Web analytics is a necessary discipline when working in online marketing. It’s what proves whether your initiatives, campaigns or designs are working or if they’re a total flop. With the wide adoption and ease of use of Google Analytics, it’s easy to take web analytics for granted. So easy in fact that those who don’t have an analytics background are taking the data as-is with no further thought behind how it was collected or why it is the way it is.

Web metrics can easily be skewed leading to the wrong decisions if certain principles aren’t adhered to.

The following is a manifesto for anyone responsible for web analytics for their organization.

web-analytics-manifesto

I <insert your name here>, who is responsible for analyzing and disseminating web metrics vow to adhere to the following web analytics principles and guidelines:

  1. Interview business stakeholders for their web analytics expectations and requirements before launching a new website or starting a web analytics initiative
  2. QA new websites thoroughly to ensure all pages are tagged appropriately, making certain data being collected is as accurate as possible
  3. Stay involved throughout the evolution of a website as new pages are added or updated making sure they are all tagged appropriately
  4. Question the accuracy of any sudden spikes and dips in traffic or activity
  5. Investigate the cause of all traffic or activity spikes and dips
  6. Visit the site’s web pages as I am analyzing the data to provide context
  7. Keep up to date with all offline and online promotions, marketing campaigns and publicity that could affect website traffic
  8. Hold back distribution of web metrics reports until all data inconsistency issues are either addressed or the cause has been identified
  9. Do my best to only supply web metrics that are actionable
  10. Do not send dashboards to key decision makers without being accompanied with a thorough analysis
  11. Do not assume stakeholders will analyze or see the data in the same way I do
  12. Segment the data as best as I can to provide context regarding activity on the website
  13. Distribute web metrics not only to executives but to content developers, designers and web developers to see the impact of their work
  14. Train key stakeholders on web metrics basics or provide them with succinct training materials
  15. When appropriate, combine web analytics with data from other sources such as the CRM, forum systems or blog platforms
  16. Remember that web analytics is only part of the equation and to do my best to push for more qualitative data from surveys or interviews
  17. Stay up to date in my training and continually improve by learning from other practitioners

=============================================================

If you think I missed any, I’d like to hear from  you.

Hey there, I'm Marc!

I'm the Director of Growth Channels @Hootsuite. I've worked in digital marketing since 2000 for agencies, startups and companies like Electronic Arts & SAP. Thanks for checking out my site where I share digital marketing strategies on how to increase website traffic and generate more revenue.

Comments

  1. S.Hamel says:

    Good start Marc! I like it!

    (You should use a numbered list so people can provide feedback more easily)
    #1: Don’t ask stakeholders for their web metrics requirements, ask them about their business objectives. As an analyst, it is your job to figure out which metrics will measure outcomes – and speak the business language.
    #2: Not only QA new websites – put in place a continuous auditing process because once the site launch, your tags also needs to evolve (and be checked)
    #9: Metrics themselves are not actionable, recommendations are. As an analyst you don’t want to provide metrics & reports, you want to provide insight.
    #15: Web Analytics is not only clickstream data from Google Analytics – a true analyst will always seek other sources of information that will offer another angle (just like in a scientific experiment). Be it voice of customer, performance monitoring, CRM, call center, back ends, competition – you should always use more than at least two sets of complementary data. My preferred ones to start with: clickstream, voice of customer, performance monitoring – because they are fairly easily accessible and more under the control of the web team (as opposite to getting info from other departments such as IT, customer service, etc.)
    #16: quantitative AND qualitative data are both important
    #17: Some references: the Web Analytics Association, Web Analytics Yahoo! forum, Twitter #measure, UBC Award of Achievement in Web Analytics, eMetrics conference

    I have spent the past two years working on the Online Analytics Maturity Model where I address those issues (and more). See http://immeria.net/oamm

    Stéphane Hamel
    Online Analytics Advocate at immeria.net

  2. Ruy Carneiro says:

    – Test … Test … Always be testing 🙂 Great post!

  3. Very useful post. Google Analytics can definitely provide a lot of interesting data (which isn’t always entirely accurate when compared with more real-time analytic suites) and it makes a great traffic counter, but knowing how to use it to create a better web experience for visitors and to increase a website’s sales is much better!

Leave a Reply

Your email address will not be published. Required fields are marked *

*