Harris Data on the ‘What’ & ‘Why’ of Social Media

Earlier this week pollster Harris published some interesting data from a consumer survey conducted in April (n=2,131) about social media attitudes and motivations:

[N]early two-thirds (64%) of online Americans use social media, and most social media users (84%) reveal information about themselves via social media channels.

Among the findings is also a discussion of the influence of reviews and their various sources (click to enlarge graphics).

While 84% of social media users are providing tweets and updates, more than 20% are discussing brands and products or generating reviews. The following charts show the relative influence of reviews by content source and by age group:

The hierarchy of review influence is:

1. Friends & family
2. Reviews from newspapers or magazines (experts/professional/editorial)
3. One’s broader network
4. Blogs/message boards
5. Celebrity “reviews” (most influential on the 18-34 yr old group)

Advertisement

MyYahoo an Unleveraged Product

I keep thinking about My Yahoo! and ways that Yahoo! might develop it into something really useful and compelling. The day of the “RSS reader” has come and gone. Though many people use Google Reader or iGoogle or Netvibes these products have failed to break through into the mainstream by and large. My Yahoo! was/is the leader in the category.

When Yahoo! updated its homepage to make it capable of customization it made My Yahoo! less relevant, although the latter has many more features. My Yahoo! is in need of an update and could become a very useful and strategic product for the company — with a few tweaks and a redesign.

Why not make it a personal dashboard where users can see and update social network news feeds, read news, take notes, send email, conduct local searches, save websites and so on. It can do many if not most of those things today but the UI is not very friendly and it’s not simple enough to accomplish these tasks.

Internet users are often seeking to accomplish concrete tasks and My Yahoo! could be a kind of personal assistant in the process; this is how people generally use search today but search is an incomplete tool.

Mike Arrington of TechCrunch asks “But Seriously, What Is Yahoo?” coming off his expletive-tinged interview with Yahoo! CEO Carol Bartz last week. In his article, Arrington contrasts AOL CEO Tim Armstrong’s one line answer about AOL’s new identity vs. Bartz’s laundry list of features and capabilities.

As embattled as it is Yahoo! still has a trusted brand (among mainstream Internet users) and it presents a central place where people can access lots of online content, tools and information.

My Yahoo! could quite easily become a kind of “personal dashboard for the Internet.” But it would need to be pre-configured, simplified and redesigned to do so.

Google Algo Changes Drop ‘The Other Shoe’

Google’s first big shoe drop was the increasingly common appearance of a map + 7 in response to explicit and implicit local intent queries. That took away many of the “above the fold” SEO slots available to local sites such as yellow pages publishers. However, some of that real estate has been restored in the new Google interface:

Now the “other shoe” may have dropped as algorithm changes that may penalize undistinguished local directory sites start to have an impact. Monsieur Andrew Shotland explains:

Thanks to Vanessa Fox as always for eliciting grains of truth from Google.  At last week’s Google I/O conference Matt Cutts confirmed that there was an algo change at the beginning of May that will likely be affecting sites that generate a lot of long-tail traffic:

”this is an algorithmic change in Google, looking for higher quality sites to surface for long tail queries. It went through vigorous testing and isn’t going to be rolled back.”

I have already seen the impact on a couple of large sites that I monitor, as well as on smaller sites.  In the case of the large sites, the traffic has drifted downward, with a couple of extreme drops, and in the case of the smaller sites traffic growth has either flattened or slowed down to barely noticeable.

This shouldn’t surprise anybody in SEO land as unique content, good site architecture and strong backlinking have been the cornerstones of SEO for years.  That said, it appears that this algo change has raised the game to a whole new level.  It used to be that an authoritative domain could add pretty much any content, even duplicated content, it wanted to its site and with a little bit of effort get strong organic traffic. That does not appear to be such an open and shut case anymore.

Andrew already sees this impacting YP publishers and others in the local space that have historically relied on SEO for traffic.

I also had conversation with Matt Cutts at I/O about this same general phenomenon, except I was thinking about and focused on Demand Media, Associated Content and the like. I wasn’t thinking about YP publishers.

If what Andrew is saying is playing out — and he’s observing it apparently — then local ad networks like CityGrid become even more important, as well as improved user experiences (for direct traffic) and brand building.

Social media and mobile also become even more strategic for branding and traffic (although in the case of mobile it’s all incremental right now).

Google Ranks the Web’s Top Sites

Barry Schwartz at SEL posts on a new Google/DoubleClick AdPlanner ranking of the Internet’s top 1000 sites. This list is global, and the data come from installed Google toolbars and other sources.

Here are the top 25 sites according to the list:

Google has removed its own sites from the list, so no data on Google Maps traffic: %$#@!

The world’s top site is . . . Facebook with 540 million unique users. Yahoo is second and Live.com third (curiously). Bing is 13. Twitter comes in at 18, with 96 million uniques. Craigslist is 49th with 14 billion (with a “b”) page views.

Unfortunately there’s no ability to search the list, you have to page through it.

PaperG Bringing Automated Display to SMBs

AdReady couldn’t do it. Google and Yahoo! haven’t been able to do it. And Facebook hasn’t done it either. I’m referring to dramatically simplifying the process of display ad creation to reach the small business market. However, PaperG’s PlaceLocal it appears has achieved that lofty goal.

PlaceLocal was the subject of a profile in the NY Times on Friday:

New software called PlaceLocal builds display ads automatically, scouring the Internet for references to a neighborhood restaurant, a grocery store or another local business. Then it combines the photographs it finds with reviews, customer comments and other text into a customized online ad for the business . . .

Then PlaceLocal takes over, gathering basics like telephone number, hours of business, maps and directions, and adding positive comments extracted from local blogs. Samples of ads may be seen at www.paperg.com, the PaperG Web site.

PaperG’s program is up and running on 32 local media Web sites, including Time Out New York and Time Out Chicago, and on 29 network TV affiliates owned or managed by Hearst Television, said PaperG’s chief operating officer, Roger Lee. The company has also signed up the McClatchy newspaper chain and will soon be on some of its Web sites, he said.

PlaceLocal/PaperG is undoubtedly part of the mix in the Gannett Local offering as well.

And because CPM inventory is cheaper than paid search I would expect more local online marketing platforms to incorporate this with or without the knowledge of local businesses. It could for example be a part of a traffic arbitrage strategy or sold directly to SMBs as a product or part of a product bundle.

____

Related: PaperG Challenges AdReady, Expands Reach

Facebook Beats Yahoo in Q1 Display

A significant milestone, reported by the WSJ:

In the first quarter, Facebook pulled ahead of Yahoo for the first time and delivered more banner ads to its U.S. users than any other Web publisher, according to market-research firm comScore Inc.

Overall, Facebook.com served 176.3 billion display ads on its website over the first three months of 2010, or 16.2% of the total, said comScore. Yahoo served 131.6 billion banner ads to Yahoo users, and Microsoft served 60.2 billion, according to comScore. The data don’t include ads that Yahoo and Microsoft delivered to other Web sites through their networks, a major source of revenue for each.

By revenue, Facebook has a long way to go to catch up to its more established rivals. The social-networking site earned more than $500 million in revenue in 2009 and is forecasting revenue of more than $1 billion in 2010, according to people familiar with the matter. Yahoo earned $6.5 billion in revenue in 2009, mostly from advertising.

Obviously there’s enormous potential with SMBs on Facebook as well. It’s another possible distribution point for local ad networks too. Just take PaperG’s automated display ad creation tool for SMBs and distribute on Facebook — voila.

When location kicks in we should see some really interesting new ads/campaigns on Facebook. Along those lines . . . the company, I predict, will also become a mobile ad network (or its equivalent) in the not-too-distant future.

Are Content Farms Creating ‘Digital Serfs’?

AdAdge recently ran a piece discussing the move toward content outsourcing in traditional media, from “in-house” journalists and editors to third party “content farms” such as Demand Media and Associated Content:

Hachette is using Associated to supply some content for its Woman’s Day site. Thomson Reuters experimented with Associated for a limited period of time last year but plans to ink similar deals in the future, whether with Associated or another content provider. And Cox’s Atlanta Journal-Constitution published a handful of articles from Associated last summer; it is now running regular articles supplied by Demand Media.

Such partnerships further the ongoing shift among established news operations to capitalize on the availability of cheap content, such as USA Today’s recent deal with Demand Media, which is using its network of freelancers to supply pieces for a new Travel Tips section on USA Today’s website.

In the past I’ve criticized Demand and Associated Content as being fountains of crap content, much like the oil spill in the Gulf is spewing junk into the ocean. But that’s unfair; many of the writers that work for (freelance for) these organizations are former journalists or people with expertise in the areas they’re writing about.

At least some of the content being generated then will be of reasonable and maybe even high quality. The issue as the AdAge piece points out is cost; these sources are just much cheaper than headcount:

Associated pays its contributors anywhere from $5 to $30 per article, sometimes upfront — and in some cases pays a performance fee of up to $2 for every 1,000 impressions the story generates within Associated Content’s site.

The margins on these articles are quite large; but the publisher pays no health insurance or other overhead costs — Demand says it provides “access” to health insurance — so the bottom line on outsourcing is highly favorable vs. hiring staff writers and editors.

Maybe someone will come forward and claim to be making “six figures” from writing for one of these companies but I’m skeptical. What we’re seeing develop, accordingly, is a class of underpaid writer-content creators or “digital serfs” who are largely anonymous and fungible. Some people may, of course, be using content farms as a kind of SEO strategy to promote themselves or their businesses. But those savvy folks are likely in the minority.

Branded publishers know that they can’t maintain their reputations by relying primarily on this stuff but they can back-fill or round out their content online with help from Demand or Associated Content.

Perhaps these companies are simply filling a void and need created by the disruptive influence of the Internet, which won’t support ad revenues comparable to print and thus the staffing. Perhaps these are the new economics of content creation online. But the creation of a class of digital serfs is a lamentable outcome in my opinion.

More on Facebook, Privacy & Data Mining

Here are two unrelated pieces on Facebook that I ran across nearly simultaneously:

• An Online Alias Keeps Colleges Off Their Trail
• Facebook VP: Why We Want Marketers to Like “Like”

The first is from the NY Times, about how high-school students and college applicants are trying to make it harder for colleges to find them on Facebook (using aliases), for fear of the adverse consequences of institutions knowing too much about them:

For high school students concerned with college acceptance, Facebook presents a challenge. It encourages making public every thought and every photo, an opportunity for posturing and bravado nearly irresistible to teenagers. But this impulse for display clashes with the need to appear circumspect and presentable to college admissions agents, who some high school guidance counselors have warned are likely vetting applicants by trolling the Web.

Whether admissions officers really do plumb Facebook is up for debate, said Dr. Frank C. Leana, a prominent independent college counselor in New York City whose services cost $1,000 (for a one-time consultation) to $9,000 (for ongoing counsel throughout the college process). His students believe they are being watched, he said, but “it’s really hard to know how accurate their suspicions are.”

One of the big pitches for the Open Graph is “making the Web less anonymous,” more transparent. But Facebook being the de facto identity management platform across the Internet is not so desirable to everyone, as the above article suggests. And young people do care about privacy it would appear — or at least that third parties not be able to access their information.

Real and open identity is good as an abstract matter until the “big brother effect” kicks in and others are passing judgment and denying jobs or college admissions, potentially because of one-too-many drunken party images on Facebook. While there’s no documentation of the latter, the fear is clearly present.

The second article, an interview of sales VP Mike Murphy by eMarketer, reveals some of Facebook’s ambitions surrounding data mining and ad targeting vis-a-vis the “Like” button. Here’s eMarketer’s summary conclusion of Murphy’s comments:

Whenever a person clicks to “like” something they see on the Web, that information will go into their Facebook profile and marketers will be able to use the information to target advertising within Facebook.

This is a key point: all off-site “Like” activity will factor into targeting on Facebook. In other words, Facebook is mining actions across the Internet for targeting on its site. Brilliant and/or creepy? A little of both I think.

What many (most) people don’t realize also is that “Liking” something across the Internet will enable marketers or publishers to push content and “publish into [users’] news feeds.” This requires prior approval but it may not be clear that people are signing up for an ongoing stream of information or marketing messages.

Promoted Tweets: ‘Pull’ and ‘Push’

Twitter’s business model has emerged. Ironically, yesterday, Bill Gross of IdeaLab (and founder of Overture) announced TweetUp, a keyword-based search marketplace built on a proposed better search engine for Twitter and a group of syndication partners. Last night Twitter announced its own, very similar idea with Promoted Tweets:

Q: What are you launching? What are Promoted Tweets?
A: We are launching the first phase of our Promoted Tweets platform with a handful of innovative advertising partners that include Best Buy, Bravo, Red Bull, Sony Pictures, Starbucks, and Virgin America — with more to come. Promoted Tweets are ordinary Tweets that businesses and organizations want to highlight to a wider group of users.

Q. What will users see?
A. You will start to see Tweets promoted by our partner advertisers called out at the top of some Twitter.com search results pages. We strongly believe that Promoted Tweets should be useful to you. We’ll attempt to measure whether the Tweets resonate with users and stop showing Promoted Tweets that don’t resonate. Promoted Tweets will be clearly labeled as “promoted” when an advertiser is paying, but in every other respect they will first exist as regular Tweets and will be organically sent to the timelines of those who follow a brand. Promoted Tweets will also retain all the functionality of a regular Tweet including replying, Retweeting, and favoriting. Only one Promoted Tweet will be displayed on the search results page.

Image source: AdAge

The NY Times echoes how this will work:

When a Twitter user searches for a word an advertiser bought, the promoted message will show up at the top of the results, even if it was written much earlier. The posts say they are promoted by the company in small type, and when someone rolls over a promoted post with a cursor, it turns yellow.

This is paid-search advertising around keywords (though pricing is apparently CPM to start). Ads will also appear in third-party clients and syndicated streams.

This is all very familiar, well established, no big deal. That’s the “pull” dimension. But there’s another “push” (AdSense-like) dimension to all this (per the NY Times):

In the next phase of Twitter’s revenue plan, it will show promoted posts in a user’s Twitter stream, even if a user did not perform a search and does not follow the advertiser.

For example, if someone has been following people who write about travel, they could see a promoted post from Virgin America on holiday fare discounts.

Anyone who uses Google has grown accustomed to seeing ads alongside their search results, but Twitter users could resent seeing promoted posts in their personal content stream.

Twitter is aware of that risk. It is still figuring out how to determine which promoted posts should appear. It could be based on topics they are writing about, geographic location or shared interests of people they follow.

This second, “involuntary” dimension of the program will apparently roll out later and very carefully. Though all the ads are substantially text-based, what you’ve got in the two components is direct response and awareness ads.

Predictably there’s positive and negative reaction to the second part, which is premature. We can’t begrudge Twitter a way to make a living.

______

It’s worth revisiting the Twitter “business model contest” that happened a little over a year ago. Search and contextual advertising are of course among the many suggestions in the submissions.

AT&T Formally Announces YP Brand Change

This morning AT&T put out a press release formally announcing YP.com as the flagship brand for its yellow pages site:

AT&T Interactive today introduced YP.COM, the new YELLOWPAGES.COM, as its evolved flagship web property and brand. Powered by its proprietary local search technology, YP.COM draws from AT&T’s history and success in local search to provide innovative solutions that connect consumers with businesses. . .

From the start of the search experience, YP.COM promotes discovery through predictive text search technology that automatically suggests potential search terms as a user enters a business name or category. Search results, supported by YP.COM’s database of over 21M business listings[i], allow users to filter results by neighborhood, Video Profiles and business coupons. Giving users more reasons to explore, YP.COM leads to more consumer engagement, which means advertisers have a better opportunity to connect with engaged consumers who are ready to make a decision about where to dine, catch a movie or stay while away from home. While in beta testing, advertisers averaged a 28 percent increase in calls and clicks per search[ii] as a result of placement on YP.COM.

The site expands beyond traditional YP to include events and other information. I wrote about the change yesterday when I stumbled upon it.

I agree with the brand change because it makes the site much more flexible for the inclusion of additional types local information and content.

Chatmeter Launches SMB Reputation Tools

Today a new SMB-facing reputation management and SEO tool, called Chatmeter, entered the market. It joins Marchex and AmIVisible in the effort to help SMBs (or those that work with them) discover problems with their data and manage their content and online presence. The company was founded last year by former V-Enable executive Colin Holmes.

Localeze is also in this category but with a very different service.

Holmes will offer a free and a premium version directly to the market but also will be seeking to enlist sales channel partners that have direct hooks into the small business market.

The following is a screen from a client/customer dashboard. It shows inconsistency in the listings data, identifies where SMBs profiles are incomplete and tries to put all that in the context of competitive sites and how they compare across categories. The system assigns numerical scores to social media mentions, reviews and references.

I didn’t see a demo or have a chance to actively use the system, but I did get a lengthy briefing from Holmes last week.

Based on my review of the material and discussion with Holmes it looks like there’s valuable information and useful tools here. The question is how to make the information accessible to SMBs in a simple way, as well as make it more and more actionable.

YPG Launches New Image, Campaign

Oh to be a “branding consultant.” Would that I were, I’d be making considerably more money . . . Anyway, Canadian directory publisher Yellow Pages Group has introduced a new logo and a new marketing campaign to update the company’s image:

With a new logo and a new national advertising campaign,Yellow Pages Group (YPG) today launched a new brand identity reflecting its evolution into a performance media and marketing solutions company.

Here’s the new logo; what do you think? Does it say “multi-platform”?

The new homepage also integrates content from the recently acquired RedFlagDeals site.

ChaCha Search Data Provide Brand Insights

ChaCha has done a number of interesting things lately to expand its audience and utility, such as Facebook integration. The company is now seeking to offer marketers access to its data around brand affinity and purchase intent. ChaCha offers one of the most direct marketing channels to the youth market in the US and has fielded millions of questions about products and brands by teens and young adults (13-24).

It recently commissioned an analysis of its data and queries to expose trends and preferences among its users. Here are some charts from a report released in conjunction with this analysis:

There’s more data and discussion in the report. The idea here, however, is that ChaCha has a massive database of queries (like Google) that show trends, etc. The difference is that ChaCha’s audience is very demographically specific.

CEO Scott Jones also told me previously that the company has great success with polls, which offers something like a massive youth focus group for brands.

ComScore: AOL the Top Ad Network

AOL has long owned the greatest ad network reach online. But people forget that because they treat AOL as though it’s “over.” But the recent comScore report (which is essentially monthly) on the top US ad networks reminds us that AOL is still formidable in the display ad segment:

These networks are where 99% of the roughly $7.5B in annual online display ad revenue is concentrated. There may also be other revenues in here, not formally counted by the IAB in the display category.

Top Web Brands Today . . . Tomorrow?

Here’s Nielsen’s list of top Web brands in the US market for this past December. This is about reach:

Will this list be very different a year from now? Will Facebook be #1 or #2? Will Twitter be on the list; will Fox be off? What about AOL, will that have fallen farther?

Local Display Ads: Where Are They Going?

The Kelsey Group’s latest release projects an increase in geotargeted display advertising:

The geotargeted display (or banner) advertising market will grow from $897 million in 2008 to $1.9 billion in 2013, representing a compound annual growth rate of 16 percent, according to BIA/Kelsey. The geotargeted segment of the display market will grow from 10.2 percent of all display ads sold in 2008 to 15 percent by 2013.

The locally bought portion of the market, which primarily comprises small and medium-sized businesses, will see the highest growth, at a CAGR of 66 percent. The segment will grow from $45 million in 2008 to $565 million by 2013. Further, the SMB market will swell from 5 percent to 30 percent of the total geotargeted market over the same period.

Most small advertisers are not using display advertising today. Three different surveys (with different sample sizes) I’ve conducted in the past year and a half have shown between 13% and 22% of SMBs saying they have done online display advertising of some sort (definitions are an issue). That means, on average, roughly 80%+ have not done any display advertising. There’s a much longer conversation to be had about how SMBs execute display campaigns directly or via surrogates.

My belief is that national entities will still be the primary ones taking advantage of increasingly accurate display geotargeting online and on mobile devices in the foreseeable future.

The 10.2% starting point in the Kelsey forecast likely derives from a comScore study, which was based on a creative but pretty rough methodology for determining which display ads were geotargeted in four markets. It probably under-counted actual geotargeted display inventory.

The stated rationale behind geotargeted display growth in the Kelsey release — that local search resellers are looking for cheaper local impressions/clicks — is sound. But in my view it’s too narrow in terms of what may actually drive geotargeted display over the long term.

There’s also the problem of the ad creative. While that can be automated it has to be done carefully and thoughtfully or those ads simply won’t perform. Creative is the most important part of a display campaign and most of the local search resellers are not thinking about creative; rather they’re just thinking about alternative sources of cheap local inventory.

Compare Borrell’s numbers (definitions are always an issue in forecast comparisons), which have locally targeted “banners” at $5.3 billion in 2010. In addition Borrell sees standard display giving way to myriad other forms of local advertising, including video, email and search over the next few years.

Get Ready for ‘Extreme Local’ Marketing

As one of the very early users of the term “hyper-local” I’m here to say it’s “used up;” it’s been drained of meaning by overuse. How about a new and potentially more irritating term: “nano-local”? Really just kidding about that. 

But get ready for “extreme local.” EXTREEEEEME LOCAL! In a world where nothing is as it seems . . . and nothing will ever be the same . . . 

The NY Times profiles multi-platform local marketing and advertising campaigns being run in NYC by AT&T and AMEX:

How extremely local are the promotions being advertised? They last 24 days; cover 15 neighborhoods in Brooklyn and Manhattan; involve two sponsors, American Express and AT&T; and include more than 650 stores, shops, restaurants, cultural attractions and other places where residents of and visitors to New York City can spend time — and money . . .

The campaign includes print and outdoor advertising, ads online, social media like Facebook and Twitter and content on a microsite, or special Web site (nycgo.com) . . . 

American Express is devoting more time and attention to localized campaigns, some of which can be found on a Web site (amexnetwork.com/travel). There are deals for cardholders in markets like Chicago, Los Angeles and San Francisco, offering discounts on hotel rooms, restaurants, entertainment and other attractions. For AT&T, the partnership with NYC & Company represents its first such locally focused campaign, said Chris Schembri, vice president for media services at AT&T in Dallas.

This campaign illustrates the way brands may be starting to “get” the value and importance of local.

YouTube Pre-Roll and Cost-Per Engagement

The other day I was surprised to see a Geico car insurance commercial (couldn’t find today) as pre-roll on YouTube. But now I understand it’s part of a test program that allows users to skip the ads to see which ones are working. In most pre-roll video ads you cannot bypass the commercial (think Hulu) and are compelled to sit through the entire thing. But Google is testing a new approach that may yield a video “cost per engagement” model. MediaPost explains:

Google will begin testing “skippable” pre-roll ads in videos on YouTube Wednesday that could lead the Mountain View, Calif. search engine toward a new advertising model. The small sampling, which runs indefinitely, will allow people who find the videos to click on the link and skip the ad, which takes them directly to the content. The ads will run on videos from content partners, which have already opted into the test.

The goal to move the industry toward more engaging high-quality ads requires a lesson in human behavior. The test that determines if and when people watch the video clips will provide Google with insight into the type of person who may skip an ad, what type of ad they might skip, and what piece of content does better than another. Google also will look at whether some ads are skipped in a specific portion of the session. Does the person skip the ad in the first video versus the third during a 30-minute time slot while on YouTube? . . .

[Product manager Phil] Farhi says Google eventually sees a model where the advertiser only pays for a completed view of the ad. Quality and user signals would determine the correct place in the video to serve up the ad.

The idea of letting users freely skip pre-roll is fairly radical. Most users won’t watch most pre-roll video ads. Most TV advertising is built on the premise that people must be compelled to watch ads, which partly gave rise to TiVo and the DVR. But if the ads are interesting enough people will watch them and even virally promote them (online). 

This move coincides with one toward “addressable advertising” at Google, having just added Nielsen (former Claritas) “Prizm” segmentation to its TV Ads offering: 

This partnership brings more layers of audience data to our existing targeting tools. In addition to demographic and interest dimensions from Nielsen and Equifax already available in our targeting tools, advertisers will now be able to layer on PRIZM segments to find their audience on TV through Google TV Ads. PRIZM categorizes US households into 66 unique segments, using attributes such as lifestage, income, social group, home ownership, employment, and education . . .

How does it work? Advertisers simply select a PRIZM cluster in the Google TV Ads interface, and receive recommendations on which TV networks, dayparts, and programs are popular with that cluster. Advertisers can then add the media they’d like to target and air ads within 24 hours of building a campaign.

These movements portend not only the “convergence” between TV and online video advertising (which parallels audience user behavior) but a dramatic shift in how “TV advertising” is bought, measured and billed.

Is the Yahoo! Campaign Working?

I’ve been seeing the “It’s Y!ou” Yahoo! brand campaign everywhere, in the real world and online. Certainly it will help raise awareness of Yahoo! but do you think it’s doing anything concrete for the company?

I’m curious . . .What do you think?

comScore: 10% of Display Local/Geo

From comScore this morning:

The study found that between 9 and 11 percent of display ads in the four markets among all publisher sites were locally targeted. San Francisco (11 percent) and Washington D.C. (11 percent) had a relatively higher share of ads being locally targeted, while Atlanta (10 percent) and Chicago (9 percent) were slightly lower. Not surprisingly, in the regional/local site category – which includes sites like Yahoo! Local, Citysearch and Yelp – the share of display ads that were locally targeted was substantially higher at between 23-33 percent among the four markets.

And now for some crude math:

Using the IAB numbers . . . if online display advertising represents 22% of online ad revenues ($23 billion) that’s $5.06 billion. Ten percent of that would be just over $500 million. This is probably low. But that’s the crude math.

I spoke to comScore this morning about the methodology and here’s what they said:

There was no manual review of ad copy. They also didn’t capture which ads were IP targeted. The way they determined local or geotargeted advertising was by estimating the percentage of ads that would appear normally in the particular DMA they were examining. Then they compared that to the actual percentage of ads appearing in that DMA. If there was a higher percentage of ads than the normal average in the DMA they assumed those ads were geotargeted.

There’s ambiguity and under-counting here to be sure but it starts to get at very interesting data.

Any thoughts?