Identify an opportunity to improve the effectiveness of Home Depot sales associates in terms of their ability to accomplish essential tasks without compromising customer experience.
Home Depot associates of varying levels of expertise who juggle customer care and bay/inventory upkeep on a daily basis.
UX Designer: prototyped experiences at low and high fidelity through concept sketches, wireframes, and high fidelity interactive prototypes. Coordinated and coached team’s efforts on brainstorming and ideation exercises.
UX Researcher: conducted field observations, contextual inquires, interviews, and usability tests with target users and industry partners in a systematic fashion.
Team Presenter: represented the team’s research findings and designs through considerate slide deck design and oral presentations.
Client Relations Manager: coordinated with industry partners to define success parameters, schedule weekly meetings, and share resources and represent the team’s interests.
User Research Methods
In order to identify painpoints within the current system, the team organized a series of three qualitative research efforts:
Field Observations: visiting local Home Depots for the purpose of orienting the researchers to the real-world context that Sales Associates at Home Depot operate within, and use gathered insights to formulate meaningful and directed questions.
Semi-structured Interviews: engage stakeholders at the associate, management, and corporate levels with open-ended questions built upon insights from previous observations to gain a greater understanding of how tasks are managed, tracked, and selected.
Contextual Inquires: shadowing associates while they complete tasks within an everyday context to better understand behavior surrounding tasks. Observing associate interactions and potential distractions or failure points.
For each research effort, the team formed complimentary roles such as Interviewer, Note Taker, Auxiliary Note Taker, and Photographer. Notes, photographs, audio recordings, and any other artifacts were captured systematically for easy documentation (i.e. naming conventions).
After each user research session , the team would gather for debriefing sessions were insights were discussed and compared from two or more person’s perspectives to ensure accuracy. Left: a sampling of photographs that later became meaningful to the project:
Analysis & Painpoint Synthesis
Processing large amounts of diverse qualitative data required a strategic approach to systematically recording, tagging, grouping, and presenting findings to clients. Notes and audio recordings gathered during user research were transcribed systematically within an affinity map for the purpose of finding painpoints related to tasking. Over 450 individual insights (yellow) recorded and synthesized into 19 unique painpoints (red). This large document was then formatted into an easily browsable PDF for our clients (bottom left, right)
“ Processing large amounts of diverse qualitative data required a strategic approach to systematically recording, tagging, grouping, and presenting findings to clients “
The team discussed the potential value of each pain-point in terms of quantity of supporting research, relevance to client needs, and respect for the ecosystem of Home Depots products. Below: five painpoints with the most evidence and potential client impact isolated for the next phase of design.
Storyboarding & Client Feedback
The team ideated through rounds of sketching and discussing concepts in the double-diamond fashion of switching between divergent and convergent approaches. Over 50 unique ideas were created, which were then distilled into three approaches to design:
Task Capture: platform for capturing, describing, and delegating tasks in an efficient and systematic fashion
Expert Directory: connecting the right associates at the right associates at the right time for rapid assistance.
Departmental Waypoint: Information kiosks for customers to signal a need and for helpers to meet then at a designated point in the store.
To illustrate the context surrounding these designs to the client, I rapidly generated storyboards to represent the team’s vision. Presenting qualitative research findings with complimentary storyboards proved an effective means of eliciting insightful discussions with the client about possible design directions
“ Presenting qualitative research findings with complimentary storyboards proved an effective means of eliciting insightful discussions with the client about possible design directions “
Storyboards were presented to seven corporate stakeholders at Home Depot’s headquarters, including UX Designers, Product Managers, and Store Operations Managers. A form with Likert scale and open-ended questions was given to each participant, and collected after the presentation. The form allowed the team to gather qualitative and quantitative data to rank-order each of the three storyboards, and isolate one for further development.
Of the three concepts, Expert Directory was chosen for further development by its overall better qualitative performance and its favorability among the corporate stakeholders. Here are some of the more impactful positive comments:
“ …Saves unnecessary waste of time of associates having to search for other associates “
“….Amazing, this [ expert directory ] could really solve major problems and avoid loss of sales down the line “
“ This is the most practical idea to implement in our current system “
System Flow & USe Scenarios
The team started off the design phase by creating a system flow. With limited time for prototyping, the team developed two personas based on the strongest use cases available from previous research. App screens and features were driven by the contexts of these two cases.
“ With limited time for prototyping, the team developed two personas based on the strongest use cases available from previous research. App screens and features were driven by the contexts of these two cases. ”
CASE #1 | Mike
Mike is a sales associate at the Home Depot working in appliances. One day while walking past the Lumber department Mike is approached by a customer who needs help picking plywood for their kitchen cabinets. Mike knows other associates in the store might know, so he pulls out his store-issued phone and opens the “Ask Homer” app to quickly get in touch.
Case #2 | Jane
Jane is a junior sales associate at the Home Depot working in the Gardening department. While walking the fertilizer aisle, she notices a spilled chemical on the floor that looks dangerous. She doesn’t recognize the chemical, so she pulls out her store-issued phone and opens the “Ask Homer” app to find a coworker who can help.
User Experience Flows
With evidence-based user scenarios and a system flow in hand, the team began sketching wireframes and app features ( below ) . I led this phase of sketching, directing the team’s efforts towards an appropriate level of detail and consideration of composition. Since Home Depot associates use smart-phones issued by the store that run on android, interface development was guided towards Google’s Material Design for its intuitive mental models and well-built visual design systems.
Iterating on the previous wireframe sketches, initial user experience flows were created using Adobe XD for its rapid mockup and interaction capabilities. Effort was focused on “mini-flows”, or quick tasks within the grander scheme of accomplishing a goal. The team met to discuss these small batches of UX on a daily basis, which led to valuable comments on specific interface improvements. Below, a “mini-flow” of Mike’s story where he initializes an inquiry through a voice interaction.
Once UX flows for both use cases were created, the team conducted a focused critique on any obvious or blatant interface missteps in terms of usability, language, and conformity to current mental models of messaging and voice assisted interfaces. The entire UX flow was overhauled according to the feedback, and designed with XD with functioning simulations of voice interaction. Below: final UX flow, system features, and a video demoing the voice interaction.
“ Adobe XD’s newest voice interaction functionality allowed us to simulate the prototypes intended use with a higher degree of accuracy than other tools “
The team carried out two tests of the final interactive prototype at the Home Depot headquarters in Atlanta:
Heuristic Evaluation with Home Depot UX Designers to examine the UI and identify pain points, strengths, and weaknesses of the system with Nielsen's Ten Usability Heuristic to evaluate the system ( image below )
Usability Evaluation with Home Depot Associates to understand issues and problems that users have with the system using moderated think alouds to identify components that are unclear and confusing (AKA) feature painpoints)
During both testing sessions, the team split into complementary roles that included a Test Facilitator #1 (me), Test Facilitator #2, and Photographer/Note Taker. Four Home Depot associates and three UX experts participated in the test. To see an example of recorded heuristic feedback, click here. For a detailed report on the testing procedures, visit Alexandra’s portfolio, who designed our testing procedures.
Early analysis of the SUS Usability Evaluation ( below ) revealed an overall average of 84.6. Based on SUS standards, a score above 68 is considered above average in terms of usability factors. While these results are optimistic, another test of the same nature is needed with more participants. During the testing, a none-negligible amount “nudging” came from myself and the other facilitator, which may have biased the results.
Insights from the Heuristic Evaluations from UX designers brought to attention confusing and non-standard visual design language within the interfaces. The insights focused on feature details, and not the features themselves which were widely approved. Another iteration that takes these critiques into account may produce a significantly better, cleaner, more friendly prototype.
Thanks for reading!