gdgt / Engadget
Fifty routers on a shelf, one nervous shopper, and a brief with almost no information.
GDGT and Engadget asked for a shopping companion app. The brief was a paragraph long, and I couldn’t ask a follow-up question. I had to decide what the shopper needed, write down my assumptions, and show how the app would work.
What can you design from a one-paragraph brief?
The brief: a companion app with one purpose, help someone shopping for a consumer-electronics product make an educated buying decision, using the rich product data from Engadget and gdgt. The scenario that anchored it: a smartphone-owning parent walks into a big-box store with no idea which router to buy, and there are fifty to choose from.
Define, ask, synthesize.
Rather than jump to screens, I organized the whole response around an explicit three-step process and presented it that way, because with a vague brief, the reasoning path is the deliverable.
Define the problem
What don’t we know that we need to know to move forward?
Ask the questions
What can I ask to get the answers I need, and when no one can answer, what do I assume, out loud?
Synthesize
What do I do with what I’ve learned? Turn a point of view into screens someone can challenge.
What does a nervous shopper actually reach for?
I couldn’t run a study, so I answered from what I already had, a working mental model of non-technical shoppers from an earlier e-commerce launch. Three things the app had to make effortless became my guiding principles, and every screen had to earn its place against them:
Three guided entry points instead of a search box.
My first instinct, search and click through products one at a time, I rejected fast: for someone staring at fifty routers, that’s exactly the intimidating experience the brief was trying to fix. So I pivoted to guided entry points that meet the shopper where they physically stand: a curated Product Guide, a Scan UPC that points the camera at the box in your hand, and Search by Storeso you only see what’s carriable where you are.
Scan UPC collapses the gap between the physical shelf and the digital data, the real friction in the scenario.
Serve the primary user without amputating the power user.
The product detail page was built around the thing gdgt and Engadget do best, reviews and specs. I split it into critic reviews (an overall score, per-category breakdowns) and user reviews (a separate overall, plus readable text), with a full-specs tab underneath. The non-techie shopper gets a trustworthy score at a glance. The spec-hungry shopper can still dig.
The same layered detail page in context, category grid in, product page, and the full-specs tab that lets the power user dig past the review scores above.
Structure before styling.
The sketches argued for a hierarchy. The next step was to prove it held up as real structure. I translated the key screens into grayscale wireframes (the guided home, the refine-able results, and the layered detail page) so the layout could be pressure-tested before a single pixel was styled.
Where the reasoning landed, the hi-fi flow.
The wireframes resolved into a styled flow: a scannable Product Guide home, the Scan-UPC camera that pulls the exact box in your hand, Search by Store filtered to what’s actually on the shelf, refine-able results that lead with price, and a detail page that carries the whole decision. The hierarchy the sketches argued for, made real.
Five coded hi-fi screens, the same components carried into the deck. Swipe through the flow: home → scan → store → results → detail.
I wrote down my assumptions so the team could challenge the decisions.
I handed over sketches, wireframes, coded screens, and the assumptions behind each choice. I couldn’t ask a follow-up question, so I used what I knew about shoppers and made the reasoning visible. This was a design exploration, and it did not ship.