Fernando Catter
ShopSmartAutos

Independent concept / Mobile and web

ShopSmartAutos

From a description of everyday needs to a shortlist a shopper can inspect

I designed a mobile and web concept for someone who knows their budget, passenger needs, and preferences but has not chosen a make or model. The assistant turns that description into reviewable preferences, supports comparison, and helps prepare an inquiry to a dealership.

Role

Problem framing, interaction design, UI, and prototypes.

Team
Independent project.
Surfaces
Mobile app and responsive web.
Stage
Two clickable Figma prototypes using invented listings and scripted assistant replies. Inquiries are simulated.

Conditions

The listings are invented.

The makes and models are real. Every listing, price, mileage, and dealership on these screens was written for this work, and no listing service is connected.

The assistant is scripted.

The replies follow an authored path. Nothing here is a working AI integration, and no inquiry is sent to anyone.

The problem

Begin with the need, not the model

Where the shopper starts

The project explores a specific starting point: the shopper can describe how the car needs to fit their life, but has not translated that into inventory filters.

What the assistant is for

The assistant's job is to help with that translation, not to make the choice on the shopper's behalf.

Where this goes

Every screen that follows starts from that sentence and keeps it editable, so the shopper corrects the assistant rather than starting again.

What a shopper can say in plain words set against the fields a listing site asks for

The interface

Turn the sentence into something checkable

The sentence

The example starts with a used SUV under $25,000, five seats, and a preference for blind spot monitoring.

A plain sentence of needs turned into reviewable preferences

Weight and the budget question

Those details do not all carry the same weight. I separated requirements from preferences and made the budget interpretation visible, including the question of whether taxes and dealer fees are included. The preference review lets the shopper correct that interpretation before it changes the results.

Comparison

Put the explanation beside the comparison

The explanation stays beside the facts

The comparison panel keeps the assistant's explanation beside the vehicle facts behind it. A price or mileage difference can then be inspected rather than accepted as a recommendation without context.

A comparison with the assistant explanation beside the vehicle facts

A price is only useful next to the reason it’s that price.

What a difference does not prove

I avoided treating lower mileage as proof of better condition or an advertised price as the complete cost. The comparison should clarify a trade-off, not hide an assumption inside it.

The inquiry

Keep the inquiry reviewable

The draft

The inquiry panel carries the selected vehicle and the shopper's questions into a draft message. The shopper reviews it before it is sent.

The inquiry draft, carrying the vehicle and the questions, ready for review

A request is not a booking

A visit request remains a request until the dealer responds. That distinction keeps the interface from presenting an unconfirmed arrangement as a booking.

The key decision

Review the interpretation before showing the cars

The two options

I explored showing results immediately, then refining them, versus asking the shopper to review the preferences first.

Two options: results first and refine, or review the preferences first

Why the extra step

I chose the review step. It adds a tap even when the interpretation is correct, but makes it easier to catch a wrong constraint before suitable vehicles disappear from the shortlist.

Two surfaces

Give each screen size a useful role

One task at a time

The mobile prototype handles one task at a time: describe, refine, compare, and ask.

The mobile prototype, one task per screen

The request beside the results

The desktop prototype places the assistant beside the results, so the shopper can adjust the request while keeping the inventory in view. These are two presentations of the same task, not separate product stories.

The desktop prototype with the assistant beside the results

Uncertainty

Make uncertainty visible

Three different gaps

The prototype includes no matching vehicles, incomplete pricing, and missing equipment information. Each needs a different explanation: revise a need, confirm the total, or acknowledge that an equipment detail is not stated.

Three incomplete information states, each explained differently

A gap is not a no

Missing information is not the same as a negative answer. The interface should preserve that distinction while keeping a useful next step available.

Delivery & scope

Two prototypes, invented listings


The listings are invented, the assistant replies are scripted, and inquiries are simulated. The prototypes cover preference review, comparison, inquiry, and the main incomplete information states.

  • Two clickable Figma prototypes, mobile and responsive web
  • Preference review, comparison, inquiry, and the incomplete information states
  • The framing, the interaction design, the interface, and the prototypes are mine
  • Invented listings, so every number on screen had to be visibly labeled as a sample