In Analytics

Conventionally, web analytics data is used to track marketing campaigns and study return on investment (ROI). But, with the advancement and sophistication of user data tracking, it can also be one of the most powerful beacons that information architects, UX designers, and user interface developers use to find their way through the midst of research data and design requirements.

Web Analytics Can Guide Your Ship Home

In order to use web analytics as a lighthouse to guide our design ships to the harbor, it’s essential that it functions flawlessly for existential reasons like marking dangerous coastlines, finding hazardous shoals, and sometimes assisting in aerial navigation.

Now, it’s a no-brainer that unless data recorded by web analytics is accurate, all the user behaviors and characteristics that we derive from it would be susceptible to misrepresentation. Why should we care about existential reasons?

To explain, I’ll take an example from built world. Both bookstores and grocery stores are meant to sell things. So, one might argue that they should be architected and designed to be similar. Why aren’t they? Because, existential reasons define the goals and objectives for each of them. And eventually, the architecture and design is derived from these goals and objectives.

People like to read reviews, and sometimes the first chapter, to assess their buying need. So, in a bookstore often times there are seating areas, nice carpeted floors and an overall ambient environment for reading.

In a grocery store, people want to quickly find their groceries, do on-the-fly price and brand comparisons, and then checkout. So, the shelves are stocked with related products from all brands, floors are tiled to facilitate the dragging of shopping carts, and abundant fliers point to the lowest available price.

Bookstore

Source: http://retaildesignblog.net/2013/01/21/bookstores-paagman-bookstore-by-cube-architects-the-hague-netherlands/

Grocery Store

Source: http://caseyxrobertson.deviantart.com/art/Grocery-Store-2-191196289

Using Web Analytics to Architect Desired Experiences

When we are clear about our existential reasons, goals, and objectives, stakeholders can use them to come up with the metrics for measuring success.

In the digital world, the ability to measure these metrics is surprisingly easy. Moreover, the friction of making changes in the architecture and design of the website is a lot less. But as it’s often said, “With great power comes great responsibility,” so we cannot just go around making changes. We’ve got to base changes on firmer reasoning. Therefore, if we have functioning web analytics in place that track our visitors and measure the metrics that assess the success or failure of our existential mission, goals, and objectives, then we are in a good spot to determine if change is required in the current structure or strategy. If yes, then architects and designers of these digital places can work in coordination with the stakeholders and make those necessary changes.

Here’s an example of how this might look:

  1. A content publishing website wants to engage users with its social feeds from Facebook.
  2. An architect arranges the parts on the page to make sure that the social feed is placed at a meaningful location.
  3. The designer and developers then bring this feed-widget to life.
  4. Eventually, web analytics points out that the interaction of users with this widget is lower than expected.
  5. This useful insight stirs the water at the right moment before all the valuable nutrients have settled down.
  6. Finally, the dance between architects and designers is rekindled to make the necessary changes.

TUG‘s Process for Supercharging Web Analytics

The goal of our process is to create analytics data that is consumable both by IAs and designers for making design decisions, and also by stakeholders for making business decisions.

This process is executed in three steps:

  1. Analytics Audit
  2. Interviewing Stakeholders and Users
  3. Alignment on Findings and Implementing Code

Analytics Audit:

We look at the existing implementation and assess its correctness. At this point in the process, we’re focused on making sense of the existing implementation, comparing it with the set of standards, and articulating some of the “whys” and “whats”.

An example to show how this might look:

excel spreadsheet listing web analytics requirements

Interviewing Stakeholders and Users:

Before conducting interviews, we ask our interviewees to fill out a survey. This is a priming activity that gets them thinking in a direction that produces richer responses during the interviews.

The interviews are focused on getting answers that could help us triangulate on those fires whose smoke we saw during the previous step. Another important objective of these interviews is to find out:

1) The goals and objectives (in start terms, the existential mission)
2) The metrics used to measure those objectives
3) The web-based behavior that reflects those objectives
4) How much time is spent looking at or using analytics data to measure those objectives
5) The kinds of things that the organization would LIKE to know but don’t know

excel spreadsheet listing web analytics requirements, how it's implemented, the reality and concerns

Alignment on Findings and Implementing Code:

We synthesize the findings both from the audit and interviews and try to present them to the clients in more consumable ways, often using dumb models.

model using circles embedded in larger circles that visualize the relationships among sections of the site

These models are accompanied by a strategy brief that outlines:

  • the existential mission of the organization its goals and objectives,
  • the key metrics and tactics that should be used to measure these goals and objectives,
  • and the organization’s current maturity in terms of web analytics and recommendations for further growth and improvement.

The implementation guide includes the findings from the technical audit, the necessary code to make Google Analytics function accurately, the new web analytics structure detailing the properties, views, advanced segments, and reports. If necessary, we work with the IT development team to bring these changes to life.

We’ve executed this process in four different organizations and have seen similar results: more coherent reporting, better tracking of goals, nimbleness in making necessary strategy and design changes, and by far greater awareness of the data and a better understanding of it.

Recent Posts