Open any budgeting app and the first screen asks the same thing. Connect your bank. It is presented as setup — a small chore between you and the useful part — and almost everyone does it, because the alternative appears to be typing every transaction in by hand forever.
What that screen is actually asking for is worth saying plainly. You are handing your online banking credentials to a company you have never dealt with, so that a third company you have never heard of can log in as you, on a schedule, and read everything. The app you chose is rarely the one holding the keys. An aggregator is — Plaid, Finicity, MX — and the app is its customer.
The record on this is not speculative
In 2022 Plaid settled a $58 million class action (Courthouse News Service, July 2022) after plaintiffs alleged it collected more data than it disclosed and sold transaction histories without explicit consent. Plaid has since revised its practices and states that it does not sell user data. The settlement still stands as a factual record, and it is the reason a growing number of otherwise unbothered people now decline that first screen.
That refusal used to have somewhere to go. Mint let you track manually and it was free. It shut down in March 2024, and the products that absorbed its users — Monarch, Simplifi, Copilot — all require the bank connection. If you left specifically to avoid the aggregator, the mainstream replaced your tool with the thing you were leaving.
What we built instead
myClerkBook has no bank connection. Not an optional one, not one behind a setting — there is no integration in the codebase and there will not be one. It is a permanent architectural constraint, and the test any future feature has to pass is simple: does this need to reach into an account to work? If yes, it is out of scope, whatever the commercial argument for it.
That constraint is what makes the privacy claim checkable rather than promissory. A company with your bank credentials can promise not to misuse them. A company that never asked has nothing to promise about, and nothing to lose in a breach it did not plan for.
The line is drawn at the credential, not at the data. A statement you export from your own bank and choose to upload is yours to give — reading it requires no login, no aggregator and no standing connection to anything. That sits inside the principle rather than at its edge.
What it costs you
Honesty about the trade-off, since this whole argument is about honesty: automatic transaction import is genuinely convenient, and we do not have it. You record what you want recorded, or you upload a document. Some people will find that too much work, and they should use one of the other products — they are good, and the argument here is not that they are badly built.
It is that they all made the same decision, and nobody was offering the other one.