Give the people what they want (?)

Since I don’t make any money from my blog, I generally don’t pay attention to my traffic statistics.

Things would be a little different if I were attempting to make a living – or even beer money – from my blogging. At Tripod, we were obsessed with our traffic from very early on. I remember analysing logfiles from our first days online, in late April 2005, and guessing at who was looking at the site based on their IP addresses. (When we had only about 50 viewers a day, and there are only a few hundred thousand web users, this was kinda a fun game to play. Ooh! That must be Dick logging on from his office!)

Analyzing web logs helped us figure out that we were in the wrong business. Our homemade tools ignored all our user-generated content and only ranked how different pieces of professional, commissioned content were doing. When we modified the tool to count all the traffic to the free homepages we were hosting, we discovered that our edited traffic represented less tha 10% of our total traffic. Had we not figured out we were in the homepage business, not the edited content business, we’d likely have gone out of business before selling the company.

Then again, religious tracking of our logs helped us detect porn and pirated software, and build efficient tools to eliminate them from the site. In retrospect, our obsession with removing content that violated our terms of service is probably what kept us significantly smaller than Geocities… and is also what kept us from selling our company for hundreds of millions of dollars, instead of for the merely obscene fortune we sold the business for. The logfiles giveth, and they taketh away.

I was looking at logfiles today for the best possible reason – as fodder for an argument with one of my closest friends. Nate had forwarded me an article by Tom Engelhardt about US airstrikes on suspected terrorists in Somalia. Nate argued that the article was a good way to get readers to understand the obscenity of US policy in Somalia – a proxy war, supported by anonymous airstrikes with a strong potential to kill innocents, possible only in a country where we either don’t believe there will be consequences for our violation of soverignity, or don’t care. I thought the piece was pretty good, but was old news – the US government has been increasingly fond of air power, and the complication that we often don’t know whether our strikes hit their targets is well documented, especially in Iraq and Afghanistan.

And besides, I argued, no one gives a damn about Somalia.

To try to make this last point, I looked at statistics from Google Analytics collected from my blog for the past year and expected to be able to show Nate that none of my most popular stories were about Somalia.

That turns out not to be true. The 17th most popular post this past year is an old post, titled “Mapping Somalia“. I’m a bit baffled as to its popularity – it comes up as the 3rd hit for “mapping Somalia” on Google, but comes up much lower for “Somalia map”, which I would expect is a more popular search query. And it doesn’t show up in the list of 100 queries that send most of the traffic to my site. (According to Google, the largest plurality of people visiting my site come in via search engines – 42% of the total. This probably isn’t true – Google Analytics undercounts traffic via RSS.) Another Somalia post – on an American of Somali descent being held prisoner in Ethiopia – also made the top ten. That’s far better than I’d expected.

I’d expected my blog stats to reflect the web at its worst – attention paid to those rare posts where I talk about tech industry stuff or silly, viral stuff. And there’s some of that – my very silly post about Nate and my attempt to build an outdoor hot tub is the sixth-most popular post, in part because Mark Fraudenfelder kindly featured it on BoingBoing. And my defense of Robert Scoble’s honor makes the top 100… though just barely.

Instead, what becomes really clear is the value of answering people’s questions. The most popular post for the past year is a chatty, technical post about Berkman’s stop badware efforts and the situation a friend had with a “this site may harm your computer” message from Google. It’s become quite popular with folks trying to figure out why Google thinks they’re malware. Other posts that might apply directly to people’s lives are popular as well – some thoughts on Facebook and privacy; posts on attempts by LiveJournal to remove fanfiction blogs, and the fanfic community’s responses; a post about treatments for diabetic retinopathy.

But it’s not just technical questions. Between writing about obscure African topics and lesser-known speakers at conferences, I’ve been unintentionally pimping myself to Google, which has done its level best to bring me new readers. Search Google for Ugandan rebel leader Joseph Kony and the sixth match is one of my rants, titled “Just How Crazy is Joseph Kony?”. This worries me a bit, as I’m far from expert on Kony, and it’s hard to understand the situation in northern Uganda without understanding this strange figure.

At least 30% of my most popular posts are the direct result of liveblogging TED, Pop!Tech and other conferences. That’s a useful reminder for me the next time I’m three days into TED and ready to walk away from the keyboard. There are thousands of people who end up at my blog not because they had any interest in Africa or international development, but because they heard that Jill Bolte-Taylor’s talk at TED was amazing and read my notes on her presentation. (She’s the #5 search leading to my site over the past year.)

Ultimately, what’s affirming about this exercise is that some of the more provocative, Africa-centric stuff I’ve written continues to get a healthy amount of traffic, sometimes years after its original publication. “Africa’s a continent, not a crisis” is now almost three years old and still gets a healthy amount of traffic. And my long post on Cute Cat Theory not only is the third most popular post on the site, but Google tells me that the average reader spent seven minutes on the page, which implies that they actually read some of the content before drowning in my verbiage. And I’m totally thrilled that several of my posts from TED Africa, including ones documenting debates in the African blogosphere, make it into the top hundred.

Of course, not everyone gets what they’re looking for. The #3 search that led people to my site this past year as “Sheila Kennedy”. I’m guessing very few of those people were looking for information on Harvard design professor Sheila Kennedy, who presented an interesting (if controversial) solution for lighting in the developing world at Pop!Tech. No, my guess is that they were interested in the Sheila Kennedy who appeared on Big Brother, who’d previously been a Penthouse model. (You’re on your own for finding those links.) Similarly, I suspect that the reason this story about attending a Turkish bath is so popular is that it includes the phrase “naked Turkish men” – the 24th most popular search query that leads to my site. Again, my apologies for anyone disappointed in what they find here.

The lessons I take from this? Be obscure. Write about stories that other people don’t write about. Write about brilliant people who aren’t well known to the web. And if you’re having problems getting people to pay attention to your stories on Somalia, it never hurts to put in the names of obscure starlets who’ve taken their clothes off for photo shoots.