Comments on: Top Confusions in Digital Media

By: Viral Jani

Viral Jani — Thu, 07 May 2009 06:53:29 +0000

If Comscore has a 16000 panel, that is already huge compared to TV…a panel of 16000 individuals to track user habits of 40 mn ppl vs a panel of appx 35000 (8000 households) to track viewership of 450+ mn ppl…

By: Komli Launches ViziSense, Audience Measurement tool |Technology and Business Startups in India

Tue, 02 Dec 2008 05:31:24 +0000

[...] days back, I covered the top confusions in Indian digital space and lack of a common audience measurement entity was the most prominent [...]

By: antispin

antispin — Thu, 06 Nov 2008 02:05:56 +0000

Do we know _exactly_ how each of those sites chart out the numbers? All I have read are vague (and short) descriptions of what’s going on behind the scenes. I wish there were a completely transparent, open-source, solution for tracking web footprints. Come to think of it, isn’t there something in the works at Mozilla Labs?

Meanwhile, everyone is free to choose numbers that make their site look good.

By: Mrutyunjay

Mrutyunjay — Wed, 05 Nov 2008 10:31:05 +0000

Different sources and different methods always give different numbers, there is nothing new or interesting about it, apple trees gives apple and orange tree gives oranges only, you don’t compare Apple to Orange and say why are they different… but you prefer one over another or prefer them equally.

The confusion would always remain there as long as there are more than one source for any information, hence it is always the information seeker who decides which source to trust more.

Even if it is possible to doubt “inflation” or “GDP” number you can’t do anything about it and may not even write about it as it is just one source… “Govt. of India”, you don’t have a choice on “Population figure” as well, because it is only “Indian Census” which provides you that, but here in case of Internet you would as there are multiple sources.

The game in “sample based measurements” is always of “representation of universe”. If any measurement agency makes sure that their panel (people using the toolbar or utility or survey or whatever that tracks usage or asks usage) are demographically or behaviorally representative of universe than the number can not be grossly different from reality and from each other. To study stray dogs we must pick the right ones representing all kind of stray dogs in the area, the one sleeping bellow our car can’t be the universe.

- Now the next question is at “what level of depth is this representation achieved?”
# If a 2K or 20K panel represents overall Internet user base than overall numbers should be correct but if it doesn’t represent a segment within it e.g., “the dating users” on the net properly than the “dating website related data” it will throw would be misleading.

- Finally, there is always one best method, like if a city has the affordability then someone opens a “super specialty hospital” otherwise people keep on depending on the “next door nursing home”, if people can’t afford there is no point discussing the need for a super specialty hospital. Therefore the ultimate question is:

“Is the Indian Internet Measurement Buyers ready to pay for the best methodology?”

By: phaniraj

phaniraj — Wed, 05 Nov 2008 08:19:00 +0000

@sampad, isnt IAMAI supposed take care of this as per this link http://www.iamai.in/introabout.aspx,

By: Sampad Swain

Sampad Swain — Wed, 05 Nov 2008 06:47:01 +0000

Agreed! Even I think lack of legitimate & trusted authority make Indian digital market very murky.

So is PluggdIn jumping into this space soon? Would love to see one from Indian digital domain to be there!

Best of luck for all your endeavours!

—
Sampad