Nine Months to Build Login and Invoices. Then You Can Start on Your Actual Idea.

Every small team pays the same entry fee: most of a year rebuilding backend, billing, entitlements, admin and mobile before the product starts. AI made writing each piece faster — it didn't tell you which pieces you forgot. There's a third option between building it all and bending an e-commerce platform into a shape it hates.

The plumbing tax that quietly kills good products — and how to skip it.

Here's a thing that happens to almost every small dev team with a good idea. They start on a Monday, full of energy, and nine months later they still haven't built the product. They've built login. And roles. And a subscription state machine. And invoices, and tax on the invoices, and the webhook handler that fires when Stripe sends the same event three times. And an admin panel, because somebody eventually has to issue a refund.

The idea — the actual reason the company exists — is still sitting in a document somewhere, waiting for its turn.

"But AI writes that now"

Fair challenge. It's 2026; Claude Code can produce a subscription service before lunch. Isn't the plumbing problem solved?

It can write a subscription service before lunch. Having one is a different thing entirely. The afternoon doesn't cover proration when someone upgrades mid-cycle. Or dunning on the third failed card retry. Or what happens to a user's access during the grace period between a failed payment and an actual cancellation. Or tax that varies by country and product type — including the discount line that has to carry negative tax per rate, or your invoice totals quietly drift by a few cents until an accountant finds it. Or idempotency when a webhook arrives twice. Or the audit trail your first enterprise customer demands.

None of this is intellectually hard. It's huge, dull, and unforgiving — hundreds of tiny decisions where 95% correct means wrong, because the missing 5% is money. AI makes writing each piece faster. It doesn't tell you which pieces you forgot, and it can't compress the months you'll spend discovering edge cases in production with real customers' cards.

Add it up honestly — backend, billing, entitlements, admin, mobile — and a small team doing it properly is looking at most of a year. With great tools. That's not doom-mongering; it's an inventory.

The two escape routes, and why both cost you

Build it anyway. Spend the year. Watch competitors ship. Become an expert in VAT, which was not the plan.

Bend an e-commerce platform into a SaaS. The popular one, and the sneaky one. E-commerce is built around a different noun: an order for a physical thing that ships once. What you're selling is a recurring relationship with an entitlement attached. You can force the shape — everyone does — and you'll pay rent on it forever: a plugin for subscriptions, another for the portal, a webhook stapled to your real backend to sync who's allowed to log in, and an ever-growing pile of translation code between the platform's worldview and yours. You didn't avoid the work. You relocated it somewhere worse.

Option three: don't start from zero, and don't start from the wrong shape

VBWD is built on the premise that there's an obvious third answer: start from a framework that's already the right shape, then add the only part that's actually yours.

What you get on clone: a real backend API (Python/Flask on PostgreSQL and Redis, properly layered, not a toy). Subscription billing with plans, trials, the full lifecycle, invoices, and tax that reconciles. Access entitlements and RBAC — a permission check you call instead of write. A platform-wide event system, so your feature reacts to payment.captured instead of you threading logic through someone else's checkout. Payments behind a provider-agnostic interface — Stripe, PayPal, regional processors, non-custodial crypto — swappable by config. A Vue 3 web frontend plus a complete admin backoffice, so the panel you'd have built in month eight already exists. A CMS, so your marketing site isn't a bolted-on subdomain that knows nothing about your pricing.

And iOS and Android SDKs against the same backend: one API, one billing system, one entitlement model, three clients — instead of three subtly different implementations of "is this user a subscriber?" drifting apart over two years.

The repo also ships instructions for AI coding agents, so Claude Code arrives in a codebase that tells it how to extend things correctly instead of guessing.

The rule that keeps it from owning you

Frameworks that give you everything usually take something back: your freedom to change anything. VBWD's answer is a rule enforced by the build itself — the core is agnostic; only plugins are gnostic.

The core knows users, money, events, entitlements. It knows nothing about gyms, tarot decks, datasets, pharmacies, or your vertical. Your domain lives in a plugin that registers routes, models, migrations, and providers through seams the core already exposes. An automated check fails the build if core code so much as mentions a plugin.

That's why adding your feature doesn't mean forking the billing engine, and why upgrading doesn't mean re-applying your patches. You extend at the designed seams; everything else stays somebody else's maintenance burden.

What day one looks like

git clone https://github.com/VBWD-platform/vbwd-sdk-public.git vbwd
cd vbwd
./recipes/dev-install-ce.sh
make up

The recipe handles clone order, submodules, build sequence, and the database. When it stops, you have a running platform — API, database, cache, user app, admin backoffice, migrations applied, roles seeded. Configure your plans, prices, payment provider, and access levels; then write the plugin that holds your actual idea.

The difference is what your first line of real code is. It's your product — not class User.

The honest limits

It's self-hosted. Someone runs the server, applies updates, takes backups. That's a genuine cost, and for some teams a managed SaaS with a monthly invoice is the better trade — say so out loud before you choose. You'll also have to learn the platform's seams; anything this large has a shape, and working with it beats fighting it. And if your product genuinely is orders for physical things that ship once, use a real e-commerce platform — that's what it's for.

"Clone, install, configure, boom" means the foundation is done, not your product. Your idea still needs building. That was always the interesting part — it's just no longer the part you reach nine months late.

The point

Every small team pays the same entry fee right now: most of a year rebuilding a foundation hundreds of other teams already rebuilt, slightly differently, slightly wrong. That fee kills good ideas — not because the ideas were bad, but because the runway ended somewhere around invoice numbering.

Nobody ever won a market because their tax rounding was elegant. Clone the framework, install it, configure it, add your plugin — and spend the year on the thing only you can build.

VBWD is source-available under BSL 1.1 and free commercially while VBWD-attributable sales stay under the value of 6.7 BTC a year.

Learn more about VBWD

VBWD is a self-hosted, source-available platform for building subscription products, marketplaces, and AI-powered apps. Explore it further:

- 🌐 Website and documentation: vbwd.cc — see the plugins, architecture, and developer docs.
- 💻 Source code and plugins on GitHub: github.com/VBWD-platform
- 🎥 Watch VBWD in action: demo video 1 and demo video 2
- 💼 Follow the project on LinkedIn: linkedin.com/company/vbwd https://redrobot.online/2026/07/16/nine-months-to-build-login-and-invoices-then-you-can-start-on-your-actual-idea/