Thursday, February 10, 2011...8:30 am

Blogging lessons from last year’s Google Analytics

Jump to Comments

Warning: intense web stats geekery ahead

It’s been a long time coming from the last stats outing – so much so that it’s not exactly an annual round-up, as I’m several months late. But here, for your geeky pleasure, is a run-through of what I’ve learned from Google Analytics about the past 12 months’ traffic.

Traffic has dropped from last year

And here I was thinking how influential and important Freelance Unbound was getting. Oh well. Many reasons for this – but in the main, here’s how I’ve screwed up:

  1. I’ve stopped updating so often
  2. I’ve stopped engaging with my peer group (other bloggers/journalists)
  3. The focus shifted to a more serious, long-form academic blog – when it wasn’t being trivial
  4. The posts became less immediately helpful

1) Ironically, being more successful (busier, moving house, new exciting teaching work) means I have less time for blogging – how I miss the recession. Also, getting a dog has hit my blogging output (but I do get out more).
Key lesson: it’s a slog, but you have to keep it high on your priority list.
2) I spent the first year of serious blogging getting very involved with the “community” – reading, commenting, arguing. You really do need to keep up with the others in your field. Ignore them and they’ll start to forget about you.
Key lesson: spend at least a chunk of your time commenting and reacting to other people online, as well as working on your own material.
3) As Freelance Unbound began to get a higher profile, I started taking myself more seriously. Big mistake. I felt I had to make each post a serious and substantial addition to the media debate, which is frankly intimidating (for me, and probably for readers).
Key lesson: shorter, punchier material tends to be more immediately attractive. And if you start to write much longer, in-depth posts, the chances are you’ll get intimidated at trying to match them every day and your posting schedule will start to drift. Even if you produce long, reflective pieces, try to mix them up with shorter items that are easier to read (and to write)
4) Adam Westbrook has an excellent piece here about how the mistakes he’s made and how he turned his blog into the undoubted professional powerhouse it is. Mainly by not publishing anything that would not be useful or valuable to his audience. Do I do this? Clearly not.
Key lesson: From my own freelance success advicebe useful.

Top posts

Aside from the home page (and discounting bio pages and pages with student notes), which content gets the bulk of my pitifully small traffic?

  1. Tales from the trade press: Clinique’s press office hell
  2. Is People Per Hour any use?
  3. Andy Davies, the unseen casualty of the Jonathan Ross affair
  4. Facebook obsession murder
  5. UK student riots – a different perspective
  6. 6 Laws of CMS: Lessons from a web site launch
  7. Looking for a Delicious replacement?
  8. UK journalism jobs – a graphical guide
  9. Writing style: advice to journalism students
  10. Why is “I” the most important letter in journalism?

Items 1, 5, 6, 8, 9 and 10 get their traffic largely from referral – the Poynter Institute web site picked up on Clinique and gave me an instant traffic spike, the lovely FleetStreetBlues flagged up the pieces about journalism jobs, student riots and student writing style, and the essay about the “I” of journalism. And (the Guardian’s head of tech, no less) referred kindly to the Laws of CMS. also picked up the piece on jobs (mainly, I suspect, because I used their classified ads as my data source).
Key lesson: make and keep friends in the online world (oh, yes – see lesson [2], above).
The others are in the top rankings because of my carefully crafted SEO. Andy Davies, People Per Hour, Delicious and Facebook are all awesome traffic generators.
Key lesson: pimp my headlines with celebs and hot tech companies. No matter what the content is.
Another key lesson is something I discovered last time I looked carefully at my stats – a big post with a tsunami of referred visitors will give you a traffic spike that disappears almost immediately. But search optimised content will give you a steady flow of visitors, day in, day out – and that’s a much better in the long run.

Traffic sources

1 (direct) / (none) 3,102
2 google / organic 3,008
3 / referral 1,417
4 / referral 569
5 / referral 447
6 / referral 337
7 feedburner / feed 245
8 / referral 236
9 / referral 212
10 / referral 149

At the top, a roughly even split between people who know where they’re going and come here anyway and organic Google search.
For the rest, FleetStreetBlues is still far and away my most important single referring site – and they even went and voted me 9th best UK journalism blog at the end of last year. Bless.
Still useful are Twitter and (though isn’t nearly as powerful as a traffic source as I thought it would be, interestingly. Maybe its readers aren’t that interested in blogs, or maybe the site itself is too rich in content.)
That one mention on Poynter puts the site in my top 10 referrers, but that will only be for this year – I won’t see any of this traffic recognised in 12 months’ time.
My friend the Wartime Housewife, on the other hand, will probably still be up there in the rankings, even though her content is light years away from mine. It just goes to show that an intelligent, curious readership can be transferrable.
There’s also some feed reader traffic, which means it still isn’t dead as a means of distribution, as some suggest it will be, but then I’m dealing with journalists, who are probably the last holdouts on RSS.

Search engine source

Is Google under threat? Let’s have a look.

So – no. No matter what Microsoft has to say, Google is search. Until the whole online model changes again and we only care what our Facebook friends read. SEO is GEO.
Not sure what that generic “Search” slice is – probably it’s search from within Freelance Unbound itself, which means it’s probably me looking for relevant posts while I’m lecturing and not logged in.

New visitor vs old friend

  1. New Visitor  – 60.22%
  2. Returning Visitor – 39.78%

Is this good or bad? I have no idea. What is the ideal breakdown between new visitors and regular readers? How do you convert one to the other? Suggestions welcome. And a project for 2011…

Loyal vs disloyal

  1. 60.22% have only ever visited once
  2. 6.55% have visited twice
  3. 5.31% have visited 201+ times

Apart from the last, which is probably me obsessively checking my posts without logging in, a whole lot of visitors only see one or two posts. So – those one-off visitors don’t stick around. I’m clearly not as sticky as I’d like to be. Another project for 2011…

Length of visit

67.59% – 0-10 secs (oops, wrong site)
7.38% spend 10 minutes or more on the site. (Again, some will be me, obsessively reading my own posts. Some will be students staying on a tutorial page while I rant about grammar)
In between is where you want to be, and there’s less of that than I’d like. But is most of the web like this? According to Jakob Nielsen, yes – you have a very short space of time to grab your reader and get them reading – and even then they often bounce away quickly. Watch out for more usability experiments on Freelance Unbound.

Depth of visit

Most people look at a couple of pages (55.79%)
A decent chunk looks at four pages (11.32%)
A few weird obsessives (or keen students, possibly) look at 20 or more pages a visit (1.68%)
This chimes with the findings above – I clearly have a few “real” readers and a lot of transient traffic that doesn’t necessarily mean anything. Or not. Whatever.


Again – a few stalwarts (Soilman, early reader Bill Bennett) vs many lone comments from ships passing in the night. And many from me giving lovingly crafted personal answers to my valued readers (community building in action!).
Of most interest and concern is the drop in total comments from 282 in 2009 to 229 in 2010. Fewer readers, fewer comments; less engagement, fewer comments. Come back – I miss you already.

Search engine terms

What do people look for to find me? Some expected search terms – and some weird things that you’d never expect.

1 people per hour 250 (my careful attempt at SEO success worked!)
2 rotterdam 207 (weird stray photo of Rotterdamstill pulling them in)
3 freelance unbound 175 (my loyal army of stalkers)
4 facebook obsession 118 (cunning SEO…)
5 andy davies radio 2 89 (…and here…)
6 andy davies radio producer 52 (…and here…)
7 people per hour review 45 (…and here…)
8 sophia myles 37 (random post about vampires)
9 facebook murder 32 (cunning SEO…)
10 peopleperhour 28 (…and again here…)
11 raymond chandler 23 (excellent writing advice)
12 andy davies 21 (SEO still working, then…)

Mostly, the search terms drive traffic to the most-read posts – but not always, as that stray Raymond Chandler search term proves. Which just goes to show that all content has value, even if it’s not the busiest.

Weirdest search

Finally, for reasons best known to themselves, someone searched for amateur at and found my post about art magazine site Which has nothing to do with sex, honestly. But there’s a thought for boosting my profile online…


  • Amateur cumshots is the way to go, my friend. Sod all this journalism malarkey.
    [See what I just did there? Boosted your ‘amateur cumshots’ search traffic still further. AND again…]

  • What an incredibly useful article – I must do those stats on my own site. Funnily enough, my sister sent me an email today complaining about my inclusion of body parts in the titles (incidentally the stats on that were 3 out of the last 26 which is hardly overkill) and asked me if I was doing it to drive traffic. I wasn’t, but as you say….

Leave a Reply