Top 10 Practices of Successful Web Analytics Organizations
What are the key ingredients for building a successful Web analytics organization? You can be a great individual analyst, but for your team to be successful, you need a support structure to turn your recommendations into actions.
A successful Web analytics team influences the entire organization to become data-driven and provides measureable optimization opportunities. Following are 10 best practices that help create a successful Web analytics organization to support the team’s success.
It all starts with a good analytics implementation. You need a solid foundation to build both results and the trust of your business partners. If you or your partners don’t trust your data, then your partners can’t trust the analysis or your recommendations. So, how do you get a good implementation?
Control of Tags
How much control do you have over your analytics implementation? Analytics tagging must be updated constantly to keep up with site changes, fixes, and new optimization opportunities. It is important to have the developer resources that can make changes quickly. Adobe Tag Manager is an excellent solution that allows you to make quick changes to data collection outside of the normal development cycles.
Tight Integration with Reporting Systems
Once you have your development process in place, you must be able to validate the data’s accuracy. If you can easily compare systems, you can more easily identify issues quickly and build trust in the data.
Your Web analytics revenue data is unlikely to perfectly match your finance revenue numbers, but the variance should be consistent. Monitoring that variance often will alert you to any issues with the implementation.
Any variance with your backend data should be well understood and documented. It’s common for the Web analytics system to have a 2-3 percent variance with the finance systems, but any more than that might be due to a difference in how you define revenue.
A good “data dictionary” will define all the variables and important caveats. What is the definition of revenue? Does it include gift cards or discounts? This should be clearly defined so that your business partners have confidence in the data and know how to interpret it.
The data dictionary is often a variable-by-variable guide, explaining what each variable captures and how that variable can be used for analysis. No implementation is perfect, so you may want to flag variables with incomplete data or call out any caveats, such as dates when the variable was launched or issues occurred.
An “analytics events” calendar can help analysts understand anomalies in the data due to sale events, code changes, or outages. Also, you might want to make an intranet or document repository available to your business partners that contains links to the data dictionary and other relevant documentation.
Venue for Sharing Ideas and Successes
Great analysis is useful only if it’s used to improve performance. To this end, successful organizations hold roundtables of business users or analysts to share successes and recommendations for improved measurement. Typically, this is done by functional area, such as marketing or merchandising. This also helps the analytics manager understand how the business users employ the data.
Further, sharing successes with executives helps build support for additional analytics projects and resources. Building out a document repository for shared analysis is an important part of this process.
Strong Understanding of Business Processes
Learning how your business partners do their jobs gives you better insight into the types of data and analysis that will make them successful. It also helps you identify what types of measurement improvements will lead to improved reporting.
The goal of a great analytics organization is to feed the business users exactly the right information at the right time to make informed decisions that will lead to higher conversions. When working with business partners, you must know what actions they will take, not just what data they need.
Early Involvement with Projects
Before a new site feature launches, define the measures of success. Is a site feature going to increase conversion or visits or units per transaction? A true data-driven organization asks these questions before the project starts. Early involvement ensures that the implementation is updated to capture the key metrics.
The analytics team should also play a key role in developing the business case to create a new site feature. Additionally, you should ask how business users will optimize a new site feature. For example, for a new search engine, you might want to monitor search terms and null results and optimize the dictionary; this may require a dashboard for on-going optimization. Finally, for each site feature or functional group, you may want to document possible analyses and optimizations.
Playbooks are the next step up from a basic data dictionary. A playbook outlines how to do basic analysis for a specific site feature or functional area. It outlines the basic business questions and shows examples of how to answer those questions. A playbook pulls together all the details of how measurement is implemented and how a business user can use the data to take action and optimize the site.
You might create playbooks for certain functional groups, like merchants or off-site marketing—or you could aggregate examples for site features, like search or refinements. Documenting analytics strategies is helpful for increasing your organization’s analytics maturity. The documentation is also excellent training material for new team members and business partners.
A System to Log Requests
Managing a huge backlog of requests is a common problem for Web analysts. Coming up with a system to log and prioritize those requests is critical. You must provide backlog visibility to your managers and stakeholders to ensure that you handle the most impactful requests first.; this can help you justify additional analytics resources.
Your Outlook inbox is not the ideal system to log and manage critical requests. One solution is an online form, such as Adobe FormCentral, which is a great way to funnel requests into a spreadsheet that can be easily shared, sorted, and updated.
Roadmap for Improvements
What are the long-term goals for your analytics implementation and the entire company? A great analytics organization needs an analytics roadmap to identify and prioritize top optimization opportunities. Updates and improvements can be made to any implementation, but such changes must take into account feedback from key business users and analysts and management.
Adobe Consulting has an analytics maturity model that outlines the phases an analytics organization can enter and master from descriptive to diagnostic to prescriptive. It’s important to understand the next steps your organization must take to become a highly effective analytics team that drives measurable results.
These tips can help your analytics team become more effective at driving adoption and success throughout your organization. It’s important to build a strong foundation that will help your analytics team build trust with your business partners and help those partners become active participants in the analytics optimization process.