Country specific Search Engines in organic traffic with Google Analytics
One of the problems to solve while studying the keywords bringing traffic to your site is Google Analytics not specifying the ccTLD of the search engines at non-paid organic traffic report.
For example, a certain keyword came through Ask but you no way to know if it was ca.ask.com, or uk.ask.com. Another came via Google but which one: google.au, google.co.za, google.com.ph? This is particularly important when working international multilingual SEO campaigns.
Imagine what a difference could make in your SEO strategy discovering that a significant portion of your traffic comes via country ones like google.es, es.bing.com, es.ask.com, more local ones like terra.es, ozu.es or more specific ones like images.google.es, blogsearch.google.es (all from Spain) but using English or French keywords.
This quite sure reveals that part of your traffic as tourists (seasonal segment) or non Spanish residents (distributed along the year) living in Spain. High level of granularity segmenting types of visitors is what gets you closer to success.
We have two different problems here, get more detail at the list of local or country specific search engines GA identifies as organic and add the ones not identified.
Tracking search engines in detail
Filters will do the magic here but first create a new profile, it is always a must while experimenting new and advanced Google Analytics filters.
Then we need two filters.
Include only organic traffic filter
- create a new custom filter, 'Include only non paid organic'
- check 'include'
- filter field 'Campaign Medium'
- filter pattern 'organic'
Don't forget to add this filter to the brand new profile created previously.
Search Engines filter
- create a new custom filter
- check 'advanced'
- Field A -> Extract A : Campaign Source : .*
- Field B -> Extract B : Referral : ^http://([^/]+)
This regular expresion means: everything from 'http://' till the first '/'
- Output To -> Constructor : Campaign Source : $B1
- Field A Required : Yes
- Field B Required : Yes
- Override Output Field : Yes
Again, don't forget to add this filter to the new profile.
Before finish check the filters order in the profile to have 'Include only non paid organic' filter first.
Derick Ng did something similar but I wanted to take it one step further. Sure you can improve those filters or find some other interesting uses, I'm not an Analytics geek. Ideas welcome.
Let it run for several days and Voila! there we go, check Keywords report selecting 'source' in the drop box and this is what you will see.
Funny discovering sphin.com, talktalk.co.uk or gigablast.com as organic sources although they are not included at Google list. Some portals offer 'enhanced by Google' search like www.interia.pl (appearing later at GA as www.google.interia.pl) so maybe this is the reason.
Adding new Seach Engines to Google Analytics organic traffic
Not all search engines traffic counts as organic. This is a list of search engines Google Analytics identify as organic. Visits from any other outside the list are treated as referral.
To double check if there is any update grab ga.js and look for something like 'google:q,yahoo:p,msn:q,bing:q' (engine-name:keywords-parameter) to discover the whole actual list.
What if you would like to have some more search engines into Google's list as organic or reset the whole list having only the ones you want?
Add to tracking code pageTracker._addOrganic("search-engine-name","query-string-variable"); before _trackPageview(); per any of the engines you want to be included as organic. The way to tell Google Analytics to ignore it's default search engine list is adding pageTracker._clearOrganic(); before the desired ones.
pageTracker = _gat._getTracker("UA-xxxxxx-x");
pageTracker._clearOrganic(); [only if you want to reset Google's list]
Now you have the ability to enrich your keywords data adding a new layer of detail to organic traffic. Beyond that you can play with custom reports or any other Google Analytics feature to mix regional search engines, country and language in the same equation and get more insight from it.