Monday, 29 November 2010

What Google Analytics can't tell you

Like any web analytics package, Google Analytics is very good at helping you understand how visitors found your website and what visitors are doing on your website but not why they are doing it. To understand the "why" we need other tools like:

1) Site Surveys tools such as:
iperceptions (formally 4Q)
Kampyle
Google Consumer Surveys

2) Page Surveys tools such as:
Qualaroo

3) Live Chat
Live Person
Live Chat
Zendesk

4) Heuristic review tools:

Screen recording (yourself):
Camtasia
ScreenFlow

Heuristics from Nielsen Norman Group:
E-commerce heuristics
Non-profit
B2B heuristics
About us heuristics
PR section heuristics
Investor relations heuristics
Intranet heuristics

5) A/B and Multivariate testing tools
Google Analytics Content Experiments
Visual Website Optimizer
Optimizely
Adobe Test and Target
SiteSpect

6) Online usability testing:
Usertesting
Whatusersdo

7) Online usability feedback:
Usabilla Live
Usabilla Feedback
OptimalSort
TreeJack
ChalkMark

8) Online recruitment on your site for usability testing
Ethnio




Sunday, 28 November 2010

Google Analytics' Regular Expressions

Google Analytics' Regular Expressions can be used four areas:

1) In Google Analytics on-screen display filters:

















Filter Source:
facebook|twitter|youtube|myspace|stumbleupon|technorati|delicious|reddit|digg|flickr















2) In Google Analytics Advanced Segments:

Add this Social Media Segment to your Google Analytics Profile.
















3) In Google Analytics data processing Filters:














4) In Google Analytics Goal Configuration:







Here's some more information about Regular Expressions:
http://www.google.com/support/analytics/bin/answer.py?answer=55582&ctx=sibling
http://analytics.blogspot.com/2009/04/regular-expression-tips-and-tricks.html
http://www.regular-expressions.info/

Here's a great resource for testing your Regular Expressions before you deploy in Google Analytics:
http://ryanswanson.com/regexp/#start

Google Analytics IP address exclusion Filters

It may be important to exclude Visitors, Visits, Pageviews and associated data from Google Analytics for your organisation's offices, call centres and possibly business partners such as your web design agency and web marketing agency. To do this you can build a simple predefined filter:










or a Regular Expression filter ( You can learn more about Google Analytics' Regular Expressions here):










or a Regular Expressing filter for an IP range using the Google Analytics IP range exclusion tool :

Excluding query string parameters in Google Analytics

It's important to configure Google Analytics to only record your website's Pageviews that actually matter. If your website uses query string parameters it may be creating hundreds or thousands of unique Pageviews which do not actually exist. This clutters analytics, fills up table space and makes sensible analysis difficult. To prevent this data being collected you need to identify it and filter it out:


1) Check for query string parameters in Content > Top Content
2) Export your Top Content, use &limit=50000 suffixed to the Content >Top Content report
3) Export to CSV (does not work for CSV for Excel)
4) Check your Content for URLs to clean or remove in order to aggregate your data for better analysis and to prevent irrelevant Pageviews. You may need to discuss which parameters should be ignored in Google Analytics and which should stay in because they add value to your analysis.
5) Exclude any query string parameters, separated by commas in Edit Profile> Excluding URL query parameters


More on excluding query string parameters:

http://cutroni.com/blog/2006/09/21/google-analytics-configuration-mistake-2-query-string-variables/
http://www.google.com/support/analytics/bin/answer.py?hl=en&answer=55499
http://analytics.blogspot.com/2008/06/make-your-content-reports-more-useful.html

Exporting your Google Analytics data for analysis & cleansing

It's important to comply with Google Analytics Terms of Service and one condition is that you must not use Google Analytics to record any personally identifiable information about your visitors. Often, this rule is broken when URLs are posted using the HTML get function:

http://www.cs.tut.fi/~jkorpela/forms/methods.html

To check your Google Analytics installation is not breaking this rule, here's some checks you can perform:

1) Check for personal data in Content > Top Content
2) Filter for Containing “@”
3) Export your Top Content, use &limit=50000 suffixed to the Content >Top Content report
4) Export to CSV (does not work for CSV for Excel)
5) Check your Content for personal data
6) Check your Content for URLs to clean or remove in order to aggregate your data for better analysis

Back to Basics: Tip for exporting rows

How do I export more than 500 rows of report data?

More Google Analytics &limit hacks:
http://www.adwordsstrategy.com/get-more-from-the-navigation-summary-and-pivot-tables/

Edit Account Settings in Google Analytics

When you edit your account settings in Google Analytics you can do a number of things:

1) Check your Google Analytics UA (Urchin Account) number.
2) Change your Google Analytics account name. It is best to give it a meaningful name and prefix a number to the name in order to raise it to the top of your list (if you have many accounts).
3) Agree to share your Google Analytics data with other Google Products like Google Adwords Conversion Optimizer that relies on reading your conversion data.
4) Agree to share your Google Analytics data anonymously with Google and others via the Google Analytics benchmarking service.

Frequently asked questions for the Google Analytics data sharing options.

Saturday, 27 November 2010

Useful tools and reports to help check Google Analytics page tagging

Google Analytics' Achilles Heel is poor page tagging so what can be done to check how well tagged your website is? Well, here are some useful tools and reports:

Browser tools:
1) View Source in your browser and search "UA-"
2) Firebug for Firefox
3) Charles or Fiddler for Internet Explorer
4) Charles for Safari
5) Wasp (Free version)
6) Google Analytics Tracking Code Debugger

Free Scanning tools:
1) http://sitescanga.com

Paid Scanning Tools:
1) Wasp (Paid Versions) (http://webanalyticssolutionprofiler.com/)
2) Observepoint
3) REL Link Checker - You can use Relsofware's Web Link Validator Page Rules which are designed for evaluating pages against certain conditions and confirming the absence or presence of specific display text, tag text, links, scripting, forms, etc. generated by your code. Here's some more instructions on how to do this:



Paid Audits
1) ivantage's Google Analytics Audit !

In my opinion, the best tool for continuous auditing is http://www.observepoint.com/. Relsoftware's Web Linker Validator is the best tool for one-off analysis and for SEO audits, but to my knowledge you can't schedule regular audits. Our paid Google Analytics Audit uses Observepoint technology.

Google Analytics Reports:
1) The "Check Status" report in your Edit Profile setting performs a basic test of just the home page.

2) The presence of self referrals in Traffic Sources > Referring Sites report indicates a change in status of the Google Analytics cookie which may be down to domain transfer or simply poor tagging.

3) Check to see where your Google Analytics Tracking Code (GATC) is being resolved in the Visitors > Network Properties > Hostnames report.