Opinions

Build your own garden

Josselin Liebe

Published on Oct 30, 2025 Updated on Oct 30, 2025

If your business is software, your moat is ownership. Building your app primarily on Shopify or Whop might look like a shortcut to distribution-but it places your product, pricing, data, and roadmap inside someone else’s garden. That’s convenient until the landlord changes the rules.

This isn’t anti‑platform. It’s pro‑ownership. Use platforms to acquire customers and integrate where it makes sense-but keep your core on ground you control.

TL;DR

Platforms are rented land: policies, fees, and APIs can change overnight.
If your product depends on a single app store, you’re one policy email away from churn.
Own your domain, data, payments, and UX; integrate with Shopify/Whop at the edges.
Design for portability: your app should survive if a platform bans, rate‑limits, or pivots.

The risks of building in someone else’s garden

1) Policy changes can rewrite your business

App stores optimize for their ecosystem, not for your margins. A ToS update, a new fee tier, or a “quality” guideline can force product changes or kill features you rely on.

2) Fee creep and margin compression

Platforms rarely get cheaper. Between platform fees, app store rev‑shares, and payment constraints, margins get squeezed. Your pricing freedom shrinks as costs rise.

3) API limitations and breaking changes

Critical endpoints can be throttled, deprecated, or scoped behind new permissions. Your roadmap becomes a negotiation with someone else’s product team and release cycle.

4) Discovery dependency is fragile

You don’t own search rankings in an app store. A small change to ranking factors, categories, or editorial curation can crush installs-and there’s no export for that.

5) No real control over checkout and payouts

If the platform governs payment flow, you inherit their dispute rules, payout timing, and risk scoring. You can’t optimize conversion, retries, or dunning on your terms.

6) Data access is never full access

You’ll always have partial visibility-missing identifiers, delayed events, or blocked joins across systems. That compromises analytics, attribution, and support.

7) Switching costs compound over time

The longer you stay embedded, the harder it is to leave. Every feature built directly on proprietary primitives increases your migration cost.

Shopify vs Whop: same pattern, different shapes

Shopify: Amazing commerce backbone-but the app store owns distribution and takes a cut. You’re boxed by checkout rules, API scopes, and whatever Shopify prioritizes this quarter.
Whop: Fast way into the creator economy-but you’re anchoring value inside a marketplace that controls hosting, routing, and access. If your app is a Whop page, Whop owns the frame.

In both cases, the platform is the venue-not the business.

When platforms do make sense

Use them for:

Acquisition: Treat app stores as channels, not homes.
Integrations: Meet customers where they work (Shopify admin, Whop communities) with connectors.
Payments optionality: Accept platform payments as a mode-not as your only pipe.

If the platform disappeared tomorrow, your product should still work via your own domain with minimal changes.

The owned‑garden architecture (practical)

Design your stack so the platform is a plugin, not your foundation:

Own the surface: Public site and app at your domain (e.g., Next.js/Remix/Astro for web; Expo for mobile).
Own payments: Stripe directly to your account; handle webhooks in your backend. Map customers to external identities (Discord/Telegram) in your DB.
Event core: Durable, idempotent workflow engine (queues/state machines) to process billing, access, and audits independently of any platform’s timing.
Integration layer: Thin adapters for Shopify/Whop APIs. Keep them stateless and replaceable.
Data warehouse: Your analytics and attribution in your store (clicks, trials, LTV). Don’t depend on app‑store dashboards to run the business.
Portability tests: Regularly simulate “platform down/permission revoked” and confirm the product still functions at your domain.

If you must start on Shopify or Whop

Keep a standalone product at your domain from day one.
Store your canonical state (users, subscriptions, entitlements) in your DB, not theirs.
Build an import/export path for every object you create through platform APIs.
Avoid proprietary UI/hosting as your main surface; link back to your app for deep workflows.
Capture emails and ownership of the customer relationship at signup.

Founder mindset: rent channels, own the house

Platforms give reach, credibility, and trust-use them. But durable businesses are built where you control the rules: domain, database, payments, and roadmap.

Build your own garden. Integrate with your neighbor’s when it helps-but don’t mistake their land for yours.

If you’re monetizing a community on Discord or Telegram and want ownership with low fees, build on your domain and integrate platforms at the edges. Tools like Sublyna are designed with that philosophy in mind: your Stripe, your data, your rules.