---
date: 2018-08-01
type: ship
title: Shop 99
slug: shop-99
project: Independent
kicker: Conversational commerce, before chat-as-interface was an idea.
excerpt: Designed an AI shopping assistant where chat replaced search, filters, and category browsing. Each exchange was structured to teach the system the user's intent.
cover: /assets/covers/hero-shop99.webp
palette:
  accent: "#7E8B4E"
  source:
    brand: "Caran d'Ache"
    name: "Olive Brown"
  role: Lead Product Designer
  pull: We replaced search with a conversation. The conversation taught the engine what each shopper actually wanted.
tags: [ai, commerce, conversation, 0-to-1]
---

Shop 99 explored what online shopping looks like when conversation replaces search. Users described what they wanted; the system learned style, price range, and intent from dialogue alone. No filters, no categories, no taxonomy.

## Context

Early ML experiment with almost no training data. The model needed to learn preference and context directly from user input. The design had to support that process: guiding people to phrase requests clearly without breaking the flow of a conversation.

![Shop 99 message list on iPhone, against a flat green backdrop. Header reads "Shopping Requests" with a back arrow and a "Search in messages" field. Three conversation rows stack below: a Shop 99 update reading "We have completed the process. Your jacket will be send shortly." timestamped Yesterday; a second Shop 99 thread reading "Understood. Package is sent to your home address. Your tracking number is…" timestamped Last week; and a "You" reply reading "Perfect. You guys are awesome <3" also Last week.](/assets/projects/shop-99/shop99-message-list.webp)

## What I designed

The conversational tone, the message structure, the fallback behaviours. We mapped common shopping dialogues and modelled them into chat patterns where each exchange doubled as training. Visual design stayed deliberately quiet so users would attend to language, not interface.

## The decision that shaped it

Treat chat as a full interface, not an add-on. Most "AI chat" features then (and many now) bolt a chatbot onto a traditional product surface: search bar above, chat window below. That tells users the chat is supplementary. We did the opposite: chat was the only surface, with no fallback to filters. It forced the conversational model to actually carry the workload, and forced the design to be specific about timing, memory, and trust rather than hiding behind familiar UI.

![Two laptops on green pedestals showing the Shop 99 back-office. The closer screen is the operations dashboard: "Shipments" tiles read New 50, Ready for shipment 98, Pending Approval 24, Cancelled 26; "Payments" tiles read Total Receivable $9,800, Total Paid $8,520, Total Due $1,200, Refunds $500; below sit "New Members" and "Latest Shipments" tables. The further screen is a customer detail view for "Nancy Banner", with profile photo, contact and personal information forms on the left, gift-card and activity panels on the right.](/assets/projects/shop-99/shop99-dashboard.webp)

## What it left behind

Launched as a live prototype. Users completed purchases faster and explored broader categories through dialogue alone. It became an internal reference for how conversational interfaces could carry e-commerce workflows without traditional UI scaffolding.

![A loose collage of Shop 99 mobile screens with soft drop shadows, suggesting the breadth of the product surface beyond chat: a side-drawer menu (My Card, Locations, Settings, Add Money, Shopping Request) over an Order list; a transactional success screen reading "Congratulations! Transaction is successfully completed." with a $50 voucher; a "My Gift Card" account showing Jessica's $50 balance and a "Transaction History" of small invoices; a "Settings" screen with profile photo, Display Name, Change Address, Security Settings, Terms of Conditions; and a "Locations" picker with "Pickup Locations" and "Voucher Points" tabs.](/assets/projects/shop-99/shop99-surfaces.webp)

## Role

Lead Product Designer. Defined product behaviour, conversational structure, and interaction principles end to end.