Posts about measurement

Viral Nielsens

I’ve been wanting to know what viral videos had infected the most conversations and was thinking about trying to do something that looked at links and insertions into blogs and such. Well, somebody just went and did it. Scott Button unveils the Viral Video Chart. What’s interesting about this is that it isn’t purely traffic. It is about conversation. They explain:

We scan several million blogs a day to see which online videos people are talking about the most. We count the number of times each video is linked to and the number of times each video is embedded. Every morning, after we’ve had a cup of coffee, we publish a list of the 25 videos that generated the most buzz over the previous day. We reckon this is a pretty good yardstick of what’s hot and what’s not. At the moment we only look for references to videos on the three most influential video sharing sites: YouTube, Google Video and MySpace. We tried looking for references to videos on some other sites for a while, but nothing ever made the top 25 so we stopped.

Today’s top video, Keith Olberman slapping Bush around:

Stand up and be counted…. and counted

The Audit Bureau of Circulations — the agency that verifies the copies sold by newspapers and magazines — just announced a new effort to combine print and online audience numbers.

But their logic is off. The Times says that for their first test case, Advertising Age, they “arrived at a figure by combining the publication’s print readership with a day’s worth of its Web site traffic.” The Times also reports that newspapers say this isn’t accurate because it doesn’t account for people who read both the paper and its online site. And I say that measuring only one day is ridiculous, for as many studies have found, very few people come back to online sites every day (and the truth is that they probably didn’t read papers every day, either; it’s just that they were delivered every day). If the goal is to measure total audience — total reach — then this will inevitably fall way short.

I toiled for way too long on the ABC committee that first tried to measure online (I suffered through endless meetings debating the question, ‘What’s a page view?’) As it turned out, advertisers didn’t really care about auditing the audience of online sites — and thus publishers didn’t want to pay for the service. Advertisers care only about auditing their own flights of ads; the wanted to verify that they got what they paid for. And that makes sense. For in an online site, you don’t really care how big the whole site it; you care about how many people your ads reach how many times.

The ABC and other old media players keep trying to use old math with new media. It won’t compute.

Numbers game

The Times goes on an odd attack against and its claims to be an audience leader.

The Times’ real complaint is against Comscore and other providers of the numbers — like every big web site — uses. These numbers are gathered from small samples — a la Nielsen TV ratings — and they are relative bullshit but they’re what advertisers go by. Their veracity is ever less reliable the more the media world spreads. When there were only three networks, a small sample was probably a fairly reliable indicator if not measurement of viewership. But with millions of sites, the odds that a small sample will go to them in the same proportion as the rest of the world falls to nada. (This, by the way, is why blogs are doomed in a Comscore/Nielsen world; there’s no way that they can be measured. And that, again, is why we have to do a better job with our own measurement.)

This argument over ratings numbers is less important online because advertisers need not care how big the site is, only how often the ads they pay for get served — something the web can measure and verify and TV can’t. I served on the horrible Audit Burea of Circulations committee that dealt with these issues years ago (“What is a pageview?”). As it turned out, advertisers didn’t care about audits of how big sites were. They needed audits of their own ads. So whether Forbes has X million users or half that, it doesn’t matter to the advertiser so long as his ads get served to the right number of people.

One more note: I think I found one odd reason why keeps growing. For unknown reasons. GoogleNews favors way over other sites. Look at this analysis of citations on GoogleNewscompiled for more than a year since GoogleNews had a neonazi site and wouldn’t reveal its own sourcdes — and note how high is. The guys didn’t even know this until I ran into them on a panel and told them.

: Rafat Ali weighs in.

Podcasts get ratings

Nielsen released a report today on the economics of podcasting with some juicy stats to add to yesterday’s Pew numbers (here’s a only I to a PDF of the press release):

* 6 percent of U.S. adults — 9 million people — have downloaded podcasts in the last 30 days. The same number call themselves regular podcast listeners.
* More than 75 percent of them are male.
* 38 percent of active podcast listeners told Nielsen that they are listening to radio less often.
* The most successful podcasts, Nielsen says, are get two million downloads a month. (I’m curious to hear the stats for Diggnation and other big ones.)
* 60 percent said they always fast-forward past commercials.
* 72 percent of regular downloaders get one to three podcasts a week; heavy users — 10 percent of them — take eight or more.

Nielsen also said it is going to launch an iPod panel with 400 users. That’s good. But I’ll caution that you can’t measure the mass of niches that way you could measure the masses; a sample won’t get the — pardon me — long tail. Still, in a new medium, data is good because it makes the medium real.

The value of the conversation

Now you’ll see why Intelliseek and Buzzmetrics sold: Here]’s a Hill & Knowlton PR exec using the tools of buzz and the methodologies of Steve Rubel and others to figure out how to listen to the conversation about products and industries. He even creates a neat little tool for companies to chart themselves. I also hope they don’t continue to see the public as numbers on charts, though. The real way to listen is not to chart but to read. [via Rex] (Full disclosure: H&K just asked me to speak to a confab of its execs.)

So blogs will have overnight ratings and cumes

Fred Wilson reports that Intelliseek merges with Buzzmetrics and takes a majority investment from VNU Nielsen (which is, in turn, the subject of a takeover bid). To me, this means that measuring blogs and distributed media will matter more and more to advertisers and there will be a growing market for such analytics. It also is another indication of the big changes coming to the mix of ad dollars. Yes, we are ready for prime time.

A principle: I have a right to know when I am read

How about this as a fundamental principle of content and conversation on the internet:

I have a right to know when what I create is read, heard, viewed, or used if I wish to know that.

That is my followup to the whine about RSS — and content — caching below.

If this simple principle were built into applications — not the internet, per se, but in how readers and viewers work — then caching and P2P, which both serve creators by reducing bandwidth demand, would not be issues. This also would help those who want to make use of advertising (though actually serving ads is a different matter).

I’d like to see this as a technical add-on to Creative Commons: Distribute my content freely, please, on the condition that you allow applications to report traffic back to me. And applications designers should build such reporting in. The creator is still free not to require this and the end user is still free not to consume those things that require ping-backs. But simple traffic reporting is at least common courtesy.

I’d like to see this work for RSS, HTML, audio, and video.

: ALSO: Scoble, Winer.

: LATER: Just want to emphasize that My Yahoo will provide the data. It is not now because something broke in an upgrade but two Yahoo folks have confirmed that they will continue to play nice, for which I am grateful.

: LATER STILL: Matt Cutts of Google says he will mention this to the guys at Google Reader and believes there’s no reason not to build it into a next version of that new product. Bravo again.

My content, my readers, my numbers, damnit

Hey, My Yahoo, Google Reader, Pluck, Newsgator Enterprise and other RSS readers: Hand over my numbers. You are taking my RSS feed and caching it to serve more efficiently, which would be fine if only you told me how many times you are doing that. But you’re not.

Brad Feld is much more polite than I am about this. He complains that My Yahoo just stopped reporting how many subscribers a feed gets there and Google Reader never did report and many others, including those I list above, don’t report subscribers, even though there is an easy and automated way to do that.

That’s theft. If you took a song and cached it and fed it out to lots of people these days without reporting back to the owner, you’d get sued or slapped in jail.

Well, all I ask that you do for caching my feed is to report back the number of subscribers. Not much to ask. And not doing that is tantamount to theft.

Why do I care? Because I have an ego. Because I want to see how much RSS I serve and learn about it. Because I want to see how efficient my advertising is. And just because. Damnit.

RSS is becoming a ever-more-important transport mechanism but without metrics, some will refuse to be transported by it. My Yahoo and Google Reader are making hay including RSS in their new products. They should practice good citizenship and share the data those feeds generate with their creators.

I can’t go to the Syndicate conference this time, because it’s in California, but if I were there, I’d wear a T-shirt and carry a picket sign to all the players listed here and in my Feedburner report:


: LATER: I should add that I’m not against caching because it saves on my server load. But I do want to maintain a relationship with readers who subscribed to my blatherings and the barest way to do that is to get statistics. I also am not crazy about services changing feeds without my permission; some cut my full-text feed back to just headlines. Do newsstands refuse to tell you how many copies of your publication they sell? Do they cut out pages and give you only covers? No. Online distributors should operate by similar rules of the road.

: UPDATE: Jeremy Zawodny, of Yahoo, reports in the comments that the Yahoo counts will be back; it’s a bug to be fixed. Bravo. Now how about you, Google?

: LATER: See a followup post on a fundamental principle, above.

: LATER CONFIRMATION: I also just heard from a Yahoo exec who confirms that, indeed, something got broken in an upgrade and that they will feed back stats on feeds. Once again, thanks, Yahoo.

: LATER STILL: (Repeating this from the post above): Matt Cutts of Google says in the comments here that he will mention this to the guys at Google Reader and believes there’s no reason not to build it into a next version of that new product. Bravo again.