The article goes on to focus on how TechCrunch, BoingBoing and Fark are earning a substantial amount of money, together with blogging networks like Weblogs and Gawker. These are, the article says:

Real businesses, with real revenue streams from real advertisers — not overhyped next big things with pick-a-number valuations based on selling out someday to some overenthusiastic big-media sugar daddy.

(you may recognise what looks a lot like a snarky reference to the recent story about Kevin Rose being worth $60mn in 2.0’s rival publication, BusinessWeek)

There’s a little bit of a problem with definitions here. Is a blog that’s entirely made out of user-submitted bits and pieces (Fark) a blog at all? Is a blog written by a team of people (TechCrunch, BoingBoing) still a blog, or has it become an online magazine? For me, many of the Technorati top ten blogs, like the Huffington Post, have definitely gone into magazine territory, or maybe even become an online newspaper. With smaller, close-knit teams, it’s more of a grey area. I think TechCrunch and BoingBoing remain united enough that they do qualify as blogs.

The point may seem like semantic nitpicking, but the difference in posting volume that a team can bring then acts as a multiplier on your page views, and so on revenues. Advertisers mainly buy on a CPM (cost per 1000 page views) basis. A blog with two authors ought to have at least twice the page views as a blog with one — there’s double the posts for regular readers and twice as many possible extra readers who are just occasionally drawn in by your headlines.

Moving on, there are two good reasons why blogs are starting to attract advertisers. In the first place there’s an attraction to the personal touch and the integrity of blogs: “their recommendations are highly valued by readers — which naturally has made advertisers take notice”. If you wrote a blog about motorcycles, for example, then there are two, maybe three, reasons why, say, Yamaha might be interested in advertising with you. It’s a relevant audience, and it also signposts the company as one that’s interested in supporting and talking with its customers. If you’re cynical, you might feel that there’s a lot more chance of positive posts about your company, or fewer negative ones, anyway.

Secondly, there’s much more money in the internet marketing space as a whole, and it’s growing all the time. While the article says Google adverts should not be expected to make bloggers much more than “beer money” in most cases, brand advertising is in a completely different league: “Web ad agency Organic puts ad spending on blogs at $40 million this year.[…] blog ad spending is roughly twice what it was last year. With overall Web advertising expected to grow by 50 percent to $23.6 billion in 2010, it’s certain that more and more ad dollars will land on blogs.”

Blog networks are a good way to share the costs, cross-promote and create large readerships from groups of blogs. The Gawker group of sites, for example, which include LifeHacker and DownloadSquad, did 60mn page views in June. Twenty reasonably well-respected bloggers working together in a network can afford to have salespeople representing them in a way that a solo blogger cannot. This is how Federated Media works, acting as an agent for sites like TechCrunch and GigaOm.

There’s also an admission in the piece that advertising on blogs doesn’t really lead to sales. Intel’s blog advertising campaign for its Core2Duo chips attracted click-through rates of less than one percent. Perhaps crucially, the advertisers have chosen to be on the sites for the benefit of association, being seen as a company that’s interested in the blogosphere, rather than for direct sales. There’s a big difference between the advertising potential of a blog, where readers have come in order to read your daily post, and — say — a computer review site, where they are actively considering a purchase. If advertisers wanted clicks, the computer review site would win hands-down.

This is crucial to the chances of most bloggers because there’s a very big difference between the readership of the top 100 blogs and the rest of the great unread. The top two answers in a Google search receive something like 80% of the clicks. If you’re in the Long Tail, one of the other 49 million blogs with information on the subject, then you aren’t going to get nearly as much traffic. If you get an advertiser or a sponsor, then it will be because that client wants to be associated with the grassroots of the subject you write about. And it doesn’t get much more “grassroots” than having fewer than 500 readers a day!

Sadly, though, I think this association advertising — the bloggers’ bonus — will be the first thing to disappear if times get tough. I’m not convinced that many organisations, and even more so their agencies, view good will and listening on quite the same level as new orders. I think they ought to, but at some point the fat financial controller gets involved and good will becomes a luxury. The positive side of that risk, though, is that — provided it isn’t your day job — it probably won’t really matter that much either.

Click fraud occurs in pay per click [PPC] online advertising when a person, automated script, or computer program imitates a legitimate user of a web browser clicking on an ad, for the purpose of generating an improper charge per click. Click fraud is the subject of some controversy and increasing litigation due to the advertising networks being a key beneficiary of the fraud whether they like it or not.

So basically, advertisers are paying Google, Yahoo and other publishers money for every time someone clicks on their advert. Obviously, if you’re an advertiser, you’d hope that someone clicking on your advert means that person is interested in your products and services. Unfortunately, there are people who click on adverts or write software to click on adverts that have no interest whatsoever in whatever you’re selling. This has led to some very bold scare stories about PPC being doomed, the end of western civilisation and other such tosh.

It’s annoying to waste your advertising budget in any way whatsoever, but it’s part of the game. Advertising is about speculation: you can’t avoid it. “Half the money I spend on advertising is wasted. The trouble is I don’t know which half,” said Lord Leverhulme (or David Ogilvy or John Wanamaker — it’s a bone of contention). That statement has been true for a long time and it’s only PPC that’s ever helped business owners pick the right 50%.

Pay-per-click advertising is different from old media. It’s a lot more secure than advertising in a newspaper, magazine or broadcast. Advertise in a daily paper with a circulation of 1mn readers and they’ll charge you $80,000 for a page. You’ve got no idea of the ROI until you try it out. If the newspaper doesn’t reach the sort of people who buy from you or your advert is rubbish, then you’ll get nothing back. If they aren’t in the mood to buy things then they will flick straight past the advert. The advertiser doesn’t get his/her money back in that case either. Why on earth are people suggesting they should get refunds from Google and Yahoo?

With pay-per-click, apart from the click-frauders, the only thing that advertisers really pay for are potentially interested customers. The economical efficiency of that compared to old media is staggering. When Eric Schmidt was asked by ZDNet whether he thought click-fraud would slow the growth of PPC advertising and Google, of course, he said no. It never will, because despite the scare stories, it works very well.

Think through the sums. With traditional advertising, $80,000 buys you a presence in front of a million newspaper readers or a 30-second TV spot. You have no idea if any of them will respond or whether you got your messaging right. If you spend $80,000 dollars on AdWords or Yahoo, on the other hand, then you get the exposure anyway and that’s almost free. Totally free normally. Because the ads are published on the basis of relevant context, then you’re almost guaranteed the sort of audience you want. The thing you pay heavily for is clicks: the equivalent of phonecalls from your newspaper advert. Let’s go overboard and say that half of those clicks are fraudulent. That’s wayyyy higher than it really is, but let’s suppose that 50% of the clicks you got were generated by a machine. So what? You’re still getting massively increased efficiencies compared to the old model. Because of the scare stories, the networks are extremely vigilant in hunting click-fraud, so I really believe it will never represent a large fraction of an advertiser’s traffic. It’s just not in the networks’ interest so they’ll work it out.

Traditional media won’t sell you advertising on a response basis as a new customer. Probably they never will. They are in business too. They can’t take the risk that the advertiser has produced an effective piece of copy, or that they know their customers well-enough to select your publication because it will work. The advertisers themselves have almost no idea until they try it. The only reason Overture or AdSense exists is because of the explosion of internet sites. The number of blogs doubling every six months. If you’re the thousandth most-visited site about technology, and most are a lot lower than that, then you might still get decent traffic, but you won’t get an interview with an advertising agency. They have 999 other technology sites to contend with that are more important than you.

Nobody in this argument is thinking about the Long Tail of internet sites that are providing this terrific value. The ones that are 1000+ on the rankings. These are the networks’ lifeblood because the owners can’t get brand advertising on their sites. When will those people get fed up and stop offering such cheap rates, I wonder? When they realise that a couple of dollars a month is an insult to the effort they’re putting in and the audiences they’re getting? The proposed introduction of Cost-Per-Action advertising might spell the end of this particular bandwagon.

Perhaps the next step is about the distribution of brand advertising. Giving PowerPoint presentations to a bunch of ad agency guys is old hat. It just isn’t efficient any more. There are too many viable, cheap, relevant media outlets to allow for it. What we still don’t know, I think, is the next best thing.

I asked him about MySpace’s apparent victory over Yahoo! in the recent Hitwise ratings. “Well, MySpace should make around $350mn in sales revenue this year. Yahoo! will make about $4bn. That’s quite a difference. And it means that Yahoo! will have the investment opportunity to add better content and services.“

“The social networks have a real problem with advertising. That’s especially true with brand advertising because the advertiser has no control over the context in which their brand appears. That makes them very anxious. The networks need to do two things. First they need to develop tools that are able to interpret what is on the page before they show the advertisement. Otherwise, you’re putting credit card applications on pages belonging to eight-year olds, which doesn’t make sense for anyone. This intelligent interpretation of the page content and the behaviour of users will also be driven by the child safety lobby. It is possible to observe the behaviour of users to pinpoint the behaviour of predatory adults. They’ll need to be doing this in the background anyway, so if it also gives them more information about who visits which pages then that can also help them with advertising.

“Second, they need to develop areas of interest — sports, music, gaming, etc. — that will provide advertisers with the possibility of choosing more carefully the area where their copy appears. At the moment, all the users are lumped together and that isn’t appealing or useful to advertisers. More granularity is key. To go back to Yahoo!, if you think about the Yahoo! Finance page or the cars sections, then having those sections are useful for users and advertisers. Users know where to go to get information on certain topics. Advertisers will know where their copy is most relevant. And users will accept relevant advertising, especially in a free service. I can also see a move towards more localised services, with networks that are devoted to your local community.”

So where are we going? What’s web 3.0 going to be like? “The semantic web is coming. The words on a page will intelligently link and record themselves to create a much more seamless experience. At the moment, you have to apply a lot of human intelligence, but it’s the hope that as the web evolves, it will start adding intelligence of its own.” But that’s some way off, I argue, with different microformats in competition. “Sure,” says Mark, “there’s a lot of miles to go. In the meantime, we’ll see a big extension of the things that are happening now. More participation, better ease of use, greater use of communities.”