As you’re all aware, on April 1st 2012, more than 3 years of blogging at http://www.sublimeoblivion.com/ (S/O) came to an end. For the majority of that period – to be precise from May 13th, 2009, to March 31st, 2013 – I had an account with Google Analytics that provided me with very detailed states about the blog: Where visitors came from, how long they stayed for, what they shared on Facebook, etc.
On some occasions, the stats are exactly what one would expect. Other times, they are unexpected and fascinating. For instance, more people from India visited the blog than from Russia, one of its main subject matters. The Kremlin Stooge was my third biggest referral source of all time, even though he only came on the scene in July 2010 (thanks Mark!). Almost ONE THIRD of all-time visitors landed on a single post!
As I’m going to lose this data as soon as I find a new use for the old domain, I decided to save it all and lay out the interesting bits in this post. I hope I’m not the only one who will find this “meta-blogging” interesting.
Visitors, Visits, Views
From May 13th, 2009, to March 31st, 2013, there were a total of 760,581 visits to my blog. (As I recount in the Blog History, the all-time figure would be about 35,200 higher, as before that date I blogged at http://darussophile.blogspot.com/ from January 2008. The total number of visits to all my blogs before the April 1st transition to WordPress.com is therefore about 800,000). The graph below shows the dynamics. You can click on it to get a higher resolution, as with most of the graphs here.
As you can see, most of the period was one of slow growth, until it began to rapidly take off from about mid-2011. It reached a plateau of more than 2,000 daily visits by the end of the year. Then, in early March, this figure plummeted by two third. This was the result of a pharma hack that replaced links to my blog on search engines like Google with links to online shops selling viagra and the like.
This made perfect sense, as by that time, the vast bulk of my readers were coming from search engine results.
This is a graph of monthly visits. It is much easier to see the general trends, as day by day fluctuations are smoothed out. Overall, during this period, I had an average of 713 visits per day, and a maximum of 2,837 visits on November 8th, 2011. That date coincided with a translation of my post BRIC’s of Stability by Inosmi. In fact, as we’ll see later, most of those spikes in daily visits were linked to translations by Inosmi.
The above data shows the numbers of unique visitors, superimposed on visits (each visitor is at least one visit) and pageviews (each visit is at least one pageview). As the blog’s popularity expanded, the pageview per visit, and the visits per unique visitor, decreased, as a greater bulk of my traffic shifted from regular readers to people who came in via search engine.
Which continents did Sublime Oblivion readers come from?
Nothing surprising here.
Which countries did Sublime Oblivion readers come from?
Here is a world map.
Here is a list of all the countries that graced S/O with more than 1,000 visits.
|Country/Territory||Visits||Pages/Visit||Avg. Visit Duration||% New Visits||Bounce Rate|
|United Arab Emirates||5,200||1.19||0:00:49||89.35%||89.40%|
|Trinidad and Tobago||1,734||1.75||0:02:50||23.82%||71.57%|
Below is a graph of visit dynamics from the top 5 countries by total visits.
The US, UK, and Canada all have broadly similar dynamics, as would be expected of Anglophone nations with their respective populations. The other two are better explained with another graph.
Taking out the US, we can identify more details from the other countries. In the beginning, the UK clearly has a consistent lead; however, visits from India begin to multiply from around May 2011. Though I’m tempted to ascribe this with the blog publishing post with titles such as The Century without an Indian Summer or Why China Is Far Superior To India, which are basically red banners to Hindutva nationalists, the one came too early and the other too late to explain the leap.
I’m just going to assume that it’s due to the conjuncture of (1) the vastly increased secular popularity of my blog, which increased about five-fold in 2011, and (2) growing Internet penetration in India, with gross number of users surpassing Britain’s, and hence logically, them drawing even with the UK on numbers of visitors to my blog.
Russia follows a totally different dynamic. It is not an Anglophone nation, and foreign language skills are poorly developed. For obvious reasons, Russians are exceedingly unlikely to find my site through Google.ru or Yandex. As such, its “constant audience” at S/O is a lot smaller than for the US, UK, India, Canada, and even Australia. A big chunk of Russians stumble onto the site whenever Inosmi, a very popular Russian website that translates articles from the Western media, translates one of my posts. In the immediate aftermath, there is a sudden flood of Russians from Inosmi to my site, creating the spikes you see in the graphs above. Occasionally, one or two high profile Russian bloggers would then repost one of my translations from Inosmi, resulting in secondary spikes.
Which cities did Sublime Oblivion readers come from?
Here is a world map.
The Big Three are Moscow, London, and New York. Quite appropriate, really, considering these are the capitals of the three countries I’ve lived in.
Maybe I should go live in New Delhi or Sydney next. 🙂 I’m also pleasantly surprised to see Berkeley, a very small city of about 100,000 people, making the Top 40.
The US, UK, Indian, and Russian Audiences
64% of Russian visits, or 19,805 of them, came from Moscow, a reflection of the country’s super-high degree of centralization in the capital: English-language speakers; Western expats; high rates of Internet penetration (in 2010, 73% in Moscow vs. 38% in the country as a whole); etc. Another 3,067, or 10%, come from St.-Petersburg.
The top 10 Russian cities.
|City||Visits||Pages/Visit||Avg. Visit Duration||% New Visits||Bounce Rate|
If Kyiv were in Russia, it would be third (about 1,000 visits), and Minsk would be seventh (about 300 visits). Unsurprisingly, Ukraine, Belarus, and Kazakhstan follow the same patterns of visiting as does Russia; this is not surprising, as there are plenty of Russian speakers there who read Inosmi.
With the exceptions of the capital Kyiv, and West Ukraine’s cultural capital Lviv, you can almost demarcate the borders between “Blue Ukraine” and “Orange Ukraine” just from a map of who visited S/O.
Ultimately, the site’s low level of visibility on Runet is not a concern. If it was, I’d be blogging in Russian.
As expected, there is a close correlation between the population of a US state, and the number of visits I get from them.
People from more educated, richer, and connected states taking a greater interest in S/O. In total visits per 100,000 people, the District of Columbia has a clear lead (818), which I excluded from the graph above. It is followed by Washington (202), Connecticut (151), Vermont (145), New York (142), Virginia (127), Massachusetts (119), and California (111). States where people cared least about S/O included South Dakota (39), Louisiana (36), Arkansas (34), West Virginia (34), and Mississippi (28). It is worth noting that Washington and Virginia in particular have high concentrations of political/security and foreign analysts, precisely the types that are most likely to stumble across a site like S/O. All in all, none of this is in the least surprising.
The top 20 US cities.
|City||Visits||Pages/Visit||Avg. Visit Duration||% New Visits||Bounce Rate|
Visits from the national components of the UK were perfectly in line with their populations: England – 88%, Scotland – 9%, Wales – 2%, Northern Ireland – 1%.
From a cities perspective, 25% of UK visits to S/O originated from London. But many other cities got a share of the action. It was interesting to see Preston with 2%, the rather unremarkable town of 100,000 where I lived in the UK.
The most S/O-friendly Indian states were Delhi, Maharashtra, Karnataka, Tamil Nadu, and West Bengal. Above is a map by cities.
What characterizes the general S/O audience?
This is a graph of the language settings visitors had on their browser per visit. That Russian (23,851) is lower here than the number of visits from Russia (30,781) implies that perhaps 25% of the visits from Russia were made by Moscow-based expats. Which seems reasonable.
Man, who still uses Explorer these days?
The average duration of engagement was 2:11, but as shown above, the distribution is highly uneven. On most sites, including mine, most people leave very quickly.
Who were S/O’s best referrers?
Everybody who gave me more than 250 visits is in.
|Source||Visits||Pages/Visit||Avg. Visit Duration||% New Visits||Bounce Rate|
What were S/O’s most popular landing pages?
Every landing page that got more than 1,000 visits. (Needless to say, all of those links are broken, given the site change).
|Landing Page||Visits||Pages/Visit||Avg. Visit Duration||% New Visits||Bounce Rate|
I find it astounding that, counting only proper posts, more than a third of all visits to my blog have been to Top 10 Most Powerful Countries In 2011. Pareto is spinning in hiss grave. The second most popular post, with a mere 3% or so of 750,000 odd visits, is Top 10 Most Useful Languages. I think I’m beginning to see a pattern here, but I can’t quite grasp it… Oh wait, I get it! People like to read opinionated lists!! Who could have imagined? Okay, I’ll make more of them! 🙂