Driving Conversion with Concept Testing

Company: Questrade
Role: UX Researcher
Tools: UserTesting.com, Miro
Duration: 3 weeks
Team: Sr. UX Researcher
Stakeholders: Product Manager, UX Designer

Problem Statement & Goal

Questrade aimed to improve conversion rates in their point-of-sale (POS) flow, but there was uncertainty around which features would increase user trust and encourage decision-making confidence. The product team had hypotheses but needed insights on how users would react to these proposed features.

Key Research Questions:

  • Which POS features build trust and improve conversion rates?

  • Do transparency elements impact purchase confidence?

  • How do uncertainty and missing information affect decision-making?

Research Questions

  • Which POS features build trust and improve conversion rates?

  • Do transparency elements impact purchase confidence?

  • How do uncertainty and missing information affect decision-making?

Timeline of research study from initial screener write-up to stakeholder share-out.

My Role

As an assisting UX Researcher, my focus was on conducting research in close collaboration with the Senior UX Researcher and Product Manager. I led usability sessions, gathered insights, and played a key role in synthesizing and interpreting data.

Research Approach

To evaluate how users interact with the POS flow, I conducted:

  • Stakeholder Alignment

    Defined five key hypotheses based on prior user behavior and business goals, focusing on trust-building elements.

  • Participant Recruitment

    Selected 8 participants (ages 25-55) who were actively shopping for similar financial products online.

  • Moderated Usability Testing

    Led sessions to observe decision-making behaviors and validate or challenge our initial hypotheses.

Affinity mapping organizing user feedback for different areas of the POS flow tested.

Analysis & Insights

Finding #1: "I Don't Know" Checkbox Created False Data

  • Before: The checkbox was designed to help users proceed without making uninformed choices.

  • After Testing: Users preferred to input their best estimate instead of selecting "I don't know," leading to inaccurate survey data.


Decision & Impact:

  • Deprioritized this feature, recommended a hyperlink instead explaining where users could find the information needed.

Finding #2: Transparency in Coverage Increased Trust

  • Before: The “What’s Not Included in Coverage” section was an optional, collapsible detail.

  • After Testing: Users strongly favored transparency, with most saying it increased their trust in the product.


Decision & Impact:

  • Emphasized this section in the final design, making it always visible instead of hidden.

  • Increased trust scores by 29%, leading to a higher likelihood of users completing the purchase flow.

Slide presented to stakeholders summarizing user testing results on "I Don't Know" feature.

Slide recommending addition of informational hyperlink to form.

Restructuring Navigation to Improve UX

UX Research

Read More

Designing Secure & Flexible Admin Tools

UX Design

Read More

From Cart to Kitchen: Reducing Food Waste

UI/UX Design

Read More