As a web analyst, I have been playing around with Google Analytics for many years now and have increasingly enjoyed watching and waiting for new features that a)add better ability to gain insight about a web business and b) make my life easier!
Features such as custom variables and event tracking have been an absolute gift in terms of being able to understand who visits my clients’ websites, which features are interacted with and what value this delivers my clients.
Frustratingly of course, there are areas where things could just be a little bit better and that’s where (in all honesty) we get to have some fun by re-working the way Google Analytics delivers data by creating hacks and being creative with filters.
I can now setup alerts using Google Intelligence to email clients if their sites go down, if there is a sudden influx of 404 errors or if a keyword increases or reduces number of visits.
Even the simplest of features, such as ‘annotations’ make peaks and troughs in data far more explicable – server went down, untagged email campaign was pushed out etc.
One area of frustration is the lack of ability to see the regionalisation of search engines, particularly from an SEO perspective. By this I mean that it is all well and good to be able to see in Google Analytics that the Google Search Engine delivered visitors to your site.
Well congratulations, pat on your back, but where did these visits truly come from? Should you be optimising for Google.co.uk or for google.com or for google.fr?
Google Analytics has given us the ability to add custom search engines (method found here - e.g ['_addOrganic', 'google.co.uk','q'], to add the Google.co.uk TLD search engine.
However to add every single known TLD for Google search engines as well as others can be a tedious and time consuming process. Now throw in Yahoo and Bing and price comparison engines such as Pricegrabber.com and Kelkoo.com, you can see how the list quickly expands.
At last count, Google Analytics tracked 40 domains by default.
Mercifully, Brian Clifton has created a hack (this is not a freebie, but is certainly affordable) that I have been using that splits search engines out into regions.
Brian’s script adds (again, at last count) a further 265 domains. For further information see here.
Friday-Ad, one of the UK’s leading classified websites, is accessed by visits from many regions.
Formerly, when you clicked on the Traffic Sources/ Search Engines report in Google Analytics, you were presented with a similar view to the below image, where no TLD was given:
By using the script, non-paid search engines can now be seen as:
Of course now you can drill down and work out which engines and keywords are driving visits or revenue for example.
A great example of this is to understand which countries/territories are familiar with your brand name compared to those that are finding you by generic keyword.
This is gold dust if you are an international organisation. For Friday-Ad, we removed the view of keywords containing ‘Friday’ and traffic source containing ‘google.co.uk’. This gave us a massive insight into the main key phrases used by non-UK territories to find the Friday-Ad site.
Not to be confused:
It is important to remember that this report is a search engine, non-paid traffic source report. This is not the same as the Traffic source referring site report which shows TLDs for the domains.
I have often been asked by SEO’s why e.g. google.co.uk is showing as a referrer rather than a search engine and there are some good reasons for this.
Here are some examples:
- You actually have a link that visitors are following from e.g. google.co.uk – lucky you.
- Your site is being visited via igoogle.
- Your site is being visited via Google Image.
- Your site is referred via the Google Directory link (DMOZ).
- Clicks from your Google profile.
To find out the answer to this, you can use the exact referring URL hack which will capture the full referrer information and store it in the Visitors, User Defined field in Google Analytics.
Field A -> Extract A – Referral (.*)
Field B -> Extract B – (Leave Blank)
Output To -> Constructor – User Defined $A1
See below image to help.
This represents the data as follows where you can see the exact referral path to a page on your site:
Using both these solutions should give you a much clearer picture of the impact of the big search engines properties on your site in terms of regionalisation and reason.