Go to content Go to navigation Go to search

· Statistics ·

Matt Magain and I had a quick chat about web stats today, and he suggested I share which statistics I look at each month for ANTaR.

These are the stats I report on to our board's web sub-committee, staff, and database/IT volunteers to give them a quick snapshot of how we're doing:

In addition to the stats I report on, I also track other visitor stats such number of 404's, visit duration, percentage of new visitors, just to see if anything remarkable is happening. I also test our google.com.au ranking on a few key words and phrases, and analyse our email newsletter statistics in detail. I should probably also check the number of people reading our content via RSS, but I don't :(

The tools I use to do all this are AWStats (which comes free with our hosting package), and Google Analytics (which is also free). Email stats come from Campaign Monitor and are quite detailed for html emails (as opposed to text emails). I have had access to packages such as NetRatings at previous organisations I've worked for, however I feel that I get almost everything I need from these free services ... which is lucky because NetRatings costs lots!

P.S. I usually spend about half a day every month doing all this :)

Monday December 3, 2007

Categories:

  1. I do a similar set for Amnesty. I don’t give out raw numbers, just trends, it saves confusion. I also have activism rates (thousands of form submissions per month, number of subscriptions etc). Also, we are big enough and old enough now that the ‘most popular’ bits -other than individual pages – don’t really change much.

    Don’t forget that email open rates (raw numbers) are pretty worthless (trends are useful), click-throughs are really useful though.

    I do track feed subs and click rates (using feedburner) and do point sampling on other live stuff, but i am a real numbers geek. This is really useful though as we get more eyes through syndication and RSS than through visitors to our main web site.

    Also about half a day a month – and it is this week – so do you wanna swap samples?


    — Mark    3 December 2007    #

  2. Thanks for your response Mark. It’s great you can also send out activism rates … unfortunately I need a systems change before I can access those numbers easily, but I’m working on that (btw, the Sea of Hands numbers I mentioned would go in this category).

    I’m a relative newcomer to email rates so I’m still working out how to use them best. There’s a great post here from Campaign Monitor about open rates: http://www.campaignmonitor.com/blog/archives/2007/01/all_about_email_open_rates.html. I guess I view open rates as more useful for our database volunteer (to measure the accuracy of our email list) rather than how great our email newsletter is.

    I would love to swap samples, let’s chat offline :)


    — Priscilla    4 December 2007    #

  3. Hey P. I had a similar report when I was at a “previous NGO employer” (although I hate to admit, I think it largely got ignored). I used ClickTracks and Google Analytics – we had ClickTracks before Analytics became free, and the difference in unique visitor reporting was significant enough that I had to use both for different things.

    From memory, the core stats I tracked: * Total visits (compared month on month over two years) * Total pageviews (compared month on month over two years) * One off gifts – Dollars raised (compared month on month over two years) * Regular gifts – Dollars raised (compared month on month over two years) * Top 20 pages * Top 10 PDFs (a heavy emphasis on PDF publications made this worth tracking) * Top 5 referrers * % search referrals – this increased from around 14% to 35% in my time – an indicator that SEO was working * Top 20 search terms * Email open rates (unique opens) * Top-clicks on email (a monthly email)

    I think there were one or two other stats that I’ve forgotten. Used to take me 3-4 hours to complete each month.

    I didn’t go into daily tracking except when we wanted to specifically focus on the impact of a big media splash or campaign launch. We also used to track specific campaign action success, but this was a rare occurrence with that particular NGO.

    I empathise with Mark on the idea of using trends rather than raw numbers – I did use raw numbers, and thankfully the improvement was significant enough that it worked to my advantage ;) However, I did have to explain that the drop in page views to visits was a result of good IA, not falling interest :\

    However, I found that providing access to management of Google Analytics for one of our campaign sites really helped them focus on numbers – unfortunately to the detriment of perspective – the result was spending many $$ on advertising campaigns to drive traffic, rather than investing in the experience. More money was spent on advertising than the site itself, which I think is indicative…

    Although all of these “number” metrics are good – they don’t necessarily capture some other important aspects of the site.

    P.S. the captcha on your comments form is nearly impossible to read/decipher :(


    — Grant    4 December 2007    #

  4. Awesome, thanks for sharing P!!


    — mattymcg    4 December 2007    #

  5. Thanks Grant, a very interesting response :)

    I’m intrigued about your online advertising spend being more than online experience spend … what a shame … although this doesn’t really surprise me!

    P.S. yeah I know the captcha is really annoying but it’s the only one that I could find that comes with Textpattern, and I was getting a bit of comment spam, so it’s the lesser of two evils :(


    — Priscilla    4 December 2007    #

Commenting is closed for this article.