Online news registration may not be working [Thanks paidcontent!].

Among the reasons offered are that there's a lot of false information being posted and people are sharing registrations. Some ten to twenty percent of registered email addresses are bad, and no one knows how much of the demogrpahic data is phony. But newspapers seem to think they need demographics to sell ads.

"Our view is that we need help from you: We've got to pay for what we do, we've got to convince advertisers into looking at us and tell them that these are the demographics we now know about our readers," [Atlanta Journal-Constitution ombudsman Mike King] said. "The old standard — advertising geared to people who live in the areas we cover — doesn't work anymore."

But Google built the largest, fastest-growing advertising business on the Web without any demographics at all. And a huge chunk of that business is never touched by a sales person.

The other part of the justification is that it's becoming more expensive to publish a newspaper on the Web. That's weird. It should be getting cheaper.

Belden has released a study that says unique-user counts based on cookies are too high. Can this be news to anyone? I remember having this same conversation with my webmaster (remember webmasters?) back in 1995.

I've used cookies to count users on MediaSavvy, but I've never told anyone what this number is, or believed it myself.

Underlying dissatisfaction with this crude method of counting users is no doubt one of the many information issues that are driving news sites to user registration. While I'm unenthusiastic about registration as a method of gathering useless demographics, it does help create countable and measurable identities for users.

The Perseus blog survey that says two-thirds of blogs are inactive has gotten a lot of attention lately.

Cyberatlas does a good job of comparing Perseus' study to The National Institute for Technology and Liberal Education (NITLE) Blog Census, which shows that two-thirds of blogs are active.

The big difference is that Perseus took the easy way out and only looked at hosted blogs, while NITLE took the trouble to gather a reasonable sample. It appears that NITLE has missed a lot of blogs that Perseus counted.

The two surveys' blog counts are striking. NITLE measures 1.4 million and Perseus estimates that total is 4.1 million. In any event, the number of bloggers, active and otherwise, is in the millions.

There is some great information on the NITLE site that is worth checking out.

What is a marketer supposed to do with that information? Target his ads to the smallest, poorest states in the country? I'm trying to imagine why this information is useful. This is consistent with a quote on Digital Deliverance about how companies compulsively collect worthless information.

Don't target your customers by where they live. Target them by what they do.

According to ComScore Media Metrix, monopoly newspaper penetration of their home markets ranges from 4% of Web users in Philadelphia to 16.6% in Atlanta. I left off a couple of nonmonopoly markets and the Washington Times from the bottom, and the more-than-local Washington Post from the top of the list.

Given that major metros typically have around much higher penetrations of their home markets -- especially among Internet users -- this is a pretty pathetic performance for a free product. Local newspaper sites have plenty of useful local information, yet they can't muster 20% of their home markets.

The era of repurposed online newspapers must end if newspapers are going to defend themselves against smaller, hungrier rivals.

Online personals aren't content. People don't pay to read personals--they pay to post them. Or they buy access to a poster's contact information.

Clay Shirky calls it communication. I'd call it advertising. But, I suppose paid communication is advertising, isn't it?

Whatever you do call it, observers increasingly agree that the Online Publishers' Association shouldn't count it as paid content. The good news is that the OPA is finding a much more skeptical audience this year than in the past.

Their information is so good and so useful that it's a shame to see them get use the same dopey definition of content. No doubt they will continue to do so as long as news outlets report their big market size number uncritically.

Vin Crosbie have taken on the Online Publishing Association's hype and loose interpretation of the data in their latest online content sales report. Vin was the first to cry foul on the original OPA report's fast-and-loose definition of "online content" last year.

Rafat Ali ridicules the following hopeful statement in the report: "While slowing growth is indicative of a maturing market, we may also be in the midst of a quiet period during which content providers are readying new premium paid services for an increasingly receptive public." He also has come around to the conclusion that online dating services don't belong in the online content market.

We need a new definition of online content sales.

With our per capita income, US broadband penetration should be about twice what it is. Take a look at the free executive summary [PDF] of the International Telecommunications Union's Birth of Broadband report.

There's a great graph (Figure 3 on page 9 of the executive summary) that shows we're lagging behind the adoption curve that the rest of the world seems to be riding.

What forces are retarding the implementation of broadband in the US?. Clearly, it's not content, as the copyright holders would have us believe. There's more US-developed content than anything else on the Net.

The ITU believes the single most important driver of broadband penetration is competition. It makes you wonder why the FCC thinks eliminating competition is the best way to get the telco's to deploy fiber optic networks.

The Online Publishers Association has released a demographic study that compares online content buyers to the Internet users as a whole.

Most of the data show that the demographic differences between content buyers and everyone else are meaningless. They're the tiniest bit younger, a little more heavily represented among people with incomes over $100,000 (like buyers of everything else), and their households are smaller (probably because they're younger).

The real difference is behavioral. Internet content buyers spend about twice as much time on the Net and view more than twice as many pages. And they're a little more likely to have broadband service (about 60% versus 50% for all Internet users).

The most interesting fact is that they spent less on conventional ecommerce ($235/quarter vs. $315/quarter) than the average online buyer.

This data confirms my thesis that demographics are meaningless to Internet marketers. Internet content buyers look like everyone else on the Net. We should be looking at behavior.

The market for online content sales is growing, but contains the seeds of its own destruction. That's my interpretation of the new online content sales report [PDF] from the Online Publishers Association. The report covers the first half of 2003. Here are the highlights, in my opinion:

• The core online content market grew a respectable 14% between the first half of 2002 and 2003. The OPA's definition on "content sales" is still too expansive my taste. They include personals (up 77% since 2002), streaming audio and video, games, credit repair, fantasy sports, and greeting cards. Core online content, in my analysis, includes business content, research, and general news. This market was $255 million in the first half of 2003, or about a third of the OPA's content market.
• General news is flat at about $35 million/year.
• Online advertising is an order of magnitude larger than online content. The Interactive Advertising Bureau hasn't released their numbers for the first half of 2003 yet, but it's likely to be more than $3 billion.
• Subscriptions dominate the online content market. About 88% of the market was subscription-based. Because the OPA doesn't break this out by category, it's hard to know how this applies to the core content market.
• Single payments under $5 are less than $15 million/year (8% of total single payments which are 11.5% of the approximately $1.5 billion/year market). The OPA calls these "micropayments". I'd call them "minipayments".
• Revenue per content buyer has been stalled at just under $25 for nearly two years. All growth in revenue comes from the conversion of Internet users into content buyers. This isn't too surprising. It's more than they pay for the daily newspaper. It's about what most people pay for Internet access. It's about half what they pay for local telephone, long distance, cable, broadband, or mobile service. Does anyone think that it will ever be greater than any of these?
• The average content buyer spent $.25 on single payment content in the first half of 2003.

Conclusions: The market for online content will stagnate if content sellers can't continue to find new buyers. Publishers of general news and information are misguided if they think that selling subscriptions or archives will ever be a serious business. Meanwhile they risk conceding their core markets to free, advertising-financed competitors if they pursue it.

The International Telecommunications Union has released a report that says The US is 11th in broadband penetration worldwide, with only 7% versus nearly 21% in Korea. I added emphasis to a key paragraph below;

More than 10 million of the world's high-speed Internet users are in South Korea alone, a rate of 21.3 broadband subscribers per 100 inhabitants. Hong Kong was in second place with 14.9 percent and Canada was third at 11.2 percent.

The United States was in 11th place in the per-capita broadband rankings at 6.9 percent, though it had the highest overall total with 19.9 million subscribers.

Japan was in 10th place, with 7.1 percent broadband use. But ITU experts expect Japan to move up because it is now offering the world's fastest speeds and lowest prices. Broadband service that is about 520 times faster than a dial-up modem is available in Japan for about $24.19 a month.

...

[Taylor Reynolds, one of the authors] said a key reason why Japan and South Korea are so far ahead is because of heavy competition among broadband providers. The Japanese and South Korean governments have taken steps to encourage the use of broadband, such as requiring telephone companies to let competitors use existing lines at low cost.

The FCC has taken the opposite approach, granting monopolies to local phone companies in exchange for them granting citizens broadband access when it suits their purposes.

Meanwhile, Peterme points to a Business Week commentary that says we need to grant telcos a monopoly on DSL, subsidize their deployment, and improve content protection to increase broadband penetration. He notes correctly that more content isn't what we need to create broadband demand. There's now plenty of evidence in worldwide Internet penetration to show this is nonsense that only serves intellectual property owners.

Right now, there is more money to be made selling ads online than selling news.

With online advertising continuing to climb (Emarketer says online ad spending will be up 4.8% in 2003), and with online newspapers getting an outsized share of that growth, who is going to be willing to jeopardize their seat on the gravy train by charging for content?

The Online Publishers' Association says that their members (all big publishers) are seeing a 38% growth rate in online advertising. This is consistent with what the big newspaper publishers are reporting.

It's no surprise to anyone that the Web now dominates the real estate advertising market.

In response to the question, 'What resources did you use in your home-search process,' 65 percent of respondents listed the Internet, while 49 percent mentioned newspapers. Two years ago, 43 percent of respondents listed newspapers as a primary information source while 43 percent listed the Internet as a primary source."

For the true believers (like me), it's a bit of a surprise that the Web was never able to disintermediate the real estate agents' MLS monopoly. However, no one should be surprised that the agents recognized the Web as not only a cheaper, but a superior method to promote houses (and their services).

Not only did newspapers abuse their local classified monopoly for decades, but they never were able to offer an efficient buy in a business where the only thing that matters are location, location, and location. Why advertise to an entire metropolitan area when you only want buyers who are interested in a single neighborhood?

Bay Area real estate advertising never recovered from its decline in the recession of the early 90's because the advertisers in their desperation found newer and cheaper ways to sell houses. The business was already pretty damaged by the time the Web came along.

Given the poor prognosis for newspaper classifieds in general and employment in particular, I was startled to hear that the Conference Board is still promoting its help-wanted index as a measure of the employment market, saying "Because ad volume has proven to be sensitive to labor market conditions, this measure provides an important gauge of change in the local, regional and national supply of jobs."

What decade do the Conference Board's economists live in? These days, employment classified ad volume is a lot more sensitive to online competition than it is to labor market conditions.

The AeA (what you probably know as the American Electronics Association) says that broadband growth is slowing:

"Now, the limiting factor [to the growth of broadband] is access and price," said William T. Archey, president of AeA, which was formerly known as the American Electronics Association. Only half of people living in rural areas have access to high-speed Internet service, and many others don't want to pay $50 a month for it, he said.

The pressure on the broadband access duopolies to increase access and lower prices is growing. They're not exactly a friend of the citizen, but the AeA should provide a political counterweight to the duopolies and make it clearer to everyone that they're holding back the economy with their current approach to broadband access. [Thanks Poynter E-media Tidbits]

I hate media concentration as much as the next guy, but this baffles me. An "advocacy" group called Children Now says that the amount of kids' television programs in Los Angeles fell sharply when one company owned more than one of the city's TV stations.

Put aside their misuse of the word "duopoly" to describe one company owning two TV stations in the same market. My question is whether anyone really believes that kids are better off if there are more TV shows for them to watch.

The Online Publishers Association says its members, the largest online publishers, are reporting their online ad revenue is up 40.7% in the first quarter over the first quarter of 2002. Interestingly, total revenue was up 37.6%, suggesting that non-advertising revenue sources are not keeping up as a share of total revenue.

The members of the OPA are the largest online publishers, including of About.com, Bankrate.com, Belo Interactive, CBS MarketWatch, CNET Networks, CondeNet, COXnet, Edmunds.com, ESPN.com, Forbes.com, Hearst, Internet Broadcasting Systems, Knight Ridder Digital, Meredith, MSNBC.com, New York Times Digital, Scripps Networks, Slate, SportingNews.com, Tribune Interactive, USATODAY.com, Wall Street Journal Online, weather.com and Washingtonpost.Newsweek Interactive.

The number of broadband users grew 50% last year, according to the latest memo from the Pew Internet Project.

CyberAtlas is running an interesting chart based on Pew's data that shows broadband users and experienced dial-up users are pretty similar in their online activities. Not surprisingly, they use a lot more streaming media. However, the only traditional Web activity whose use seems to be increased by broadband is reading news.

Pew predicts that the pace of broadband adoption is likely to slow as most ready buyers cannot get access.

Slowing growth rates for broadband adoption could lead to price cuts in the near future. Don't sign any long-term contracts.

The Online Publishers' Association has released a new survey of at-work Internet use by daypart[ PDF].

The Internet is the best way to reach your audience while they're at work, and this study does a good job of showing the behavior of this audience.

I wish the study had spent less time wallowing in demographics, which are easy to collect but pretty meaningless, especially in this context. I also don't understand why they're looking at the at-work market as a consumer, rather than business, market.

PaymentOne's new study concludes that security concerns are the main reason why consumers don't buy more content online.

That seems a little weird to me: Not high prices, not the inconvenience of payment, but security.

But if you take a look at what content they're buying online, 52% of it is advertising and another 9% is Internet services and communications. Only 12% of what they're buying is news and information.

It seems to me that the value of classified ads and communications services are well established on the Net and that security (the principal bugaboo of all mainstream ecommerce) would be the main problem keeping consumers from buying them. But for news and information, the value proposition and ease of payment are much bigger issues.

Until we treat online news and information as a distinct market, most research into the "content market" will be useless to online publishers.

It's not just Canadians. Koreans are two and a half times as likely to have broadband connections as US households (57% vs. 23%). (See also Part 1 and Part 2)

Korean DSL connections are not only half as expensive, they are a lot faster, especially upstream.

Drawing conclusions about Americans from Korean behavior is riskier than using Canadian behavior, but it seems pretty obvious that the telcos the the main obstacle to widespread adoption of broadband. And I find it hard to believe they can't make money at $25/month by lighting up wires that are already in the ground.

Korea's secret? Encouraging competition with the monopoly provider. Didn't we try that?

The United States has gone through a similar shakeout, except it happened before the broadband network was extensively built. The Telecommunications Act of 1996 set off a surge of expansion that collapsed when the Internet bubble burst, driving many of the broadband start-ups, like Rhythms NetConnections and NorthPoint Communications, out of business. While fixed-line operators in Korea and Japan were cajoled into making D.S.L. service available at low cost, analysts say that the Bells are reluctant to cut prices.

At around $50 a month, broadband costs about twice as much in the United States as in Korea and Japan. Worse, broadband in the United States is slower and less suited for interactive entertainment and other two-way uses because it relies on an asymmetric system that receives data much faster than it can send it.

The Bells say they are doing everything they can to promote broadband. But critics say the phone companies view broadband as more of a threat than an opportunity, so they have done little to rectify these problems.

We also failed to enforce the Telecommunications Act, allowing the Bells to starve their competitors.

The current broadband market is a drag on the economy. Billions of dollars in consumer investment in computers, peripherals, software, networking, downloadable music, and content are being held back back because of the low penetration and poor quality of our current broadband connections.

TNS Telecoms says that per-user telecom spending in the US decreased in the fourth quarter of 2002, the first quarterly decrease in more than three years.

For years, telcos have been relentlessly focused on increasing ARPU (average revenue per user, not the guy who runs the Kwik-E-Mart) to the detriment of customer service and rational pricing. Between competition for zombie telcos, their inability to deliver any new value propositions to their customers, their debt chickens coming home to roost, and the Bush recession (or the Bush dip in the Clinton recession if you prefer) these guys have what positive thinkers refer to as "an insurmountable opportunity".

Short-term, don't expect the telcos to cut publishers any slack. They're going to be looking for change under every cushion in the marketplace. Long-term, look for their receivers to loosen things up after they've written off their debts.

Both Jupiter and eMarketer are projecting that consumer ecommerce will increase significantly in 2003.

This is good news for online publishers, because consumer ecommerce is a key source of online advertising.

McKinsey says Canadians are twice as likely as US citizens to subscribe to broadband. This despite the fact that, according to McKinsey, roughly the same percentage of households have access to broadband in Canada (89%) and the US (87%).

I first looked at this issue Monday, when Ipsos-Reid released broadband penetration numbers for Canada and the US, but I didn't have the access data, so it wasn't clear whether that was the reason for the difference.

Interestingly, roughly the same percentage of online citizens of each country are interested in broadband (49% in the US and 54% in Canada). Also, the reasons for subscribing are about the same -- speed, always-on connection, keeping the phone line open.

So, why are Canadian broadband providers getting so many more takers for their services?

NOTE: There are some problems with these two studies. McKinsey says that 43% of Canadian online households have broadband and Ipsos-Reid says 54%. McKinsey says 27% US online households have broadband and Ipsos-Reid says 34%. I can't account for the discrepancy between the two firms, but the message is awfully clear.

I'm working on a project that involves using more corporate Web sites than I would if I weren't getting paid to do it.

I am stunned by how many of these sites are using Flash for navigation and information and how badly this stuff works with all three of my Mac browsers (Camino, Safari, and Internet Explorer).

It's sad how much they must have paid for these ugly, broken sites they present to the world.

One of the less-celebrated aspects of the blog phenomenon is how much activity there is in using CSS and simple, standard HTML to create fast, elegant sites that work really well. I've got a folder full of beautiful sites that I plan to steal ideas from for my next redesign.

Flash on a Web site is beginning to feel more and more like tailfins on a car.

it's still less than £200m for 2002, but it's up 19% in a year when US Internet advertising was down 17% . [Thanks, MarketingFix]

The only explanation I can think of for this remarkable performance is that the US was still digesting some bad revenue data last year. if that's the case, it's further evidence that the US Internet advertising market has already bottomed out.

The Internet Advertising Bureau and PriceWaterhouseCoopers, who released the study, say that Internet advertising was 1.4% of all UK advertising and "on track to hit its target of taking a 2% slice of the advertising market by 2004". That would be a very satisfying 20% annual growth rate.

The Pew Internet & American Life Project has a interesting new report on who's not online and why. One sixth of Americans are former users and a quarter have no direct or indirect experience with the Internet.

But an equally significant reason why people are not online is lack of desire—they do not want the Internet, do not feel that they need it, and do not feel that it holds anything of interest or value for them. They believe they are not missing out on anything by not being online. For some, this disinterest is based on incorrect assumptions about online content, but for others it is a reasoned choice, based on personal preferences for communication style and information retrieval or past Internet experience.

There's a ton of information here and I'll be going back to this report for more items later, but I wanted to make sure you knew it was available.

54% of online Canadians are using broadband, versus 34% of US citizens who are online, according to comScore Media Metrix Canada. From CyberAltas:

"With more than half of our online population using broadband, it's clear that Canadians are hooked on speed," noted Brent Lowe-Bernie, president of comScore Media Metrix Canada. "The cable and telephone infrastructure in this country has allowed broadband service to flourish, and Canadian activity within the Digital Media landscape has responded accordingly."

This raises more questions than it answers about a staggering gap between two nations with so few cultural differences. Why are Canadian Internet users 59% more likely to have a broadband account? How much has to do with the percentage of Canadian homes that can have broadband connections, price, marketing differences, regulatory policies?

In other words, what can the US learn from Canada about how to increase our use of broadband Internet connections?

Advertising may not be sufficient to support Internet media.

MarketingSherpa has an important study of marketers' discouraging experience with Google Content Targeted Advertising on general content sites. [Thanks, MarketingFix]

Generally newsy, how-to, and highly targeted articles on niche sites tend to get far better ad clicks than newsgroups, bulletin boards, general interest sites, or stagnant info pages.

Unfortunately Google didn't take this factor into consideration when designing the program. They chose the partner sites for contextual ads mainly based on traffic (sites had to have more than 20-million pageviews a month, which very few niche sites do) and "quality" which seems to mean being G-rated.

What they found is that click rates are abysmal for many contextual ads, and that conversion rates (share of clickers who take the desired action after clicking on the ad) have been disappointing for many advertisers as well.

MarketingSherpa says low click rates aren't too big a deal if you're paying by the click and you're getting decent conversions. True enough, but it also means that pay-per-click ads aren't going to pay the bills for an online publisher.

The low conversion rates are even more troubling. This may doom general content sites to seeking price-sensitive, rich-media-laden, dubiously-effective branding ads instead of light, direct-response advertising that is the fastest-growing category on the net. According to the NY Times, search-related advertising has grown "from an estimated $400 million in 2000 to $1 billion in 2002 and even higher this year".

If this admittedly limited research holds true, what are the implications for online publishers?

First, don't assume this means you should charge for content. Just because it's still challenging to sell ads on general content, that doesn't mean you can sell it to your readers.

Second, seek services revenue. The services sector is producing more revenue and growing faster than general content.

Third, consider publishing content that supports direct marketing. Get your readers when they're considering a purchase. For local newspapers, real estate comes to mind.

While 2002 Internet ad revenue was down a whopping 17% from 2001, the fourth quarter was down only 10% from the fourth quarter of 2001, and it was up 2% from the third quarter of last year. This is the first quarterly increase since the second quarter of 2000. This information is from the Internet Advertising Bureau / PriceWaterhouseCoopers advertising estimate for the last quarter of 2002.

Remember, ad numbers that are more than a couple of years old contain a lot of junk, including the tail-ends of some awful long-term deals and hundreds of millions of dollars in phony advertising that AOL has since written off. Under the circumstances, this quarterly increase is very encouraging and may mean that Internet advertising has hit bottom and will grow again on an annual basis when the ad market as a whole recovers.

The IAB release mentions in passing (1) 15 online publishers are getting 80% of the revenue, and (2) "the majority of online publishers are profitable, and their revenues continue to rise year-over-year".

IDC has lowered its forecast for IT growth in 2003 because of increasing uncertainty about the economy.

I conducted a survey of CIO's back in August and found them to be fairly optimistic about 2003 spending. But clearly we still have no idea what's going to happen and no one's going to spend money in that kind of environment.

"The outlook for the next six months continues to be extremely volatile and a double-dip IT recession can't be ruled out in a worst-case scenario. But the fundamental drivers remain solid," said Stephen Minton, director of IDC's Worldwide IT Markets group. "Once the fog of war has cleared, there will be a gradual recovery in corporate profits and business confidence, and this will translate into increased IT spending. We expect to see improved market conditions in every region in 2004. And by 2006, the global IT market will generate $1 trillion in revenues."

In other words, someday we'll get past this, but we're not sure when.

More and more online publishers are requiring users to register if they want to read content. Adrian Holovaty correctly points out that if you don't do it well, site registration is an invitation to fraud, but Jay Small counters that eventually we'll learn how to do it and readers will accept it.

In my opinion, demographics (even when they're correct) are usually the wrong way to target online advertising.

Demographics are almost always proxy for something else. You may believe that the most likely user for your product is a woman between the ages of 25 and 44, but what you're really looking for is anyone who might want to use your product. It's just more efficient to buy magazines and TV shows that are viewed by women 25 to 44.

On the Web, you can target users by context and behavior. That's a lot more powerful than demographics and it's the reason why Overture and Google are changing the way we think about Internet advertising.

Booz Allen and Neilsen//NetRatings, in an article titled "Seize the Occasion! The Seven-Segment System for Online Marketing", say:

The focus on demographics — the outward and visible signs of inward attitudes — grew more out of marketers’ need for analytical criteria than out of any inherent link between a person’s demographics and shopping behaviors.

...

But this approach simply applies traditional marketing methods to the e-world, without exploiting the Web’s unique strengths. The abysmal performance of targeted banner advertising on Internet portal sites, where click-through rates today average 0.1 to 0.2 percent, underscores the failure of this conventional wisdom.

Wherein lies the flaw? An exclusive study by the Digital Customer Project, an alliance between Booz-Allen & Hamilton and Nielsen//NetRatings Inc., shows that the most effective segmentation scheme for online consumers first groups them by their individual behavior at a point in time, not by demographics or psychographics, or even by aggregate online behavior.

There is also an excellent interview with one of the authors at AvantMarketer. Both the article and the interview are a little long-winded for my taste, but the point that demographics is a lousy way to target Internet advertising is dead-on.

Nora Paul distributed a list of research about kids online for last week's New Media Conference at Berkeley. Two reports in particular were especially recent and interesting:

Connected to the Future is a report on children's Internet use from the Corporation for Public Broadcasting. It addresses, among other topics, use among underserved groups.

MobileYouth2003 covers marketing mobile products successfully to the Youth, from a European perspective

Jupiter says that online content sales will rise from $1.5 billion in 2002 to $2.0 billion in 2003.

Jupiter didn't define online conent in their public statements: "The $2.0 billion forecasted for paid content spending is fragmented across over a dozen categories ranging from news to sports to health to adult content, making it difficult for any one company to collect a significant share of that spending."

Without further detail, it's impossible to compare that number to the Online Publishers Association number of $1.3 billion in content sales in 2003. The OPA didn't include adult content and Jupiter may not have included services and personals.

"For at least the next 18 to 24 months, most online media companies should generate 60% to 70% of their revenues from advertising." says Jupiter analyst David Card. That average confuses the issue, probably by including some sites and sources that shouldn't be lumped into the average. I estimate that non-adult online publishers are getting less than 10% of their revenue from content sales.

UPDATE: David Card tells me in email that their number does include such services as personals and greeting cards. Also, they put pornography at less than $250 million/year, with little growth.

Emarketer has a good overview of current research supporting online advertising:

• The audience is more affluent
• It's effective in changing users's perceptions of brands.

Robert Losch at Marketing Fix adds

• It's the best way to reach the work/school audience
• Consumers are using it to find product information

DoubleClick claims that by increasing the amount of Internet advertising in your media mix, you can increase frequency and brand awareness.

I would add that it's the best or even the only way to reach people who have begun to limit their consumption of commercial media: TV nonviewers, DVR users, or NPR listeners. Many of the people are as influential as they are difficult to reach.

Buyers last year acquired 1,087 Internet-related properties for total spending of $23.3 billion, 15% fewer companies for 70% less money. In 2001, acquirers bought 1,283 Internet-related companies for a total deal value of $78 billion.

Top 10 Internet Categories - 2002

Apparently, even the bottom-feeders aren't interested in the content business. At least not yet.

Kevin Werbach points out that two of the top five sites on Alexa's Top 500 are in South Korea.

While Korea may lead the US in broadband and cell phone Internet access, I find it hard to believe that two South Korean portals are more popular than Google.

Could it be that Koreans are more likely to install and use Alexa's software? I would take any data from Alexa with a grain of salt.

It's nonetheless amazing how much of the Web may now be simply invisible to English-speakers.

MSNBC reports that 43 percent of Americans say online advertising is “a nuisance" and 53 percent of active Web users are irritated by online ad clutter:

To be fair, clutter is considered a problem all over, with 75 percent of Americans saying that advertising is shown in “far too many places,” Roper found. TV commercials annoy 65 percent of Americans, and 56 percent object to the ads in print publications, Roper found.

But impatience with online ads is growing faster than with traditional media, according to Roper, posing a bigger problem for Web publishers who are still trying to prove the Internet’s marketing power.

While 38 percent of consumers found online ads to be “useful and informative,” an almost equal number, or 37 percent, found “almost none” of them to be useful and informative, the survey found. By comparison, 78 percent of Americans believe at least some TV commercials are “useful and informative.”

“This is a critical time for online advertising,” said Jon Berry, senior research director of Roper Reports. “In terms of usefulness, online ads have a long way to go to catch up to other media.”

Roper loses credibility points for having no information about this study on their site.

Let's face it, the advertising industry doesn't care whether it's irritating or intrusive, only whether it's effective. I'm not sure anyone in the ad industry puts a lot of stock in whether consumers find their ads to be useful and informative, only in whether they take an action or change their attitudes.

Having said that, it's pathetic that online advertising isn't considered at least as useful and informative as (gasp) TV advertising. Utility and information are what the Net is about, yes? [Thanks, MarketingFix!]

CMR/TNS Media Intelligence says that Internet advertising will grow 7.4% in 2003, after declining 12% to $5.7 billion in 2002.

This makes Internet advertising five to ten times the size of the paid content market, depending how you define content.

The biggest sites will probably get the largest share of this increase.

I'm winding down bashing the OPA's billion-dollar-content report, which does contain a lot of good information. But I have one last question: why are personals included in their definition of content, when you can read the listings for free, but have to pay to add your own listing?

Shouldn't personals be classified (no pun intended) as advertising?

I got a thoughtful and irritable reply from Rick Bruner of MarketingFix on my analysis of the Online Publishers' Association study of the size of the content market. Rick has gotten me to refine my thinking (but not to change my mind).

Defining content is really hard. I'm leaning toward dumping the term entirely because it's so vague as to be meaningless. I think the OPA has exploited this vagueness to overstate the size of the market that online publishers can address.

I would include the following in my definition of content:

• News & Information: General news, magazines (consumer reports, salon), horoscopes. Excludes Business & Investing. Includes archive database revenue from a single site. Excludes aggregated content databases, such as Northern Light.
• Business & Investing: Wall Street Journal, TheStreet.com, Smartmoney. Does not include stock prices.
• Databases: Hoovers, Northern Light, Ancestry.com, stock prices. Includes database that aggregates information from multiple sources for flexible searching.

I would exclude the following from my definition of content:

• Services: Includes greeting cards, online games, fantasy sports, email, and personals. Personals aren't content, they're a service. You can browse the listings on Match.com for free. You pay to add your name to the database and to get contact information on people you want to meet. It's not clear how much of Yahoo's and MSN's "content" revenue comes from services.
• Music: This seems like a different thing altogether. It's produced, delivered, and consumed differently, using different media. It's more like software than it is like published content. This includes streaming and downloadable music.
• Electronic documents: Credit reports, people trackers, and vehicle histories feel more like ebooks--single, indivisible lumps of content sold and delivered electronically. Aside from the fact that they're delivered electronically, this business is a lot like selling books.
• Pornography (hard & soft core): Includes webcams, amateur sites, Playboy.com, etc. No one is buying Playboy.com for the articles.

What's my objection to the OPA's definition of content? Rick says on MarketingFix:

For those who maintain "users won't pay for content," I reply "neener, neener, neener." Paid content would appear to now amount to roughly 20% of all revenue for online publishers, and spending for content roughly doubled in 2002 over 2001.

Rick seems to be saying that "online publishers" are getting all $1.3 billion--about 20% of the $6 billion in online advertising in emarketer's 2002 estimate. He's making the mistake the OPA wants everyone to make: inferring that there's $1.3 billion dollars out there for online publishers to get their hands on.

But (a) Two thirds of this revenue will go to companies that cannot be described as "publishers", (b) ten percent of the remainder is going to one property, the Wall Street Journal [NOTE: this used to read 1/3, but I overestimated the revenue of the WSJ], and (c) the growth rate of the revenue available to publishers is a lot less than 100%.

My conclusion from the OPA study:

Paid content is less than 10% of all revenue for online publishers, and that share will not grow in 2003. However, there does appear to be an opportunity for online publishers to increase non-advertising revenue by using their sites to sell services to their readers. The market for online services is twice as large and growing five times as fast as the content market.

Emarketer has an interesting description of how they arrived at their broadband forecast.

I've done a lot of forecasts, and taught forecasting at IDC, and I think this is a good description of one way to approach the problem. This is worth reading. Emarketer is very smart, their forecasts are reasonable, and I really admire their presentation.

The Online Publishers Association is promoting a study [PDF] that says online content sales in the U.S. totaled $1.3 billion in 2002, an increase of 95% over 2001.

However, a closer examination of the data makes this billion-dollar figure suspicious. Here are some of my observations from their top 25, which I have classified into Content, Service, Entertainment, and Database:

• The single biggest category is more properly described as "services". It includes Yahoo, MSN, and personals.
• Yahoo is seeing results from its efforts to increase revenues from fees, and MSN is falling behind.
• Financial information is the biggest contributor to true content sale, with sales of $292 million.
• The OPA describes financial content as "maturing", with a growth rate of 18%.
• Sales of "General News" rose from $52 million to $70 million. No "General News" site was in the top 25.
• Consumer Reports and Encyclopedia Britannica are the two true consumer content brands on the list.
• Consumer Reports has fallen from #5 to #11.
• The IEEE.org site sells conferences, so it's not clear how much of their revenue is really for content.
• Pressplay, the much-maligned music service, is on the list at #22.
• ESPN's revenues are apparently mostly from fantasy sports games.
• It's not clear why Playboy is on the list, since "pornography" is excluded from the survey.

Here are the top 25 content sites, according to the OPA:

The share of Americans who use Internet audio monthly has tracked closely the share who have broadband access, but the share using Internet video has not.

This extraordinarily clear message can be found in the Arbitron/Edison survey released last week. It puts the lie to the idea coming from Los Angeles that potential broadband users are waiting around for copy-protected video content before they sign up.

Despite its intimidating name, the National Technology Readiness Survey isn't very impressive. It has a small sample size and the questions seem either generic or poorly thought-out. I'm including this one mainly in the interest of completeness.

But it's worth checking out. I used some of their data to calculate that the share of Internet users with personal Web sites hasn't changed in the last five years.

This report came out a month ago, but I've been making an effort to note free Internet research reports on MediaSavvy. This is probably the best free report on the state of the Net. The previous two reports are available from this site as well.

Jupiter Research says "more than 60 percent of companies that have deployed Web content management solutions still find themselves manually updating their sites."

This research confirms a study by the Asilomar Institute for Information Architecture that found most users' experience with content management software was unhappy.

Other findings:

• "Overcomplicated, end-to-end packages can as much as quintuple site operational costs over human labor alternatives."
• Most companies felt they overspent on content management platforms.
• Two thirds said they still rely on manual processes to update their Web sites.
• Nearly half felt their deployment "barely scratched the surface of the functionality they originally licensed."
• Only a quarter planned to continue using their Web content management systems as they do now.
• One quarter (27 percent) said they had so many problems they would build another system from scratch.

One media company spent over a year and $250,000 working its content management package into its site production process. "The company recently realized that its content had little structure to speak of, and that because it had not made a strict separation between content and presentation, the company's broader needs for reusing content elsewhere were effectively blocked."

One surprising conclusion: "(o)rganizations should not look to content management systems to publish pages."

The Washington Post and Nielsen//NetRatings have posted a study of affluent online users [PDF]. Not surprisingly, they found virtually all affluent adult shoppers use the Web to make or research their purchases.

They made less play of the fact that the affluent also spend a lot less time with newspapers (19% spend more than 1 hour/weekday) than the Internet (68% spend more than 1 hour/weekday).

While their numbers are growing, Neilsen//NetRatings says African-American Internet use still trails other ethnic groups by a significant degree. Meanwhile, Arbitron and Edison say African-Americans have achieved virtual parity with white Americans when you include school and library use.

Since N//NR's ratings methodology measures at-home (and guesses at-work) use, they are almost certainly undercounting African- and Hispanic-American use of the Internet.

Steve Outing notes that the bottom has dropped out of newspapers' employment classified business. Every category of advertising, including other classified categories, are up. But employment classifieds are down 23% -- a sickening drop that goes right to the bottom line.

It may already be too late to save this business. When employment recovers, employment classifieds will not.

The Digital Divide has been bridged by libraries and schools , according to Arbitron and Edison Media Research. The report includes a lot of information on streaming media users as well.

When you look at Internet use at home and work, The divide is small, but significant. 48% of Hispanics and 6% of African Americans have Internet access at home or work compared with 70% among whites.

When you include school and library access, 75% of the total population has access to the Internet compared with 74% of African Americans and 65% of Hispanic Americans.

These numbers are interesting in light of eMarketer's recent assertion that US Internet penetration will peak at 80%. Their conclusion seems about right, but their line of reasoning doesn't prove it.

We have a few more details on ComScore Media Metrix's recent restatement of their at-work measurements, thanks to the Internet Advertising Report.

First, ComScore acquired Media Metrix's sample, but not its software. Second, ComScore was combining two samples collected by very different means (telephone and email recruitment), and combining two very different measuring systems (end-user and proxy server based) at the same time it was expanding its sample.

Without the tracking software, ComScore incorporated its own server-based measuring system with the November 2002 release of Media Metrix 2.0.

In addition to the 50,000 home Internet users recruited via Random Digital Dial sampling methodology, Media Metrix 2.0 expanded its measurement panel to 120,000, including 35,000 samples of at-work and university users.

No wonder there were some problems with their numbers.

Today, Comscore confessed that its numbers for last fall are wrong. The principal source of the problem was its estimate of at-work use. At-work use is notoriously difficult to measure.

But the Times hints at a bigger problem with web site ratings: what are they for?

When advertisers are paying by the exposure, or even the click or the sale, is it really important which site is number one?

When advertisers are (properly) more focused on response than branding, and when its so easy to test a campaign on the Net, is it really important what the demographics of a site are?

The Times tells us some alarming things about differences between the services:

• Comscore has cut its estimate of the time users spend on some sites by as much as 75%
• Comscore and Nielsen use radically different methods for sampling at-work users (email vs. random-digit-dial)
• ComScore says Yahoo had 107 million users in the United States in December. Nielsen says it was 81 million.
• ComScore measures usage on college campuses and Nielsen does not, but this is not enough to account for the difference in the Yahoo numbers.

Web advertising must be conducted like direct marketing and not like broadcast. Overreliance on ratings is malpractice.

According to Nielsen//NetRatings ["The extra slash is for slashing prices!"] Raleigh, NC's active at home online population grew 29% from January 2002 to January 2003, and Nashville and Sacramento grew by 19%. That seems...implausible.

N//NR continues:

The fastest growing local market experienced some big shifts in income levels over the past year. In Raleigh, the number of people with at home Internet access reporting a household income of $75,000 to $99,999 a year increased by 72 percent (see Table 2) from January 2002 to January 2003. Other high-income groups online grew as well with the $100,000 to $149,999 income bracket growing eight percent and the wealthiest bracket of $150,000 to $999,999 moving up 17 percent. By contrast the lowest household income bracket for those with at home Internet access saw its active online audience decrease by 26 percent.

“Raleigh has experienced major growth in high income households that have Internet access over the past year,” said Bloom “This will make it a prime regional target for marketers looking to push big ticket items like luxury cars and appliances.”

This recommendation would be wrong even if the data were correct, which I doubt. Wouldn't it make more sense to target your ads to markets with the greatest concentration of affluent customers and not the fastest growth?

WebWord has an great discussion of Jakob Nielsen's latest screed -- "Homepage Real Estate Allocation". The consensus seems to be that he hasn't had a new idea since the nineties and that he's more concerned with selling books and seminars than on increasing usability.