Caroline Zenss's profile

UX Research - Build an IA for a Digital Workplace

2017

UX Research - Multiple Methods
Redefining a digital workspace's information architecture

Task & Context
Deliver UX Research to a customer in order to build a new information architecture for a digital workplace, as part of a migration exercise. 
Customer was the IT department of a large Canadian company operating in the financial services industry. Budget was very limited, and access to current users was restricted.

Approach
In order to come up with the right combination of methods, I considered the following:
- What is the current problem?
- What is the workplace used for? What should it be used for?
- How can I help my customer shine rapidly in front of their stakeholders?

What is the current problem?
The previous intranet was barely used by employees because of its lack of search capabilities and navigational elements, and overall lack of user-friendliness. 

What was the workplace used for? What should it be used for?
As a central repository for all corporate information, the intranet was perceived as outdated and irrelevant. Departments were used to providing information in person, over the phone and via email to employees on a daily basis, on top of making it available online.
Stakeholders were mostly unaware of the site flaws, traffic patterns and employee dissatisfaction towards the current experience. 

How can I help my customer shine?
Through my conversations with my client, I realized that they were facing a mentality of "build the new one, and they will come", and that changing their superiors' thinking would be challenging. My client worried a lot about history repeating itself with money spent on a new platform that would end up not being used at all.

UX Research Methods Diagram
With that context in mind, we ended up agreeing on using a combination of user research methods that would show value to the stakeholders, by presenting numbers and insights together. Being key players in the financial sector, we knew that numbers would speak to them, but that an iterative and creative process would also help create excitement along the way.
We agreed on 4 methods as per the diagram below.
1- Website analytics : helps identify current content categories, popular pages, unpopular ones, mobile traffic, bounce rates, traffic sources.
2 - Card sorting : engages representative end users to organize topics into categories that made sense to them. The first round was done in person with employees from various departments. The second round was done online through Optimal Workshop, with employees from across the country.
3 - Employee Surveys : collects feedback on current usage of the portal, including tasks that are performed most often, frequency of visits, current perception of its overall friendliness, and suggestions for improvements. Employees who participated to the card sorting exercise and the screenshot testing were also asked to participate to the surveys.
4 - Usability Testing (Screenshot testing) : quickly collects feedback from real users in order to make user-friendly decisions about visual elements, semantics and placement of functionality such as navigation and search.
I delivered a sitemap as part of each method. The first one to kickstart the in-person card sorting exercise was based out of Nielsen Norman Group's 2017 Intranet Design Annual

At the end of each exercise, I provided my client with a new iteration of the sitemap based on outcomes, along with key findings for the stakeholders to absorb and review. 

Key highlights
- When it comes to building a digital workplace, there was a significant difference in usage between internal departments (such as Marketing, IT, Communications) and external departments serving clients.
- Asking for feedback from employees across the country through the Optimal Workshop research tools was not only efficient and low-cost, but it also helped create excitement for the new portal, and put the old portal in the past.
- Testing the information architecture visually (through a representation of a menu rather than a sitemap) by asking users to perform key tasks such as filling out a timesheet and checking vacation balance helped fine-tune labels.

Screenshots
I cannot disclose the customer's name or final information architecture. Below are a few screenshots to show my work. All references that could help identify the customer have been removed.
1 - Website Analytics: segmenting data to identify differences in behaviour between two distinct segments.
2 - Closed card sorting results : expected (green) versus unexpected (red) sorts.
3 - Survey questions and question types, post-card sorting exercise.
4 - Online screenshot testing: task results and takeaways.
This project in a nutshell:
Combining hard numbers with user feedback helped shape the navigation in a unique way that will work for most employees in the long run. Looking at only one or the other would not have provided enough information to make the online screenshot testing of the new proposed navigation a success. This approach also helped my client build a strong case for every decision they made along the process, whenever they were requested to report back to their superiors.


To find out more about UX Research or UX design services, 
contact me at caro@carolinezenss.com 
UX Research - Build an IA for a Digital Workplace
Published:

UX Research - Build an IA for a Digital Workplace

UX Research Methods, Information Architecture

Published: