White label, no code or custom build: choosing a route to SaaS

List the ten steps your customer repeats. Now find a ready product doing the same job and count the steps it handles without any change, from sign-up through to the final invoice. Nine fit? White label is probably enough.

If four fit and six need workarounds, you are really selling your own method. Customers pay for that method. A rented product will fight it at every release, and each vendor update can break the workaround your team built last month.

Write down the three rules that make your service different. Examples are a special approval order, a pricing formula tied to usage tiers, or a regulated record such as a signed consent form. Those rules must live somewhere you control.

White label means you rebrand a vendor’s product and resell it. Launch is quick. The vendor carries the engineering, and you inherit their roadmap, their pricing changes and their outages, including the night their payment gateway fails.

No code means you assemble forms, tables and automations on a platform such as Airtable, Bubble or Softr. It proves demand with ten users. Cost and limits rise sharply once each customer needs its own data space, its own permission rules and its own automation quota.

Custom build means your own codebase, hosted where you choose. The first release takes longest. Every rule, report and billing step is yours, from the tenant table in the database to the wording on the invoice footer.

Ask where one customer’s records sit relative to another’s. Some platforms keep every account in one workspace, separated only by a filter on a customer column. That is fine for a newsletter tool. It is risky for payroll, patient notes or contracts.

Check branding depth too. Ask whether each customer can have its own logo, colours, sender address and domain, and whether invoices and emails carry their name instead of yours or the vendor’s.

Ask for the vendor’s backup and export terms in writing. A customer asking for their data back after cancelling is a normal request, and you must be able to say yes without a week of manual exports.

Every route has a ceiling. Find yours before you hit it. It may be a record limit, an API rate cap, a missing integration or a per-seat price that eats your margin.

Moving later is painful if your customers’ data sits in a format you cannot export cleanly. Test an export now. Open the file and check that the links between customers, invoices and users survive, and that every attachment comes out too.

A common path is no code to prove demand, then a custom build once ten or twenty paying accounts expose the real rules. Plan that move early.

Score each route from one to five on four tests: workflow fit, data separation, branding and exit. Add the scores. A route that scores one on any test is out, whatever its total.

Take the shortlist to one real prospect and watch them use a prototype. Watch where they hesitate. Their confusion tells you more than a feature table. Then commit to the route that passes the tests and write down the ceiling that would make you reconsider.