Jeff Duntemann's Contrapositive Diary Rotating Header Image

The Twilight of the Ad Era

I made a decision late last year without saying much about it: I won’t be using AdSense ads anymore. Now, I’m not going to remove them from existing pages, and I’m not going to shut my account down, but as you might have noticed if you’ve perused my articles over on, my new layouts do not contain ads.

There’s not a lot of point. The curve is heading in the wrong direction.

When I first used AdSense in 2006, my goal was to bring in a dollar a day on average, and I either met or beat that for the rest of 2006 and the first few months of 2007. After May, 2007 things went into a slow decline. My page impressions grew slowly, but revenue slumped, and over 2007 I averaged only 85c per day, which is still worth pursuing. Across 2008 I was averaging only 61c per day, even though page impressions were higher than they had ever been. People just seemed to stop clicking on ads. (“Ad-numb” is a coinage that I’ll offer here if no one else has.) 2009 has earned me an average of about 20c per day, and that’s really not enough to warrant the effort of designing ad spaces into my layouts, especially if it’ll be down to 10c per day next year.

An interesting thing has happened over the course of 2009 so far: Google-tracked page impressions have plunged, even though my overall page hits continue to climb. Some of this is doubtless the rearrangement of my Web content that I began last fall, but it was also true for individual pages (like my Homebrew Radio Gallery) that had not changed significantly since 2006. Daily AdSense page impressions for that single page were always up in the high 30s to low 50s, and are now down to 15-20 tops.

I didn’t start doing anything differently. I’ve never worked at building traffic to my site, and in fact the only way AdSense makes sense to me is if you don’t have to screw with it. Spending time and effort trying to drum up traffic for the sake of ad clicks is time and effort I can’t spend researching and writing new articles (or heaven knows, fiction) so I’ve never bothered.

I think I know what happened: Malware delivered from Web ads has gotten enough publicity that people in large numbers are starting to install ad blockers. This is the only way I can reconcile imploding AdSense page impressions with steadily growing traffic to my site as a whole. Google only counts a page impression when an ad is served; block the ads, and viewing the page does not generate an AdSense page impression.

I’ve never used an ad blocker before, and it was eerie surfing around using the Iron browser, which blocks ads from a huge number of major ad sites (including AdSense) by default. Eerie–and fast. Malware isn’t the only issue with Web ads: Overloaded ad servers slow down page render time, sometimes hugely. This is not new news, but until I saw it myself I couldn’t appreciate the scale of the problem. Iron may not be intrinsically faster than IE or FF, but it looks faster because it doesn’t wait on ads. Blocking ads still makes my conscience twinge a little; here is an interesting discussion on whether it’s wrong to block Web ads. The tipping issue is malware: If all it costs me is time to render your ads, then that may be the cost of viewing your pages. But if there is some significant chance that your ads are serving malware (whether you knew anything about it or not) I feel that I have a right to protect my system and my network. Remember that I can’t tell if your site even has ads before I go there, and if your ads serve malware, my system gets nailed faster than I can back out. The only way I can reliably protect myself against ad-served malware is to block ads entirely, so until each browser instance is a thoroughly isolated VM, there’s no other way.

Thus fades the Great Hope of “free” content supported by ads. What replaces it is obscure. One barely hears the term “micropayments” anymore, and those sites that have retreated behind paywalls don’t seem to be doing well. Among the pubs I read, The Atlantic Online dropped its paywall last year, and the only paywalled site I still read is The Wall Street Journal. Money does need to be involved somehow: I write better material when I get paid for it, and when I pay for material, I have higher standards for it than for what’s lying around free. That being the case, I intuit that a paid Web would be a smaller but far more useful thing than a free Web groaning under the weight of pages (you see them all the time) that exist solely to serve ads. Still, I’ll be damned if I can see the way there.


  1. Jim Tubman says:

    Perhaps some kind of paywall consortia will arise. You would pay $20/month (or whatever) to Amalgamated Websites for all-you-can-eat access to content for all sites associated with it. Maybe it might have tiered levels of access, like cable TV. The consortium would sort out payments to its members based on measurements of use.

    We may also see more of the world’s journalism getting done by organizations like the BBC or CBC, instead of by private publishers. This is not ideal (they have institutional biases and filters that can be hard to overcome) but it is better than nothing.

  2. The newspapers tried this (badly) back in May, and everybody jumped down their throats yelling “Antitrust!” Google “NAA meeting antitrust” (without quotes; these are separate search terms) and you’ll get a sense for it. I’ll take the contrarian view and say, *let them do it*; if we don’t, we’re going to lose more newspapers. It would indeed take an antitrust exemption to allow this, but Obama (somewhat peculiarly) is against it. (Pelosi understands a little better that the newspapers are virtually all joined at the hip to the Democratic Party, and she favors it. Without newspapers, all that’s left is TV, talk radio and blogs. One out of three IS bad…)

    The BBC can offer journalism about the UK and perhaps the world situation, but they can’t cover local issues outside the UK like local organizations can.

    I still don’t see a way out for print media generally, though I will continue looking.

  3. Carrington says:

    Jerry Pournelle’s site seems to make money using the public radio model. I think Jerry is a special case in this.

Leave a Reply

Your email address will not be published. Required fields are marked *