fobbr
"I don't carry my elevator keys anymore and wear my Apple Watch every day." — fobbr beta tester
46.2
NPS (beta study)
40
Hand-made prototypes
Create-X
Incubator
2 years
Duration
“Your wrist already tracks your heart rate, unlocks your car, and sends messages — but to get into your apartment building or office elevator, you're still fumbling for a plastic fob. 36% of the commercial RFID market runs on 1990s technology. The gap between what wearables can do and what access control has evolved to is a market failure — and a category-creation problem: no reference price, no existing behavior to research.”
My role
Co-founder and sole designer-researcher. Product vision, industrial design, user research methodology, pricing strategy, brand identity. Co-founded with Carl Demolder (ME, Georgia Tech).
Where judgment was required
The moments that shaped the product.
A photo grid. Images include: the nylon band material before sewing; a partially-sewn band with the RFID chip pouch visible; the finished beta band from above; a detail of the sewn closure. The visual should communicate craft and commitment, not polish.
40 hand-made beta bands — research commitment over research convenience
The original plan was injection-molded silicone bands. When it became clear that path wasn't available on our timeline, I pivoted to premade nylon bands and hand-sewed RFID chip pouches into each one — 40 units, cut by hand, photographed for documentation. The study ran on schedule. This is not a story about resourcefulness as a personality trait. It is evidence of a specific research commitment: the study design matters more than the materials. The question being asked is more important than the prototype being used to ask it.
A timeline visualization. Day 1: setup + Calendly invite sent. Day 2-3: first journaling prompts. Day 5: mid-study Qualtrics survey. Day 7: final USE survey + Van Westendorp pricing study. Annotation: 'Progressive disclosure: context builds before the core questions arrive.'
Progressive disclosure study over a week-long journal
Rather than front-loading all questions in a post-use survey, the study revealed prompts gradually over seven days. Participants journaled about where they used the band, what context changed, what behaviors shifted, and how it felt across the week. This produced richer qualitative data than any single-session survey could — because behavior change requires time to observe. A week-long study is harder to run than a 10-minute survey. It is also the correct instrument when the thing you are studying is whether behavior changes.
The four-curve Van Westendorp chart: 'Too cheap,' 'Cheap,' 'Expensive,' 'Too expensive' intersecting. The acceptable price range shaded between the intersection points. Key annotation: 'Acceptable range: $11-$54 (expected: $50+).' Endowment effect note visible.
Conjoint analysis assumes known alternatives. This product had none. Here's what I used instead.
Conjoint analysis assumes known alternatives. fobbr had none — it was pricing a category that didn't exist. Van Westendorp finds the acceptable price range from scratch by asking four questions about price perception rather than comparing options. The results: acceptable range for fluoroelastomer bands was $11-$54, lower than our $50 assumption, likely due to an endowment effect from the free beta band. Knowing which research instrument to pick for a novel problem is a design skill. Applying the default method would have produced misleading data.
Three equal-width panels (triptych). Left: microscopy image of thorny devil lizard skin at 200-500um scale, hexagonal cell structure visible. Center: the abstracted hexagonal pattern derived from the microscopy image. Right: brand identity application — the fobbr logo or pattern applied to a surface. Reading order: natural observation -> design abstraction -> brand application.
The brand pattern came from a lizard's skin at 200 microns. That's not a metaphor.
The fobbr brand identity includes a hexagonal pattern derived from studying thorny devil lizard skin at 200um microscopy scale. The material research that produced the product concept also produced the visual identity. This is not metaphor — it is design grounded in the physical world the product inhabits. When visual identity has structural reasons, every downstream decision is easier and more defensible. Decoration is the thing you do when you don't have a reason.
Process
Proxmark 3 RFID analysis. Mapped fragmented tag formats.
Hand-sewn nylon bands with RFID chip pouches. 40 beta units.
Week-long progressive disclosure study. USE + Van Westendorp + journaling.
Won Convergence Innovation Competition. Graduated Create-X.
What Shipped
87.3%
Usefulness (USE)
96.5%
Ease of Learning
46.2
NPS
40
Hand-made bands
USE Survey scores all A-grade on the SUS benchmark. NPS of 46.2. Won the Convergence Innovation Competition — the golden ticket into Georgia Tech's Create-X startup incubator. Graduated Create-X. Award: https://research.gatech.edu/ipat/cic/spring-2021/fobbr
- USE Survey: Usefulness 87.3%, Satisfaction 85.8%, Ease of Learning 96.5% — all A-grade SUS
- NPS: 46.2 (53.8% promoters, 7.7% detractors)
- Won the Convergence Innovation Competition — entry into Create-X
- Graduated from Georgia Tech's Create-X startup incubator
What I Learned
The hardest design problems in fobbr were not visual. They were structural. How do you price something nobody has seen? How do you study a behavior that doesn't exist yet? Those questions don't have answers in most design curricula. Van Westendorp answered the pricing question. A week-long progressive disclosure journal study — not a post-use survey, but a series of prompted reflections revealed over seven days — answered the behavior question. Knowing which research instrument fits which problem is a design skill. It only develops by doing research on problems the standard toolkit wasn't designed for. The brand identity derived from thorny devil lizard skin at 200um microscopy scale is worth naming as a principle too: when visual identity is rooted in the material world the product inhabits, it has reasons. Those reasons make every downstream visual decision easier and more defensible.
What this demonstrates