Rojo

select Quantcast’s Promise Overshadowed by Online Audience Measurement Marketplace Realities

+ 0

• HipMojo.com - IT, Video, Web, Technology, Gadgets (+subscribe)
• By froosh
• 1/10/2007 18:45 PM

expand close

Summary: 2006 saw a growing backlash against the all encompassing page view.  I don’t quite know why the page view is so hated.  Of course, when page views are used as the gauge of a site’s popularity, there is a problem.  But to suggest that the page view...  Click to expand...

2006 saw a growing backlash against the all encompassing page view.  I don’t quite know why the page view is so hated.  Of course, when page views are used as the gauge of a site’s popularity, there is a problem.  But to suggest that the page view must in 2007 find its place alongside Pol Pot, Benito Mussolini and Adolf Hitler in the annals of history is lunacy.  But when page views are given too much importance in the context of a site’s popularity, problems arise.

In this article, we look at some of the established players in the online audience measurement tool (OAMT) service sector and look at one startup.  Before doing so however, it’s important to consider the variables that create success in the industry.

For relative newbies: some definitions. 

People visit sites over a month and are considered its visitors.  Each visitor is a unique user.  The total pages they generate represents a site’s page views, and based on the number of ad impressions per page, a site’s ad inventory represents the number of ad impressions they can sell to clients.

As such: page views drive ad impressions, and ad impressions drive ad revenue.  It’s not a perfect model, mind you, but it has driven the world wide web’s business and revenue models for years.  Sure, Google became the ad platform by pushing clicks, and not page views, but the fact remains: page views should not go anywhere. 

Of course, page views drive ad impressions because page views also give a glimpse into a site’s popularity.  It’s not rocket science, more page views; more readers.  Right?  Wrong.

What needs help is online measurement of audiences.

For the longest time, marketers and publishers relied more on unique users than page views. But as consumption of content evolves though the proliferation of syndication tools such as XML and the return of widgets, there is a new race to develop the tools that marketers and publishers can turn to to better assess the size and demographic of online audiences.

But that’s not the problem.  The talk of page views should die misses the point.  A more underlying question is: are panels the way to go?  The OAMT all use panels and while panels are better to remove the duplication of an audience due to cookies, there are many bad things about panels.  If panels worked, Al Gore and John Kerry would be President of the US.  Oh wait…

Anyway, back to technology matters: much the same way that the Web’s 2003-current era rejigged companies, products and features of the 1994-2003 era, we are now seeing a number of online measurement companies vying for a leadership position. We disagree with people who say that such new players will eat the lunch of established players like comScore Media Metrix and Nielsen//NetRatings (the old guard), but we certainly see a spike in competition which will in turn lead to innovation and invention.

One such upstart is Quantcast. Quantcast ”is the world’s first open internet ratings service. Advertisers can find reports on the audiences of millions of web sites. Publishers can ensure their sites are represented accurately by tagging them for direct measurement. The service is free to everyone.”  

I first came across Quantcast on Steve Rubel’s blog. Mr. Rubel wisely and correctly asserted that in a contextual, vertical niche landscape, the old guard will have some difficulty remaining relevant, and in that context, will see other players eat their lunch. We disagree on the latter part. I personally do not see how an upstart can dominate an industry like this. Like what, you ask?

CPCs, CPMs, CPAs, and CYA.

Between Arthur Andersen’s creative accounting and Enron’s woes, accounting has taken a hit.  But, that does not mean that a new player in the space would be welcome.  Yes, accounting is different from online audience measurement, but they are also similar in that they are both CYA industries.

A CYA industry is a Cover Your Ass sector that relies on deep research, analysis and mainly reputation and credibility. We are a society that quotes things and lists sources. While the blogosphere has accelerated the fact that online, we sometimes seem to content ourselves with a 95% accuracy rate, the simple fact of the matter is that online audience measurement services are deeply rooted in the psyche of decision makers. Even if these know the shortcoming of said products, there mere existence and being is enough to maintain their relevance. In other words, it’s not “relevant” whether comScore (or NNR) is flaky or not, as asserted Michael Arrington of the popular blog TechCrunch, or that comScore VC Fred Wilson refuted Arrington’s claim.

What is relevant is that they are there, have been there, list the major advertisers as their clients and have built defensible businesses. At a time when everyone is raving and ranting about advertisement-supported, free services, there is much virtue in listing clients who subscribe to your service and pay you. Furthermore, there are major conflicts of interest for online audience services to be free, ad supported businesses (imagine pulling a report on who is larger, Cars.com or Autobytel.com and seeing an ad for a Chrysler, while that is targeted and useful advertising, it’s just not right). In this context, the success an online audience measurement service attains in being able to generate interest from marketers and publishers is essential and fundamental to their long term success.

The little not-so-secret secret of online audience measurement tools is that all of its users use them suspiciously, reluctantly but ultimately, religiously.  Why?  Because they need to cover their asses.

Suspicious Use of…

- Why media planners and buyers actually rely on OAMT: think why companies use accountants.
- Why don’t media planners and buyers actually rely on OAMT: site-specific analysis is the way to go.

Reluctant Use of…

- Miss out on opportunities
- Herd mentality
- Institutional Imperative: “everyone does it, so…”

Religious Use of…

- CYA theory
- Nothing better: Alexa, CSMM, NNR all have a lot of weaknesses

Enter one Startup: Quantcast

As a result of this marketplace reality, startup online audience measurement services face a very tough challenge, one that other free, ad supported services do not necessarily face. MySpace could overtake Geocities or Friendster because having clients use it was

Quantcast aims to provide open access data on Web audiences.  “As the Web has grown phenomenally and democratically, so has the content found online,” starts off Quantcast co-founder and CEO Konrad Feldman.  It could be argued though, that “online measurement tools are anything but democratic and open,” he continues.

What on earth is open access data in this context?  Quite simply, it entails simply relying on a snippet, site-specific data, as I call it.  Make no mistake about it, until a site operator adds Quantcast’s code onto their site, Quantcast is a panel like all panels.  There are nuances with regards to the “collection mechanism and similarities in terms of statistics.”  Feldman is quick to add that the main, major methodologies in all panel services ultimately adhere to statistical theory and practices.

In case you are wondering, Quantcast’s panel size is 1.5 million people.

It is true that compared to offline media, online is a very fragmented media.  Sure, magazines, newspapers, TV and radio is fragmented too, but the level of fragmentation is simply breathtaking on the Web.  You can have an audience of 1 and still have a cost effective solution (but who wants that?)  With every other media, you need large numbers and economies of scale.

According to Feldman: “the tools used in offline media, such as ratings points, simply do not work in a long tail environment,” exasperating the matter, he argues, is the fact that “the bulk of ad dollars go to the Top 5 websites, meaning that a lot of good content gets overlooked.  These are high quality audiences, it’s not like the audience is worse, in some instances it is a more homogeneous group that is better for marketers.”

It’s certainly true that advertisers will welcome new tools such as Quantcast that provide more and not less clarity into a website’s demographic makeup. 

While the OAMT are anything but a perfect mousetrap, I then ask Feldman if he thinks that marketers will welcome a better mousetrap.

“Search is a response driven media and TV is a demand generation media, [display] online advertising is not yet seeing the ad dollars migrate as much as they could because advertisers have some trouble finding the real audiences they are looking for.

And what about publishers?

“The biggest concern publishers have is misrepresentation of their site and their audience.  A free Quantcast is far more accurate…” and is quick to add that “open access data will help advertisers.”

It is certainly true that publishers will essentially welcome and embrace a product like Quantcast.  I have been openly lobbying Google Analytics to allow publishers to make their site data public.  We have Google Analytics code on our pages, I would gladly open the data to visitors on our site and Google’s main page (in fact, hint hint Mountain View:  if you so desire and aspire to become the next next-generation advertising platform, that might be a fantastic tool to allow advertisers to access… publishers can make their profiles anonymous whereby an advertiser can see the demographics and contact the publisher through google for more data, at which point Google can get a cut of the revenue, if any.  Or, the publisher can make their name public. 

In this case, Google bypasses revenue but becomes a place advertisers return to for marketing needs… which helps them achieve their goal of becoming a platform for advertising.  After all, in the information age, the new generation ad agency might be more of an information broker than a provider of labor (mainly).  But, we’re getting sidetracked somewhat, so let’s return to Quantcast, who incidentally to has aspirations of becoming a resource marketers turn to.

I asked Konrad how he plans on generating revenue for the company, and he has plans “to assist marketers by facilitating and enabling a marketplace.”  He was candid to admit that this has yet to be determined.  Or, maybe, he was being coy.
I asked directly about the threat that Google analytics poses.  Here are some previous things I have written in the space.  “Google Analytics is a phenomenal tool, we use Analytics, but it helps you know where people come from, and the website is used.   We tell you more about a site’s demographic.  Google Analytics does not tell you about demographics or your audience’s characteristics.”

Basically, one can extrapolate: Google Analytics helps you reduce marketing costs, Quantcast would help you get marketing dollars.

Light bulb flashes: “Hmm,” I thought, “acquisition play, perhaps.”

Hailing from the space of transactional behavioral analysis, Feldman’s experience lies in assisting large banks in the search space “understanding audiences to detect patterns of money laundering and fraud.”  Joining him at Quantcast is Paul Sutter, AltaVista’s former VP of Engineering.  Naturally, operating in a sector that is under the spotlight and having successful backgrounds, I asked if the company was self-funded or venture-capital funded.  Feldman mentioned that
Quantcast was self-funded but that venture capital was something they were “considering.”

Of course, when it comes to the company’s options, they are (either alone or a combination thereof):

0 - status quo, assuming they won’t run out of cash
1 - raise financing from companies with interest in the space (newspaper companies, magazine companies, etc).
2 - raise financing from a venture capital firm, who would require an exit strategy, with it being selling the company to someone over time
3 - trying to become cash flow positive and remaining a stand alone entity or selling eventually to another company
4 - sell now

So, what are the company goals for 2007?

Feldman lists two:
1. continue to develop a quantifiable publisher service and
2. add features for marketers to better understand the product and the publisher’s audience.

We then decided to actually take a look at WatchMojo.com’s audience, we are a bad example though, I will first explain why: initially I set up the company as X number of different properties.  I was influenced by the models if IGN (not IGN.com, the structure of the company when it launched as Snowball.com, as a network with many sites, or the Conde Nast company’s web network, Conde Net. 

Specifically, what Conde Net does with the numerous titles cross-promoting one another online.  Of course, these are high quality, offline properties that have been around for decades.  Setting up a website automatically converts to traffic.  With our sites, I initially set it up like that, but for practical and marketing purposes, I realized it makes most sense to simply redirect all channels under one comprehensive portal of sorts.  Technically that would have been Mojo Supreme (supreme as in “greater,” so to speak, and with all channels being somethingmojo.com, but now all of the content is under WatchMojo.com (while the search units are all under MetaMojo.com).  That is why HipMojo.com now redirects you to WatchMojo.com/web/blog, for example.  The simple candid truth is that this allows us to bolster a stronger reach and larger audience though we dilute the demographic targeted-ness thereof.  It’s a tradeoff, frankly.

What is our demographic?

“Your audience has an interest in music, cars, film, right?” boasts Feldman.

Hmm… yeah, I guess.

Like I said, we’re a bit unique due to the origins of how the network came to be.

Digging deeper:

As it stands now, Quantcast has ratings for 20,084,867 sites.  WatchMojo.com’s rank is 54,587.  For purposes of comparison, Alexa’s ranking for WatchMojo.com is 18,420.

As per demographics:

- 1.08 times more likely to have women than men
- 1.72 times more likely to be 18-24
- 1.37 times more likely to make $100K and more
- 1.31 times more likely to be Hispanic
- 1.23 times more likely to have no College
- 1.80 times more likely to have children 8-17 in the household they reside in.

Gender-wise, we are evenly balanced, with a slant to Women (index 108).   I will be the first to argue that our site has men and women on it.  The slight edge to women might have more to do with our strength in fashion (covering fashion shows, giving tips etc.)  Also, we have a lot of travel clips: women historically plan trips, be it at the workplace or in the home.  I would guesstimate that

[This, by the way, shatters once and for all my erstwhile colleagues’ argument that I violate my non-competition.  They were “a poor man’s Maxim magazine,” in the words of an Alexa commenter.  Forget the rest of the evidence…]

We predominantly have 18-24 year olds.  This might be due to our excessively large video game, music and film sections.
While I firmly believe in the niche, in the end, numbers are important and size matters.  By getting all excited about Long Tail theory, we sometimes forget that there is an ecosystem out there, and being rigid on one side of an equation does not facilitate a marketplace, it complicates and hinders it.

There seems to be a lot of households with kids 6-17.

Income is strong, almost too strong in my opinion for the age profile.

Education is consistent with the Web and country in general.  Also, since it’s a young demographic, they have completed less education.

All to say, to me, this suggests that a) Quantcast is a nice addition to the landscape and that b) WatchMojo.com is in fact a barometer of the Web (that was the goal), and through each vertical, we can target audiences for advertisers.
Great, now, will marketing decision makers turn to Quantcast and seek more information on our site and others?  Who knows.

All I know is that marketers also want something to cover their asses.  Hitherto, that boils down to reach and size.  I am not sure, then, that Quantcast really meets that.  Well, it does in the sense: by adding their code to your site, you provide an open and accurate assessment of your traffic.  We only added the code early this month, so I will let you all know after a month, so in a couple of weeks.  One advantage of adding the code is that you open up the data to global stats, since the panel is US-only.

I have no doubt that other publishers like me will want to add data.  I was personally looking for something to show my former colleagues that I had an even audience between genders.  Oh, yes, I also like the extra data.

The key however is will agencies turn to Quantcast.  Therein lies the catch 22: agencies will look at Quantcast because it’s free and because more data is always better than more data.  But unless enough publishers rely on Quantcast, then it will not be worth the agencies while to turn to the service, since they spend enough time pouring on the OAMT

If they want adoption, here is what I recommend: develop data on portals’ channels.  This will increase adoption on both sides of the equation.

But the real challenge is that other online audience measurement tools can simply offer something simple like this to publishers that they can add to their site.  As great as Quantcast could be, not sure how defensible it is.  That being said, there are numerous competitors and one of them might opt for a buy instead of build strategy, in which case Quantcast is a perfect takeover candidate.  Then again, by buying a site-specific date site, aren’t the panels admitting that panels are not 100% reliable?  Well, more data is better than less data, I presume.

Last but not least, I could see Google making a run for this company, since it fits well with Analytics, but Google could build such a product to ensure it is more in tuned with all of their existing data mining projects.

When everything is said and done, we welcome new players in the space and wish Quantcast all the luck in the world.

Check out Quantcast here. 

 

All Mojo'rs for Quantcast’s Promise Overshadowed by Online Audience Measurement Marketplace Realities

• 

• Flag
• Email
• Rojolink
• 

close