Website Global Navigation Research

Financial Services Website Global Navigation Research

2023

Customer: NerdWallet

My role: Sole researcher

Methods used: Open card sort and tree test

Tools used: UserTesting, Excel, Google Suite, Slack

Stakeholders: NerdWallet Content Strategist and Product Design Manager

Timeframe: 10 weeks

Challenge

NerdWallet, a leading personal financial company, wanted to take a fresh look at how their users contextualized, grouped, and labeled financial topics.

Early information architecture strategies were driven by employee assumptions and product revenue goals rather than user needs.

This research was part of an initiative to revamp the global navigation taxonomy to align with user expectations while also increasing traffic to revenue-driving landing pages. Primary focus was on the taxonomy level 1 groupings.

Solution

First, I conducted an open card sort. This generative approach helped us first understand how users label and group financial information without predefined categories. It also provided an opportunity to explore if users preferred primary content grouping by topic (e.g., credit cards) or type (e.g., calculators).

Utilizing the card sort insights, I partnered with a NerdWallet Content Strategist to create a high-level navigation structure and tree test tasks. We selected tasks that would help us understand user behavior for scenarios where the path inconspicuous or multiple paths were likely.

Influence

Participant groupings and nomenclature varied depending on the depth of their financial knowledge. However, the findings clearly indicated a preference for content grouping by topic rather than type (the existing navigation taxonomy).

NerdWallet has since updated their global navigation with a more user-centric approach.

Study Overview

Timeline

Research Questions

How do users contextualize the information provided to them?
How do users group and label topics?
How do user groupings align with, complement, or diverge from the defined taxonomy?

Approach

Below is a high-level overview of considerations I made as part of my research strategy:

I reviewed prior research to gain an understanding of the website’s global navigation evolution and prior research findings and recommendations.
Considering the team’s desire for new IA constructed from a user-centric perspective, I recommended starting with an open card sort. This generative approach helped us first understand how users label and group financial information without predefined categories.
Based on the card sort learnings, I partnered with the company’s Content Strategist to create a high-level navigation structure and 11 tasks for the tree test. We selected tasks that would help understand user behavior for scenarios when multiple paths were likely.
We recruited for three primary personas and analyzed data for the entire sample size as well as the individual persona types to help identify any similarities or differences amongst users.
Through the 10-week timeline, I kept the stakeholders informed with regular communications via, Zoom meetings, email and Slack. In addition, I provided a preliminary card sort high-level findings report to help aid collaboration as we began the tree test planning.
In the final report, I included a combination of tables and charts to showcase the quantitative data as well as clips and quotes to highlight the quantitive findings and build empathy amongst stakeholders. I also provided a list of actionable recommendations with priority rankings.

Methodology

Unmoderated, open card sort
- 90 participants (30/persona)
- Completed on desktop devices
- Participants were asked to group 43 randomized cards into categories they created themselves. The category names were then standardized across participants for reporting purposes.
- Cards included a sampling of topics, products, and facets (calculators, comparison tools)
Unmoderated, tree test
- 90 participants (30/persona)
- Completed on desktop devices
- Participants were asked to locate 11 randomized items
- Tasks were designed to understand user behavior when searching for complex or ambiguous content

Result

Original Global Navigation

Leads with business content types

New Global Navigation

Leads with user-centric topics

Reflection

What went well

Open communication between NerdWallet and myself through weekly meetings, Slack, and email
Collaboration with the Content Strategist to identify cards and tree test tasks that would be most insightful
Despite a tight timeline, I was able to deliver the findings and actionable insights by the deadline
I strengthened my IA method skills which helped improve efficiency when conducting future IA studies

What I’d change

Dedicate more time to discussing pros/cons of different card sort approaches with non-researchers to ensure the output will meet their expectations
Be more frank about the impact of segmenting analysis by persona (included despite being out of scope)
Factor in additional open card sort analysis time needed to standardize categories created
Prepare ahead for limitations with the UserTesting IA tool which resulted manual calculations (this tool was required as part of their Professional Services contract)