Handwork

Problem: How can you take product learnings from a retail launch to earlier in the development process?

Role: Hardware Prototyper, Videographer, Design Researcher

Responsibilities: Spearheaded the development of a 'smart' retail display to gather quantitative data.

Skills used: Arduino development, videography


Rapid, real world product testing.

In a traditional product development cycle, an idea is prototyped, built, and years later, launched in store. In many cases, the learnings that could have helped a product succeed are gained after its launch.

The Micropilot is a methodology that takes these learnings and brings them to the front of the development process. It tests the most crucial part of the innovation in a real environment to gather insight on real customer behavior.

As an internal project, our team created a fictional line of unisex hand-care products positioned for makers. We hypothesized a maker positioning would appeal equally to both men and women. To test our hypothesis, we "launched" five hand-care products in three retail environments and developed a "smart" display to log real world engagement.

I lead the development of the quantitative “smart” display to measure product engagement. The display logged each pickup and place with a timestamp and product ID, enabling us to calculate length of customer engagement for each individual product.

With the quantitative data from the display and using traditional observational research methods, we learned 52% of the customers who engaged with the products were male and 48% were female. 3/4 of the interactions were with only two of the products, the lotion and the hand salve.

The display was built with an Arduino Uno, an Adafruit SD card logging shield and a combination of IR and Hall Effect sensors. Checkout the code on github

DSC_6795.jpg
The guts of the smart display.

The guts of the smart display.

Explaining the process of calibrating a weight based sensor setup.

Explaining the process of calibrating a weight based sensor setup.

Tackling the inconsistencies of weight based sensing.

Tackling the inconsistencies of weight based sensing.

Initial testing of the IR sensors. Alongside the data logging development, we prototyped different retail display designs.

Initial testing of the IR sensors. Alongside the data logging development, we prototyped different retail display designs.

Self logging to verify the effectiveness of the display.

Self logging to verify the effectiveness of the display.

One of our in context setups at the Stonestown Galleria. We also set up the display in Wink SF, a small boutique in Noe Valley, and Betabrand, a clothing apparel company in the Mission District.

One of our in context setups at the Stonestown Galleria. We also set up the display in Wink SF, a small boutique in Noe Valley, and Betabrand, a clothing apparel company in the Mission District.