nicer for TAM?

No, we take it as it comes. For us the most important thing is to indicate what is happening in the behaviour front. Business decisions are taken by users.


Post a Comment 

Comments are closed.

Today's Top Stories