Wikileaf

Usability Evaluation

In 2014, Wikileaf was founded to empower cannabis users. Their website provides detailed information on cannabis, including use information, price points, and related news. To assess usability, our methods included a heuristic evaluation, cognitive walkthroughs, and usability testing. We found fourteen usability issues, including a lack of accessible terminology for novice users, unclear visual cues for clickable icons, and the lack of filter/sort options for prices.

Overview

Role

UX Researcher & Team Lead

Duration

9 Weeks

Zoom, Otter.ai, & Figma

Tools

Deliverables

Report

Team Members

Me, Asato Kobayashi & Alison Hsu

My role: As the team lead, I oversaw the project and was involved in all the phases, particularly with the study design, data collection, and analysis.

Process

Discover

Background research
Target user identification

Explore

Heuristic evaluation
Cognitive walkthrough
Pilot tests

Test

Remote usability tests

Analyze

Completion rate
Ease of use
Likelihood of use Severity ratings
Feedback

Deliver

Report with recommendations & mockups

Due to legalization movements and the growing use of cannabis to treat health conditions, Wikileaf must be useable for new users.

Problem

Assess the current state of Wikileaf and create actionable design recommendations.

Goal

Initially, the project began with understanding Wikileaf's goals, services, and uses. We focused our target user group on wellness cannabis users with high information needs.

Target User Profile

Wellness Cannabis Users

Wellness cannabis users are interested in learning about cannabis to improve their quality of life and address their health concerns. Additionally, they are eager to access information that can help them make informed choices. Their information needs can include general guidance about cannabis, such as what types of cannabis are available, its uses, how to use it, and where to access products.

The general characteristics of wellness cannabis users are:

  • Age: 21 years of age or older

  • Location: anywhere in the United States

  • Experience with cannabis: varying (none-expert)

  • Purpose of cannabis use: chronic pain relief

We completed a heuristic evaluation and three cognitive walkthroughs to identify usability issues.

  • Heuristics insights: there are inconsistencies in styling, feedback, and labeling.

  • Cognitive walkthrough insights: cannabis jargon may confuse newer users & clustered areas may prevent users from finding what they need to complete a task.

We used the most severe findings (ratings of 3 or 4) to narrow our scope to the Strains and Dispensaries page for usability testing and to create tasks.

Heuristic Evaluation & Cognitive Walkthroughs

We conducted five moderated tests with five tasks remotely. Tasks focused on the core features of Wikileaf that could represent a flow of finding cannabis information and then preparing to purchase a product. Our recruitment strategy focused on people without previous Wikileaf experience but with varying levels of cannabis experience.

Testing

  1. Will cannabis users be able to effectively search through the strains page?

  2. Will users be able to find information about how cannabis helps with specific symptoms?

  3. Can users find a dispensary near them easily?

  4. Can users compare prices at different dispensaries?

  5. Can users find products based on price at a dispensary?

Objectives

  1. Find a type of cannabis that addresses your arthritis.

  2. Find information about the strain "Wedding Cake" & how it can help with your arthritis. 

  3. Find dispensaries within 10 miles of Seattle.

  4. Select the dispensary with the lowest price.

  5. At Grass & Glass, find the lowest price for 1 gram of Sativa flowers.

Tasks

Task Completion & Ease of Use

Results

There were mixed results. 2 out of 5 tasks were highly successful. These tasks focused on location information and in-store dispensary pricing.

3 out of 5 tasks had low success rates. These tasks involved price comparison across dispensaries and finding health and strain information that would be vital to our target user group.

Despite the challenges and perceived difficulty, all users indicated they would likely use Wikileaf.

Likelihood of Future Use

Key Findings & Recommendations

Users' overall impression of Wikileaf was positive. Novice users appreciated the wealth of information available on Wikileaf. Additional positive comments included satisfaction with the medical focus of the information provided and within the reviews.

Positives

Areas of Improvement

We found 14 usability issues and rated them using the Nielsen severity rating scale. Common issues included complex language, a mismatch between design & user expectations, ambiguous labels, and the lack of filter/sort options for prices. We made actionable recommendations with feature mock-ups whenever possible.

The user’s current location is missing from the map

  • Severity rating: 3

  • Recommendation: Consider adding a pin icon that shows where a user is located. The pin icon should be distinct and vary in color and shape from the blue pricing labels.

Examples of errors & recommendations:

Recommended design change: adding a user’s current location so they can see where they are relative to dispensaries

Clickable icons are unclear

  • Severity rating: 4

  • Recommendation: We recommend including consistent feedback for clickable icons by adding a dimming effect on the mouse hover to match the icon's interactions on the buttons and containers on the homepage.

Users thought clickable icons were illustrations

No filter or sort by price option for products

  • Severity rating: 4

  • Recommendation: We strongly suggest adding options to sort and filter by price.

Users struggled to find a cheap product in Task 5 without a filter

What worked: Collaboration was critical to this project. This is project is where I learned to match a recommendation to every finding and having a team made the process more thorough.

Reflection

What didn’t: We mistakenly assumed the users' technical skills for computer usage. A few tests had delays during setup. Testing was occasionally delayed when users struggled with browser specifications and sharing their screens on Zoom. We quickly changed our testing setup to run checks before each session and provided detailed instructions in advance.