Shopify or a custom AI-built app? Choose the first experiment your business needs

In this guide

1. Should I choose Shopify or build a custom app?
2. Describe what the customer buys
3. Compare responsibilities, not screenshots
4. Use a small decision brief
5. Give the experiment a boundary
6. Count contribution and maintenance
7. Pick one route and learn
8. Questions you might be asking
9. Sources and further reading

Should I choose Shopify or build a custom app?

Start with what the customer buys. Shopify is worth considering for a defined product and order workflow. A custom app may fit when the product is ongoing interactive software. Compare the full delivery, support, maintenance, and operating responsibilities before choosing.

A new business idea can turn into a technology project surprisingly quickly. Before the first customer has seen the offer, you are comparing themes, databases, checkout tools, and automation platforms.

Choose the experiment first. Then choose the system that lets a real person try it.

Describe what the customer buys

If the customer buys a physical item, digital kit, or repeatable product, a commerce platform may cover much of the operational work. If the customer buys access to a calculator, saved workflow, or software behavior that changes with their inputs, a custom app may be worth exploring.

Write down the transaction in plain language. “Pay, download the hosting kit, and use it this weekend” suggests a different first version from “sign in every week to monitor client estimates.”

Compare responsibilities, not screenshots

A store needs product information, checkout, order handling, delivery, and customer support. A custom app may need those commercial decisions plus authentication, data ownership, application maintenance, and incident handling.

AI can help create either experience. It does not remove the need for someone to own the recurring work. Include your attention in the operating model, even when the initial build feels fast.

Use a small decision brief

Compare two ways to deliver this offer: a Shopify store and a small custom application. Start with the customer's task. List the minimum capabilities, setup work, recurring cost categories, maintenance responsibilities, and uncertainties for each. Recommend the simplest first experiment. Use current official documentation for product claims and do not assume we need both.

For a custom route, the familiar division is source code in GitHub, an application hosted on Railway, and saved data or accounts in Supabase when needed. For a Shopify route, keep the first catalog small and test the transaction and fulfillment.

Give the experiment a boundary

Choose a time window, spending ceiling, and observable result before adding features. A first test might ask whether five prospective buyers can understand the offer and use a sample. A later test might offer a paid pilot with clear delivery terms.

These are experiment designs, not predictions about what five people will buy. Record refusals and confusion as well as interest. A polite compliment is different from a purchase or a return visit.

Count contribution and maintenance

Use your own numbers for revenue, variable costs, platform costs, refunds, and support. Keep one-time build effort separate from monthly operation. If a product needs an hour of help for each small sale, that matters even when the hosting bill is modest.

Also decide what would make you stop. Perhaps the problem occurs too rarely, customers already have a simpler solution, or the support burden exceeds the value. Ending an unpromising experiment can free time for a stronger one.

Pick one route and learn

For a product storefront, explore Shopify through our affiliate link. For a custom application, explore Railway through our referral link. Check the current costs and eligibility on the provider's site.

There is no requirement to move every project to one platform. Choose a business question, build enough to answer it, and use what happens next to decide the next investment.

Continue with An AI build experiment: turn one useful idea into a working app.

Questions you might be asking

How can I compare the cost of a store and a custom app?

List setup work, recurring platform and app charges, payment costs, hosting, database usage, support, and maintenance where applicable. Include your time and any specialist help. Use current provider terms and a realistic small pilot rather than comparing only headline subscription prices.

Sources and further reading

• Shopify: Starting a store checklist
• Shopify: Placing a test order
• Shopify: Affiliate program
• Railway: GitHub autodeploys
• Railway: Affiliate program

Sources checked October 19, 2026. Review the linked publisher information for current details.

Put one idea to work

Use the idea revival worksheet to turn a recurring interest into a small test with a clear decision.

Choose an idea worth testing →Explore more from Wealth ExperimentGet the Wealth Experiment field notes →