A finance company that writes the same kind of loan fifty times a month is typing the same numbers fifty times a month. The processing fee, the documentation charge, the interest basis — identical on every one, re-entered on every one, and occasionally mistyped.
So we added reusable loan configurations: set the charges up once, then pick the template when creating a loan and watch the fields fill themselves in. Auto-filled fields carry a small badge so it is obvious which values came from the template and which were typed.
Then an owner asked us to add a switch to turn it off entirely.
Why that is not a disappointing outcome
The instinct is to treat an opt-out request as a verdict on the feature. It usually is not. It is a statement about a different business.
A lender running one product with one fee structure does not have templates to choose between. For them the picker is a row of chips above the form offering a single option, plus an empty-state card nudging them to create a template they do not need. That is not a small cost: it is a permanent addition to the busiest screen in the app, bought for no benefit.
The feature is right for a business with several loan products and several charge structures. It is clutter for a business with one. Both are real, and neither is a misunderstanding of the product.
A feature that helps one kind of customer and taxes another is not finished when it works. It is finished when it can be declined.
The shape of the switch
Turning it off removes the whole Loan Configuration card — chips, empty state, badges — and the create form reads exactly as it did before the feature existed. Not similar: identical.
Two decisions underneath it are worth stating:
The gate is purely in the interface. The configuration table and its endpoints stay available whether the switch is on or off. A business that declines templates today and wants them in six months gets them the moment the switch flips — no migration, no data backfill, no release.
Off is the default. Existing customers were not opted into a changed create screen because we shipped something. They see what they saw yesterday until somebody decides otherwise.
The general rule we keep rediscovering
Almost everything in Vasool is behind a per-business switch, defaulted off: voice entry, approval workflow, offline mode, deposits, rentals, the chat interface. The reason is not caution for its own sake. It is that our customers are not one customer.
A pawn broker with a gold book, a daily-line operator with eighty doorstep collections, a cart renter and a chit foreman are all running "a finance business", and a screen that serves all four equally well serves none of them particularly. The alternative to per-business configuration is not a cleaner product — it is a product that quietly optimises for whichever customer the team spoke to most recently.
The cost is real: every flag is a code path, and every code path is a thing that can break while nobody is looking. The discipline that makes it affordable is the one above — flag off must be byte-identical to before the feature existed, so the untouched path stays genuinely untouched rather than merely similar.
If you run more than one loan product and are tired of retyping the same charges, the loan types page covers what the platform handles, or talk to us.