NebuAd has been in the news this week, the CEO Bob Dykes testifying at a House hearing on behavioral targeting and privacy. Apparently quite a few NebuAd employees were laid off in the past week, as the company prepares for a long, protracted battle to be able to do business despite the privacy concerns being leveled at them.

Companies like NebuAd have definitely been pushing the boundaries and many in the industry (myself included) have felt that they would be on the front lines of any kind of regulatory action surrounding opt-out and opt-in for online tracking. My main concern here is that the good progress that is being made in improving advertising relevance through the use of mainly cookies and other anonymous non-download methods, doesn’t get undermined by and pulled in with some more-aggressive types of behavior and the (potentially, alleged?) more intrusive types of data mining being conducted by the likes of NebuAd and Phorm.

The industry itself has a dearth of knowledge and understanding on these issues — lots of advertisers and publishers aren’t up the curve on what the differences are between some of these networks, companies and their methods; let alone having regulators and the public at large understand (many of whom I doubt actually really care… but certainly that is not an excuse to avoid being more clear and providing better notice and opt-out capabilities). We may be back at a point where the online ad industry needs to “come clean”, set some boundaries and start educating everyone on the merits and details of various consumer-based targeting systems.

Note: Like other targeting firms such as aCerno (which uses online purchase data for targeting, in a data co-op variant), NebuAd buys advertising inventory from many online advertising networks, including from CPM Advisors. Networks like ours have very limited insights into what advertisers or kinds of ads are being run, nor into what users are being targeted. In this case, we are really just a dumb conduit for ads that these companies buy because they are “looking” for users with cookies that they have previously placed there via various shared cookie systems.

RockYou has improved the quality of its reporting for app developer-publishers quite a bit in terms of look and feel, but the export capability is still lacking. When you export to Excel, the data comes out as text (those little green corners with the ! that encourages you to convert the text to numbers).

Just a hint - if you are going to create a reporting system for your publishers from scratch, please please please spend some time working on the export functionality. Pivot-ready tables would be awesome (hint hint)!

Of course we should mention that Cubics (part of AdKnowledge now) doesn’t even have an export option… cut and paste is the method here.

Our latest aggregated look at the advertising data made available by Facebook as part of its Social Ads program. All of this data was gathered on July 16th via their advertising interface. We looked at the number of users in the “Any” to “Any” age range and then also looked at the suggested-bid CPM figures it spat out. Note that when we looked at the suggested cost-per-click bids, the implied clickthrough rate for Facebook ads was pretty consistent across the top 10 countries by size, at around a 0.044% to 0.047% rate (someone is clicking on these ads, on average, about 45 out of every 100,000 times the ad is shown).

Here’s the data - when you add it all up, almost 83 million users across these 93 countries with a weighted-average “suggested” CPM of $0.21 to $0.27 to run your ads. We’ve tried some ads and seen a fair bit of volume below these levels, so it’s doubtful they are conservative. If you want an Excel file of the data, join our advertiser mailing list on right and we’ll send the data to you!

Don’t be afraid to experiment with different ads and targeting. Be sure to check in on your ads regularly to monitor their performance. Try new combinations of targeting, creative,or bid amounts using the data available in your Ad Manager from past ads to help guide your edits.

Facebook Ads Newsletter, July 1, 2008

Gang - don’t be afraid of experimenting. Shudder, that nasty word is “optimization” and you too can “optimize”! It’s kind of like doing your own soldering with a soldering iron. Definitely something you want to do yourself.

The WSJ ran a story about LinkedIn’s $53 million cash infusion here. And I found this line interesting:

LinkedIn’s average user is 41 years old and has a household income of $109,000, according to Mr. Nye.

Facebook said it doesn’t release specific demographic information and declined to comment. But the company’s Web site says more than half of Facebook’s users are outside college, and users aged 25 and above represent its fastest-growing demographic group.

Hmmmmm, as we pointed out in February, Facebook still has a pretty young audience (about 23 years old, on average when we looked at it then - based on Facebook’s own numbers) . It looks like the parents use LinkedIn and their kids use the Facebook, on the whole. It’s too bad the writer of the article didn’t find places online where this data has been collated, such as our site. We’re probably due for an update on the figures, it’s true - but we’re always happy to help out.

Ad networks and ad representation firms typically charge publishers a revenue share of between 35 and 50 percent to bring advertisers to them. Traffiq launched in late 2007 touting itself as an online ad exchange where “Sellers pay 30 percent on ads sold; buyers pay no exchange fees at all.”

I guess I don’t understand how this is different from a regular ad network from a business model perspective. They don’t charge advertisers directly either (more on this in a second), and the rate, at 30 percent, is certainly in the ad network ballpark. The difference Traffiq has is that it has an open marketplace where buyers and sellers list information about their inventory available and inventory requirements. There doesn’t appear to be external validation of the publishers ahead of time, though when you run a campaign you have to grab their ad tags which they use to track delivery - so there may be other types of validation going on to ensure the ads are running on the site they are supposed to etc.

The problem with this model for advertisers, however, is that while they are not explicitly charged a fee for participating, they are paying in another way by simply not getting as much for their money. The more advertiser money intermediaries are soaking up, the less gets into the hands of publishers, and the less work the publisher will end up doing for the advertiser in helping to optimize their ad campaigns, find the right audience, and so on. In fact, the less sustainable the publisher’s business becomes… partially the publisher side gets propped up by some of the lack of transparency of prices and the fact that large publishers can often grab some of the price surplus because advertisers don’t have a view into the pricing environment (but that’s the topic of a later post!) but I digress.

I don’t mean to pick too much on Traffiq — I like their UI, I admire what they are doing in terms of adding more transparency to the ad network model, but they are indeed an advertising network. I don’t believe that if they really do want to be an exchange, that  Traffiq’s model is sustainable. The revenue share fees will have to be much lower - because otherwise the incentive is too great for buyers and sellers to find each other outside of the exchange. Even eBay, who developed a marketplace charging under 7 percent of sales historically faced and fought this issue and was able to win with the ultimate weapon: massive scale.

There’s still enough inefficiency in the online ad market for 30-percenters to make a go of it, but with an open platform I don’t see the opportunity for this type of model to achieve the massive scale that will bring with it so many other benefits.

The IAB email this morning caught my eye because it referenced a BusinessWeek article that mentions a study done by Rapleaf (a company that aggregates social network information and has come under fire for some of its marketing tactics), saying (my emphasis added) “Rapleaf CEO Auren Hoffman, citing a recent study by her firm, writes about the gender gap in social media…” - I thought this was funnily ironic because Auren is of course, a man and not a woman.

Anyway, we are always interested in new studies especially around Social Media - so I followed the links given in the BW article to the study, which sounded massive: “At Rapleaf we conducted a study of 13.2 million people and how they’re using social media”.

The “study” page on their site that is referenced in the article, amounts to an aggregation of data from other sources, none of which is quoted here by name. I suspect they utilized comScore data for most of the gender stuff. We may be wrong, but it looks like the only bit of Rapleaf-specific information here is the number of profiles they’ve aggregated from each of these sources e.g. 11.3 million (of the 13.2 million quoted in the article presumably) are MySpace users.

This doesn’t mean the article in BW is bad - but I do think with all the data bandied about these days, it makes sense to be more clear about one’s sources whereever possible. Though of course, certain companies make up entire businesses around data obfuscation and aggregation so it’s easy to fall into that habit I guess.

Comcast buys Plaxo for $175 million; I have to admit I didn’t see that one coming. Now this is also interesting in the context of ISPs who are outsourcing the delving into their users’ personal data to ISP targeting companies like Adzilla, Phorm and NebuAd. The conventional wisdom here is that lots of small ISPs (and yes there are quite a few of them) will be likely to work with these data players vs. larger ones who might be a bit more circumspect about privacy (though certain ISPs like NetZero, coming out of the ill-fated Free ISP model, find-other-ways-to-make-a-buck have been selling user data for years in various forms).

I have to agree with the article, it’s going to be weird for most Plaxo users, and lots of them will likely go away to bigger and better networks and tools.

Pubmatic’s price index shows that overall eCPM pricing for display inventory is down, according to their figures. This accords with the word around the industry that several advertisers were cutting back in April, fearing recession. It spells opportunity for certain direct ad buyers, I’m sure — if the numbers are still backing out and the sales side is holding somewhat steady — but this is a trend worth watching. We’d love to see other groups putting out similar figures… we’ll share some of our own insights here soon as well.

At the risk of attracting even more comment spam, I audaciously submit the above headline for this blog post. I had to chuckle when I read the latest press release/data bite from my former employer, Nielsen Online (the measurement company formerly known as Nielsen//NetRatings with two /’s) on ClickZ.

There follows a somewhat dubious trail of data that starts with the “Top 25 companies by sponsored link impressions“, quoting winning companies like True.com, Wu-Yi Source, Singlesnet and Anastasia Intl (?). Thereafter it shows you what their top search terms were… and naturally “chat rooms” shows up more than once, and “mail order brides”, “european.single.women.” and “beautiful women” are all well represented. Sheesh. Then some other humdingers like the top search words for Experian, whose free credit report products are a major source of online direct marketing dollars:

“Endurance saddle pads”?

Come on Nielsen, where did you get this data, honestly?

My friend and former LinkedIn colleague Eric Ly has launched Presdo (Techcrunch coverage). We’ve been trying it out for a bit and found it very useful - it’s a web tool that’s all about simplifying the process of finding the best time to get together.  Presdo is targeted at the “casual organizer”, people who may or may not use calendar programs but still need to organize one or more get togethers. It does not require users to be keeping an appointment calendar or to keep every event on their calendar to be able to effectively use the tool. It allows people to express when they prefer to meet on a per-event basis rather than when they are free to meet, especially when certain free times are not good times for the event.

I encourage you to try it out - it’s super simple and you’ll get it right away… we are always arranging meetings with people so we at CPM Advisors love it! Well done Eric, we look forward to seeing how it develops!

We’re soft-launching a site called RetainSales.com today aimed at helping marketers and publishers understand and take advantage of retargeting/remarketing. This term refers simply to a way of finding your visitors again on other websites. It is made possible by the increasing connection between various advertising networks enabled by ad exchanges and targeting intermediaries. We’ve had some good success here with a few clients and have been asked by several more to assist them in creating these types of campaigns, and we figure there are many other people out there who are similarly interested in these powerful marketing mechanisms. The concept of audience extension is another one worth mentioning - a form of simple behavioral targeting where you can sell more advertising inventory outside your own site by finding your users on other sites and networks across the web. Performance of ads on audience extension campaigns is often better than the ads running on the site itself (think about it for a bit….). Check out the new site and let us know what you think. We’ll keep adding to it including case studies and more data.

I ran into Drew Ianni, Chair for North America/Asia Pacific for Ad:Tech at the show yesterday. He told me they had 14,400 pre-registrations for the San Francisco event, which is a record for them. I was at the event and thought it was good - attendance was great, there were a number of interesting companies at the event with significant presences (e.g. Rubicon Project brought pretty much their whole company). Some were surprising (like a new adserver debuting… hmmmm) and we’ll have more to say about the event and some of the companies we met there.

Looks like Google is still ironing out some issues with their AdManager product, which has been down for us today:

We are now beta-testing Google’s AdManager product. So far it is pretty easy to use, somewhat minimalist in its UI which for many adserver veterans will come as a relief. The requirement of having header code separate from the adcode may through some people for a loop - I will be coordinating my thoughts and also want to test out and see how it does on the delivery end before sharing our insights with everyone. Stay tuned…

In advance of next week’s San Francisco Ad:Tech show, PubMatic put out its AdPriceIndex which indicates that average CPMs for approximately 3,000 publishers that it helps with optimization. It naturally shows an upward trend from January to March, from $0.42 to $0.49 or about 15%.

Naturally there is a significant selection bias here in that companies that are happier with their revenue performance and monetization are probably less likely to work with a provider like PubMatic - also, we believe that since many of the revenue figures are self-reported and not directly auditable/trackable to the source, there are likely to be some data quality issues. Note also that the data is not purely for display advertising but also spans text ads, where an improvement from PubMatic’s technology approach to trying different combinations of colors seems more likely.

Our estimates for 2007 indicate the average CPM for 2007 was $1.15 and our analysis suggests an expected 11% rise to $1.29 in 2008,  which more closely tracks the “small sites” segment in the PubMatic data.

I noticed that NexTag is a top 30 ranked site according to Quantcast… the only problem is that when you look at the detail, and the large spike in traffic that NexTag has recently experienced, most of the spike is explained by them buying (and serving) more ads through their adq.nextag.com domain. I should know, I helped put together their adserver a few years ago…

But this does highlight the point that upstart measurement companies face a fairly complex landscape today where there are still lots of ways for publishers to “juice” their numbers and buy traffic. And as an example, filtering out the major adservers is easy, but determining URLs for homegrown solutions like NexTag’s is a much more difficult task to do scaleably. I recall many years ago when I was at Jupiter Media Metrix, the brouhaha that existed concerning X10 - X10 was one of the first proponents of the much-maligned popup ad. They were able to jump to a top 5 position in Media Metrix’s rankings based on millions of “visits” to their website by people seeing popups hosted on their domain. Here’s an article about that archived on their site (X10 is still around - if you want to spy on your neighbors ostensibly - and note the article is from the Industry Standard, in a nice ironic twist) .

The web measurement companies started filtering out things like this based on complaints that it was not representative of user behavior, popup blockers came about and further muddied the picture for companies looking to buy and sell traffic and count how many of these things were popping up not for PR but for dollars-and-sense media counting, and so on. This type of thing still goes on, but it has not been seen on such grand a scale for some time. The moral of the story: having to be right for everyone is difficult: sometimes it’s better just to make money and be right for your investors.

CPM Advisors is pleased to announce we will soon be launching our Publisher survey. Aimed at helping Publishers figure out where they stand and also understanding challenges and issues facing publishers, the Publisher survey will provide valuable information to the marketplace about what’s going on “in the trenches” and how the industry can do more to service the needs of publishers.

It’s taken a long time but Google finally launched its free ad serving product, Ad Manager. Scott Switzer has a good post about it that is a variant of “do I hand over ALL of the keys to my monetization” to Google argument that will continue to rage around the online ad space for some time to come.

We’ll have a lot more to say on this topic in the near future - we’ve been quiet here at CPM Advisors as we’ve been ramping up ad management efforts on behalf of some of our publisher partners. I have long been in favor of a more open adserving approach, and have my doubts that Google will be the one to be the bringer of openness to this market. They can and will certainly bring a dose of “free”, as expected.

Total    - 63,411,040 -    100.0%
USA    - 24,741,720 -    39.0%
UK    - 8,905,320 -    14.0%
Canada    - 8,869,280 -    14.0%
Turkey    - 3,024,880 -    4.8%
Australia    - 2,430,840 -    3.8%
France    - 1,251,260 -    2.0%
Colombia    - 1,130,400 -    1.8%
Norway    - 1,075,100 -    1.7%
Sweden    - 1,059,300 -    1.7%
South Africa    - 739,340 -    1.2%
Mexico    - 688,300 -    1.1%
Hong Kong    - 639,360 -    1.0%
Egypt    - 616,760 -    1.0%
India    - 528,620 -    0.8%
Denmark    - 465,740 -    0.7%
Germany    - 451,860 -    0.7%
Finland    - 435,480 -    0.7%
Israel    - 428,920 -    0.7%
Singapore    - 396,460 -    0.6%
Spain    - 365,880 -    0.6%
New Zealand    - 319,020 -    0.5%
United Arab Emirates    - 298,460 -    0.5%
Malaysia    - 292,720 -    0.5%
Belgium    - 279,000 -    0.4%
Lebanon    - 256,980 -    0.4%
Italy    - 246,480 -    0.4%
Switzerland    - 234,080 -    0.4%
Greece    - 215,560 -    0.3%
Ireland    - 215,320 -    0.3%
Saudi Arabia    - 207,180 -    0.3%
Venezuela    - 194,740 -    0.3%
Pakistan    - 182,140 -    0.3%
Netherlands    - 172,440 -    0.3%
Jordan    - 156,900 -    0.2%
China    - 154,040 -    0.2%
Argentina    - 147,600 -    0.2%
Chile    - 130,800 -    0.2%
Japan    - 128,520 -    0.2%
Croatia    - 123,640 -    0.2%
Panama    - 116,720 -    0.2%
Indonesia    - 112,520 -    0.2%
Thailand    - 109,140 -    0.2%
Philippines    - 105,020 -    0.2%
Brazil    - 98,540 -    0.2%
Serbia    - 83,220 -    0.1%
Austria    - 78,800 -    0.1%
Peru    - 74,200 -    0.1%
Korea, Republic of    - 68,960 -    0.1%
Kuwait    - 68,620 -    0.1%
Russian Federation    - 66,680 -    0.1%
Poland    - 61,980 -    0.1%
Taiwan    - 57,280 -    0.1%
Dominican Republic    - 56,680 -    0.1%
Nigeria    - 52,240 -    0.1%

Source: Facebook

Using data made available by Adonomics (aggregating app install and active user figures available on Facebook), we did a comparison between app install and active user numbers from January 30, 2008 and February 23, 2008 — over this time period for the top 1589 applications from January 30 (all those with over 22,250 installed users as of that date) there was a 9.6% increase in number of installed users and a 16% drop in total daily active users. We are not fully confident in the integrity of all the figures since these numbers are gathered from Facebook by Adonomics and there do appear to be a few gaps in their reporting. Nonetheless it appears, on the whole to be a fairly good reflection of what is going on.

If the figures are correct, in three weeks the top 5 applications added 10.8 million installs but lost 2.19 million daily active users. Excel file with full data as aggregated is here: Application install barometer

We’ve been tracking a variety of statistics using Facebook’s active user figures as disclosed as part of their Social Ads system. As expected, there has been a good deal of growth in their older users, as is reflected in the chart below. The average age of their US user base is still very young, however, at 22.96 up from 22.28 when we first looked at the data on 11/24/2007.

Users of age 25 and older have increased by 31.7%… but the most striking thing is how far out of alignment comScore’s Media Metrix audience measurement figures are compared to these figures. For January 2008, comScore claimed that Facebook’s 35+ audience was 14.687 million, with about 5.36 million in the 50+ age bracket. I used to work at Nielsen NetRatings and we would sometimes see a 30-40% delta between panel numbers and publishers’ web analytics some of which could be attributed to cookie issues etc etc., BUT THIS is a 10x factor difference (that’s 1000%!) , which is a bit hard to simply dismiss.

According to comScore/Starcom/Tacoda, “heavy clickers represent just 6% of the online population yet account for 50% of all display ad clicks” and “heavy clickers skew towards Internet users between the ages of 25-44 and households with an income under $40,000.” It continues:

Heavy clickers behave very differently online than the typical Internet user, and while they spend four times more time online than non-clickers, their spending does not proportionately reflect this very heavy Internet usage. Heavy clickers are also relatively more likely to visit auctions, gambling, and career services sites – a markedly different surfing pattern than non-clickers.

Reading the slashdot comment thread, I love comments like this from sootman (scary as this thought is):

NOOOOOOO!!!! We NEED these people! The WWW is supported by ads, so as long as these people do all the clicking, no one will mind if the rest of use use AdBlock, custom /etc/hosts files, etc.

It just reinforces what we have been thinking about and working on for some time - a shift to per-user metrics, things like ARPU (or let’s be clear, ARPMU per thousand users!) for measuring advertising returns and not based on per-impression or thousand impressions. I have sent in a few requests to the companies that put the study together to get more details on the methodology. If this study was done the way I think it was done covering all ad campaigns, then I think the news may not be quite as bad for everyone as it seems at first glance. But I will post an update here shortly.

UPDATE:  Indeed I received the presentation from George - it looks like the study (Presentation for Conference - Heavy Clickers) is across all ads, of which clickthrough rates are much higher for direct offers that promote things like earning money from surveys, sweepstakes and other offers that would typically attract people with more time on their hands who are looking to make a few bucks online. This is really not that surprising to me. Although there are vast systemic problems with the online ad world and a lack of working “plumbing” infrastructure, I’m not ready based on this study to sound the death knell for online advertising just yet.

Esther Dyson’s column in the WSJ entitled “The Coming Ad Revolution” got my attention at first mainly because it mentioned the ISP targeting companies “NebuAd, Project Rialto, Phorm, Frontporch and Adzilla” about which not too much (yet) has been written in the mainstream press. As they work to try to sign up ISPs, these firms still face many of the same problems that have bedeviled “behavioral targeting” companies - related to the size of the user footprint they can reach, and how often they will see the user return. By working with an ISP versus an individual site, they greatly reduce the frequency problem and also the issues caused by users unhelpfully deleting their cookies (or having a spyware program do so) every few weeks. On the other hand they still have to crunch a lot of data very quickly to be able to push relevant messages out to users.

Yahoo! should have been the king of behavioral user targeting with all the visitation and search data they have about users, visitation patterns across multiple properties and so on but they by most accounts have failed to come up with much that is truly compelling in this area. Partly a problem of their antiquated adserving technology, perhaps? Several talented Yahoo!’s from this area have left the company e.g. Richard Frankel who recently left to become COO of SocialMedia.

But back to the Dyson article - I disagree with:

What might seem like a horribly complex and tedious task to their elders — categorizing “friends,” managing news feeds, handling intersecting communities of contacts — feels natural to the Facebook users of today. They want more granularity of control, not less.

People don’t use a network like Facebook in the same way nor do they have equivalent expectations about privacy (some thought Beacon was great, many thought it wasn’t). A web product has to have enough common features and bits so that you can deliver information to the user within some known bounds, but also provide enough buttons and dials to be able to configure it for an individual in such a way that it meets their needs. The configurability of a product and the instant appeal, ease of use and “gettability” of it are often at odds. It is also a bit more complex when you think about adding tweaks to an existing product where there is already a history of use, the needs and reactions of the user base to past changes are measurable and somewhat predictable versus designing for new users you hope will flock to your platform. Stepping outside of the average Facebook user of today (age 22 in the US) to the more mainstream demographics Facebook’s valuation is based on (reaching eventually via some type of advertising), we may find a much lower tolerance for having to very granularly tweak the product to get it to work properly.

In general, people don’t change default settings. That’s why things like HDTV LCD TVs typically come with pretty good standard factory settings. Sure the superuser is going to tweak stuff, but as Microsoft told me several years ago when I was researching browser defaults “95% of people don’t change the default settings”. People do tweak things more than they used to, but I believe the complexity of setting up all kinds of rules to coordinate products, advertising and events between me and my friends will prove cumbersome and be enough of a hassle that it won’t happen for quite some time. The “preference plumbing” that will enable these decisions to be made more quickly and easily is hard and will take time. Ironically, Facebook’s sugardaddy Microsoft has long been beating the drum on “[enabling] applications and services to cooperate on consumers’ behalf” but I don’t think we’re that much further along than when they were talking about Hailstorm in 2001.

There is a lot of interest in user-based targeting right now, a good deal of money being thrown behind online advertising companies and applications, but both in terms of infrastructure and coordination, we’re still a long way from an Ad Revolution.

This is a long post. But it ends with some interesting data. I recently read Ben Edelman’s piece about the Sears-comScore debacle, where Sears (”Sears Holdings Community”) was partnering with comScore to track users’ online activities, encouraging an install in a way that many users might find to be deceptive. Ben is a well-known industry researcher focusing on adware/spyware and his overall take from this piece about comScore is that:

There’s no good reason why users should share information about their browsing, purchasing, and other online activities. So time and time again, ComScore and its partners resort to trickery (or worse) to get their software onto users’ PCs.

As a former market research analyst myself, since comScore’s data gets used quite a bit by online firms, venture capitalists, bankers, etc. I am interested in figuring out what types of biases the recruitment methods of a comScore might introduce into the data. We recently pointed out large discrepancies between their figures for older users of Facebook versus what the company itself is claiming about its US audience, for example.

I know from Ben’s site that one of the recruiting sites that comScore uses is RelevantKnowledge.com - on the website they inform the user that they are gathering information and using it to create reports that get quoted in various publications, they even go so far as to show examples (partial screenshots from 2004 data) of the kinds of reports that are created. comScore’s name doesn’t appear on the reports or anywhere on the site. The terms of use /privacy policy reference that the site is owned by TMRG, Inc. with an address of 11465 Sunset Hills Road, Reston VA which is comScore’s corporate address.

This is where it gets a bit more interesting. We looked at the data for Relevant Knowledge on Quantcast.com - and according to their figures this site had 397, 653 unique visitors in December 2007, with a large bias towards women. Helpfully, the affinities of the users also skew towards another TMRG site called OpinionSquare.com. It has a fairly large user base as well of 152,023 users in December 2007 (Quantcast).

On their “About” page, OpinionSquare disingenuously states:

The information that you contribute is used by comScore, Inc., a U.S.-based market research company that is a nationally-recognized authority on Internet and general economic trends, whose data are routinely cited by major media outlets such as the New York Times, the Wall Street Journal, and CNBC, and is extensively used by the largest Internet services companies and scores of Fortune 500 companies.

OpinionSquare.com is owned by TMRG, Inc. according to the website. TMRG, Inc.’s privacy policy references 11465 Sunset Hills Road, Reston VA which is of course comScore’s corporate address. By the way, OpinionSquare also biases heavily towards women (indexing at almost 150 according to Quantcast.com) and older users with age 55-64 being almost 25% higher than the web average for this demographic.

Once I went to www.tmrginc.com, TMRG’s “corporate site”, I found these and the other two sites they use and disclose which are Permission Research and MobileXpression.

Buried in the TMRG website’s FAQ is a mention of the connection with comScore, of course if someone might be a potential customer:

My company is interested in using TMRG to conduct market research. Who should I contact? TMRG is a service of comScore, Inc., a leading Internet ratings system that provides insight into consumer behavior and attitudes. For assistance with your market research needs, please complete comScore’s Information Request form [link removed] and someone will contact you shortly.

I wondered how these sites would show up in comScore’s own data because presumably comScore’s $50,000 a year (and up) service is going to be much better quality that some of the free resources out there. The answer is that they don’t show up at all.

Hmmm does that mean these users are systematically weighted out of the panel? They were generally older and more female (the kinds of people who take surveys, respond to ads run on gaming sites, fall victim to browser exploits that install software perhaps?). And if that is the case then what does it mean to how other people get weighted? Or more insidiously are these sites meant to be ranked but are just removed from the comScore rankings to avoid any perceived taint? Surely no matter how bad the data is that the other providers could be putting out there, you’re not telling me that none of these sites are going to show up in the top 20,000 web domains that comScore publishes, where the minimum site had just 11,000 combined work, home, university users?

Did I miss a buried entry in an FAQ somewhere, comScore?