Brainstorm Design Ideas

To figure out why users drop off at each step, I made some assumptions for potential reasons based on previous research. In order to mitigate the confusion and frustration in the flow, I brainstormed multiple design variants for each step, which can potentially improve the experience. 

Why do users drop off at “Trade Selection“ step?

Potential Reasons:

• Don’t know which trade to select.

• Don’t consider their issues belong to any provided trades.

• They are not committed users and this step is different from what they expected.

How might we help users complete trade selection easily?

Why do users drop off at “User Information Collection“ step?

Potential Reasons:

• Prefer to continue as a guest.

• Not the right time to ask users to sign up before their issues are solved.

How might we integrate the “User Information Collection” step frictionlessly?

Converge

It’s easy to become over-confident in our own design ideas. Rigorous testing and iteration can help challenge our assumptions and bring clarity.

User Testing on Ideas

How do we know which variant will work and actually decrease the drop off rate? 

One way to do it is to update the features with the ones that we think would work best, track the data and compare with the current flow. However it would take a lot of time and resources for development and data analysis. Therefore, I went with a more time efficient method, Preference Testing.

Testing Methods

Besides the preference testing, I also asked the participants to try out the current RA flow on the website and give feedback as they complete each step. 

Participants

Participant Requirements: 

• People who have not used HomeX Direct Remote Assist before

• Internal employees who do not work on HomeX consumer website

Recruiting Channels:

• Company Slack channels

• Social media platform: Nextdoor, Facebook, Reddit

Number of Participants:

As found in a study, 5 participants could find 85% of the usability problems. Eight participants should be a good starting point to test the design ideas.

Affinity Mapping

I organized the qualitative data from the user testing into an affinity map to categorize user’s feedback.

Rating Results for Design Ideas

Testing Results for “Trade Selection”

Testing Results for “User Information Collection”

Insights & Solutions

Insights:

• Unexpected next steps and lack of context: During the test, I always asked participants what are their expectations for the next step, and they never got it right.

• Need reasons behind each steps: Users expect reasons for why they have to complete certain tasks, for example, why need to select the trade, why they need to provide their information, etc.

• Varied knowledge & confidence level: the amount of knowledge people have about home issues varies greatly.

Solutions:

• An overview and progress indicators for each step: It would be helpful for them to have a sense of context by providing an overview and progress indicators at each step. 

• Give clear and strong reasons for why each step is necessary: It might be helpful to give them clear and strong reasons for each step. This need to be presented in a clear and concise manner as I also found that they don’t read the secondary small texts on the screens.

• Further research on target users: We really need to do more research about our target audience for this feature. 

Iteration

Design Iteration for “Trade Selection“

Design Iteration for “User Information Collection“

Takeaways

• Start planning early: Start user testing planning and participant recruitment as soon as possible

• Importance of verifying assumptions: Some design decisions are based on our assumptions for users with the best judgement of the information we had at that time. It’s important to verify the assumptions.

• Value of user testing: 

  • Get design ideas tested to narrow down options before spending too much time on improving design details

  • No extra dev effort needed