Process

Discovery & Research

Our journey through the woods begins with a Google survey. The survey was posted to multiple woodworking groups via Facebook, in addition to the link being shared by the client with fellow woodworking enthusiasts. This survey gave us significant raw data.

User Personas

For the MVP we chose to represent all 3 personas crafted from the user research and tell their unique stories. We wanted Timber to be as simple, yet as flexible, in use as possible.

Meet Aaron

Meet Harry

Demo

Age: 47      Male      
Banker      Avid Hobbyist

User Story

"As a husband, father, & bank manager I need to be present for my family and not spend money we don't have, so that I can model discipline for my family."

Pain Points

• Hates shopping when he could be woodworking
• Refuses to put his love of woodworking before his family's needs
• Hates waiting for a specific kind of wood to be in his price range

Goals

• Wants to be present for his family
• Wants to model financial discipline for his kids
• Wants to save time & money

Meet Joe

Demo

Age: 30             Male      
Contractor     Professional

User Story

"As a contractor who wants to work in a Circular Economy I need access to reclaimed wood so I can create unique pieces for my clients."

Pain Points

• Hates online mediums that don't give enough info
• Hate seeing healthy trees cut down
• Is tired of clone style furniture

Goals

• Wants to know the pros & cons of used materials
• Wants to promote "living green"
• Wants clients to have one-of-a-kind pieces

Information Architecture

User Flow

This basic User Flow demonstrates how Timber Users can select & save filters, then apply them to view matching posts.

Wireframes & Paper Prototypes

From simple wireframes on graph paper to arts & graphs style paper prototypes, this step of the process produced tangible deliverables that directly shaped the digital wireframes to come.

This is the 1st iteration based of the paper deliverables and informal discussions with members of woodworking & carpentry communities online.

This 2nd iteration comes with a sign in sequence, as well as a redevelopment of the "Filters" page as per feedback from the client.

This is the version that was offered online for unmonitored user testing via Maze.com, as well as the 1st round of live testing via Zoom.

User Testing

This was a very eye opening experience! Not only did we learn Waze provided the click test after specifying tasks, but the results came packaged with in-depth analytics such as time spent on page, heat maps, and success rate (both direct & indirect).

We were able to test on 3 tasks:

1.  "Pretend you are using a different phone. How would you tell others you have excess wood available?"

2.  "Pretend you are new to the app. Search for firewood within 15 miles of your zip code."

3.   "Pretend you have used this app before. Set yourself up for notifications."

In all honesty, the results were terrible:

• Across all 3 tasks, an average of 33% of testers failed to reach to target page.
• Across all 3 tasks, an average of 40% of testers had success with the tasks, but through indirect routes.

And that wasn't just due to it being unmonitored! The users who completed the testing via Zoom had many pain points also.
And they were quite vocal about them...

3rd Iteration- This is the iteration that came to life after the unmonitored testing & first round of live testing via Zoom. We focused on cleaning up unclear icons, as well as adding simple text to describe them. We nailed down exactly what should be included in both the "Post" & "Filters" criteria.

The biggest thing we accomplished was taking a step back and simply following the mental model established by many apps with a main map screen. By placing the "Filters" & "Location" icons directly on the map, we  saved users from hunting for them and prevented frustration.

The major differences were all validated by a second round of monitored tests results and user feedback via Zoom. The overall reception was as different as night & day compared to the first round!

Feeling confident, we presented this iteration, along with the findings that led to its creation, to our client. We also presented the client with a color palette easily customizable from 3 different variations.

Conclusion

And this is where we are with Timber. Our client is deciding on the best application of the color palette, as well as lining up potential advertisers. We educated the client about the need for alternative text via Aria. We also made clear that we'd need to develop a landscape mode for those needing increased viewability.


Overall, it was a robust sprint in order to get the final clickable prototype ready. The ability to distribute a link for the Google survey proved invaluable during the current pandemic. Being introduced to maze.com as a source of unmonitored testing was another boon, but we do need to remember that some testers are mavericks and will click all over an app when unmonitored. That was still no excuse for those dreadful numbers from the 1st round of testing.


It's amazing how close we thought we were before that round of testing. Being able to obtain those numbers, as well as the feedback from the live testing via Zoom, really gave us concrete pain points and directly led to changes made that ultimately yielded our MVP.


Oh, and how did the guys we met earlier feel about it? Take a look at their User Journeys and see for yourself.