Executive sponsors have become increasingly frustrated that web analytics initiatives fail to deliver any real business insight.
If left unchecked, this sentiment can lead to the reduction or, reallocation of budget and marks the end of hope for a resident web analytics professional.
So, how does this situation arise and what can be done to prevent it?
Common Web Analytics Pitfalls
1. Top down, not bottom up
More often than not, web analytics failure begins during the adoption process when the initiative is being driven from the bottom up. Even if deployment is successful, this approach nearly always results in siloed data usage by Marketing, Design or, I.T. departments and culminates with pointless debates during board meetings about conflicting data sources.
Jackson's Cult of Analytics is a great place to start learning more about how to build strong foundations.
2. Integration across the nation
Just having the “big cheese” buy-in to the web analytics program is not sufficient for the initiative to deliver value. Instead, the function needs to be integrated across the business to succeed. However, exercise caution not to align any one business unit too closely as the focus of web analytics resource can become marginalised and narrow in scope.
More recommendations on web analytics positioning and governance are available from Adobe Omniture's white paper available to download here!
3. Talent, tools and training (in that order)
"We bought a Ferrari but, only use it to go to the end of the driveway and back,” remarked one head of marketing for a global clothing manufacturer recently. After a frustrating nine months in web analytics, the lessons learned were straight forward enough but, ultimately proved quite costly for the business.
“The human skills needed to bring true value to an enterprise from web analytics are scarce, but they are more important than the technology involved.” Frank Buytendijk & Astrid Van Dorst, Gartner Group, April 2001
4. Measure everything and you wont find out anything
Ask a business starting out in online measurement what they want to track and it's fairly common to get the response, “Everything!” in reply. Resist the temptation to track everything and instead answer one question to define tracking requirements…
Why does the web site exist?
The overly simplistic question usually sparks internal debate about company objectives and helps to build out a measurement framework that can be used to inform tracking implementation. Create a measurement framework for your business using design principles and kpi definitions from Avinash Kaushik.
5. Manage ecosystem evolution
Introducing web statistics into an established business data ecosystem is problematic. The reports seldom match figures from other business intelligence tools and nuances in data collection methodology can easily undermine efforts to establish credibility amongst internal stakeholders.
To ensure long term web analytics success, follows a few simple steps…
Do not circulate reports too widely after initial deployment and instead focus on reconciliation activity for a small number of key metrics.
Build a data dictionary to describe each key web metric and highlight the difference in data collection method between other tools using similar metric descriptions (e.g. “orders”). Use it to educate the business and start in the board room.
Restrict access to reporting (at least initially). The number and variety of reports in most web analytics tools is enormous. Customise reports available in the platform interface to ensure users focus on the important stuff.
Target and train web analytics “champions” for each business unit.
These people will act as knowledge satellites evangelists and hopefully act as
a buffer from requests that come in from around the business.
Start any reconciliation exercise with the six sensible steps suggested by June Dershewitz and when constructing a data dictionary for the business have a look online for examples in your industry sector.
6. Web analytics just isn't enough
Don't be surprised if using web analytics alone is not enough to produce the step change in performance expected by the business. Statistics are great at explaining what happened but, crucially not why it happened. For best results, ccompliment web measurement with customer experience and observational analysis.
7. Avoid the data monkey trap
Shocking news! Most web analysts don't actually analyse anything. Instead companies pay these resources to pull reports and troubleshoot technical issues. So perhaps a change of job description is required. To avoid the "data monkey" trap try the following approach:
Allocate a maximum of 20% bandwidth to ad hoc reporting
and have this agreed by the boss - And yes, that's just one day a week.
Setup a ticketing system and to
prioritise incoming requests, monitor service levels and track how much
resource is spent reporting. Jira is a popular approach for many businesses
Identify the root cause behind reporting
requests and don't blindly agree to deliver them - Easier said than done.
Automate at least 80% of standard reports using
the web analytics email scheduling tools or, by building pre-populated reports using API's. Most vendors have their own solution and their are really good third party ones like Omniscope and Excellent Analytics
Write a business case and secure budget to help
support reporting activities using graduates or, outsourcing companies - Don't wait until it get overwhelmed
Train people on how to pull reports themselves -
Its all about self-service these days
Interested in learning how other companies use web analytics resources? Visit Econsultancy who in partnership with Lynchpin published some interesting recent survey results.
8. Don't expect return-on-investment without change
The ultimate acid test and competitive advantage for a data driven company is generated by their ability to implement change based on insights generated from web analysis. Therefore, it goes without saying that return-on-investment can never be achieved until the development road map allows for sufficient resource to deploy code changes and implement recommendations in a timely fashion.
Build robust financial business cases to
support web analytics code development
9. Crash test, dummies
Never release web analytics code changes onto Live site without thoroughly testing it first on a Test environment. Failure to adhere to basic quality assurance procedures can have devastating short term effects on data quality and generate longer term trust issues with internal stakeholders. Each Site or situation may require a slightly different approach. We perform the majority of single page audits with Charles and undertake large automated Site-wide audits using ObservePoint. For cookie specific audits and competitor technology information we recommend Pikslme.
10. Responsive isn't just about web design
The internet is constantly changing and measurement techniques used to track it continually evolve at the same lightning pace. Stay sharp and find out how the latest technological development can help improve the existing web analytics program. Put sufficient time and resource into keeping on top of industry developments and in practical terms this means:
thirty minutes each day to read blogs,
articles and other technology news. To save time, set up RSS feeds and
have content delivered directly and don't forget there are also great
pod casts so that you can learn while you listen. A favourite of ours is
found at Beyond Web Analytics
that completes the top ten gotcha's to avoid when managing a web
analytics initiative. Thanks for reading this article and now it's your
turn to share.
About the Author
Alex Brown is a Digital Analytics and Site Optimisation expert who works as an independent freelance consultant. The opinions shared in this blog are based on personal experiences gathered over a decade of data crunching and technology evaluation. The author makes no attempt to be grammatically, politically (or otherwise) correct. Spelling was never a strong point and for practical reasons, not all vendors in the market are referenced in the article - no hard feelings.
Related Products & Services
With in-depth knowledge of both the market place and major technology vendors, allow us to help select, negotiate, support and manage web analytics implementation for your business. Contact Us for a confidential and informal discussion to see how our solutions can improve your business today.
|Resources||(In Alphabetical Order)|
|Adobe Summit||Web Analytics Conference|
|Automated Reporting||Excellent Analytics|
|Cult of Analytics||Steve Jackson|
|Customer Experience Analytics||ClickTale|
|Data Dictionary Example||Office of Disability and Adjudication Review|
|Data Governance||Omniture White Paper|
|Defining KPI's||Avinash Kausik|
|Eye Tracking||Gazehawk |
|Measurement Model||Avinash Kausik|
|Online Usability Testing||FiveSecondTest|
|Preferred debugging tool||Charles|
|Podcasts||Beyond Web Analytics|
|Reconciliation how to guide||June Dershewitz|
|Testing||Visual Website Optimiser|
|Xchange Analytics Conference||Semphonics|
Image courtesy of M Bartosch / Freedigitalphotos.net