Back to database

Vanessa Lee's AI use case

Vice President of Product at Shopify

Keyword-to-store setup workflow where a merchant describes a store and AI generates complete editable store layouts with copy and imagery.

The problem

What was broken before AI

A new merchant often had to pick a theme, decide page structure, write homepage copy, choose imagery, and configure the look of a store before they had enough momentum to know what they wanted.

What changed

What the use case made possible

AI shifted the first step from manual layout assembly to intent capture: the merchant describes the business, Shopify generates store directions, and the merchant edits from a starting point instead of an empty canvas.

Why this matters

Why this use case is worth studying

This points to a useful product design move for AI tools: make the first draft visual, editable, and close enough to the real work that the user can make decisions immediately. In ecommerce, a blank store is cognitively expensive because every choice feels connected to brand, trust, conversion, and taste. A generated layout lowers the cost of beginning while still leaving the merchant responsible for product accuracy, positioning, and final publishing.

Use this when

When this pattern applies

Use this pattern when users need to set up something complex, visual, and business-critical, but the old workflow begins with empty fields, manual layout choices, or too many configuration decisions.

Exponential Builder analysis

01

Intent is a better first input than configuration.

When users are early in a setup flow, they often know what they are trying to sell before they know what sections, templates, or settings they need.

02

Multiple drafts create faster taste decisions.

Three generated directions give the merchant a practical comparison set, which helps reveal preferences that would be hard to specify in a form.

03

AI onboarding should end in an editable workspace.

The useful handoff is a draft the user can keep shaping, with review checkpoints for the parts where accuracy, legality, and brand trust matter.

Who this is for

Best fit

Ecommerce founders launching a first store

Local businesses moving online

Product teams designing AI onboarding flows

SaaS teams with complex setup checklists

Designers exploring AI-generated starting points

Operators who want faster first drafts without skipping human review

What to avoid

Mistakes and warnings

Where this pattern can go wrong if you copy it too literally.

Do not treat generated storefronts as launch-ready without merchant review.

Watch for generic copy that sounds plausible but fails to express the actual brand.

Generated imagery may need rights, relevance, and product-accuracy checks.

A beautiful layout still needs product data, fulfillment settings, checkout, taxes, policies, and customer-support basics.

Too many generated options can slow the user down; start with a small set of strong directions.

If the merchant input is vague, the generated store may encode assumptions that need to be surfaced and corrected.

Public workflow preview

The shape of the workflow

A high-level look at how the use case works, with the reusable pattern made clear.

01

Describe the store

The merchant enters descriptive keywords or a short description of the business.

02

Generate layout options

Shopify’s AI Store Builder creates multiple storefront directions with page layout, text, and imagery.

03

Pick a direction

The merchant chooses the generated version that best matches the business intent.

04

Edit the details

Copy, images, brand choices, products, and sections are reviewed and customized.

05

Prepare for launch

The merchant checks accuracy, policies, checkout, product details, and brand fit before publishing.

Copy the pattern

The reusable idea

Pattern in one sentence

Turn user intent into several editable first drafts so setup begins with selection and refinement instead of a blank page.

Reusable idea

If you run onboarding for any product that requires setup, borrow the shape of this workflow: ask for intent first, generate a few complete options, and let the user edit instead of making them configure everything step by step. The key is to generate a useful draft that lands inside the normal workspace, where the user can keep working rather than treating AI as a separate brainstorming toy.

Steal this workflow

AI setup draft recipe

1

Ask: “What are you trying to launch?”

2

Capture: category, audience, offer, tone, and one differentiator.

3

Generate: three named directions, each with layout, headline, section plan, and suggested imagery style.

4

Let the user choose: “Closest fit,” “Most premium,” or “Most practical.”

5

Convert the selected direction into editable blocks.

6

Add a launch checklist directly beside the draft: claims, product data, policies, payment, shipping, mobile view, and accessibility.

7

Save the user’s edits as signal for better future defaults.

Suggested prompt

“I’m launching an online store for [business/product category]. My products are [short product list]. My target customer is [audience]. The brand should feel [tone/style]. Generate three different storefront directions with a homepage structure, hero headline, section-by-section copy notes, imagery suggestions, and a checklist of what I must verify before launch.”

Field notes

Get new AI use cases in your inbox

A short weekly note on how real people are using AI to save time, make money, build tools, and run their lives.

No spam. Just useful AI use cases.

Related use cases

Keep exploring nearby systems.

Browse all