How does Accountable work?

You type in what you spend, tell it about the payments that repeat, and it does the arithmetic. There is nothing to connect and no setup step between signing up and logging your first coffee — and if you are arriving with history, the importer reads whatever file your bank offers: CSV, Excel, OFX, QFX or QIF.

How do I add a transaction?

You type an amount, pick a category and save, which takes about four seconds. There is no bank connection to set up, and if you are already keeping records somewhere else you can import them from whatever file your bank or old app hands you.

Seventeen sensible categories are waiting on day one, and you can rename, recolour, add to or archive every one of them.

A date, a description and notes are optional. So is a receipt photo, stored privately and served over signed links that expire.

Already have history somewhere? Import it — your own export, or one from Mint, YNAB or Monzo, or a raw bank statement — as a CSV, an Excel file, the OFX, QFX, QBO or QIF download banks call “export to Quicken”, or the PDF statement itself. Nothing is written until you have seen a preview of exactly what will change.

Entries made offline queue on the device and sync the moment you reconnect, so a tunnel is not a reason to stop logging.

How do recurring entries work?

You describe a payment once — rent on the 6th, salary on the 15th and 30th — and Accountable books it on schedule without you opening the app. Rules run weekly, fortnightly, twice monthly, monthly, quarterly or yearly.

A rule can be paused and resumed. Cancelling the gym for a month should not mean deleting its history.

When an amount changes, you record the new amount from a date forward. The rule keeps its schedule and its past.

You can end a schedule from today onward without touching anything it already booked.

What does Accountable show me?

A calendar where green days made money and red days did not, and a dashboard that says what changed since last month. Budgets go amber at 85% and red past 100%, while there is still something you can do about it.

The dashboard opens with four figures — balance, in, out and net — so "can I afford this" is answered before you scroll.

A six-month cash-flow chart shows the shape of the year rather than just this month.

The category breakdown names your top five and lumps the rest, because a pie chart with nineteen slices tells you nothing.

What does the calendar show that a list of transactions cannot?

It shows the shape of the month. A list tells you what happened in order; the calendar colours in the days you made money and the days you did not, so a bad week is something you notice rather than something you work out.

Do I have to connect my bank account?

No, and you cannot — Accountable has no way to connect to a bank. That is the trade it makes: a live connection means handing an aggregator your banking credentials, and it also means the numbers arrive without you ever looking at them. You can still bring existing history in yourself: download a statement from your bank or your old budgeting app — CSV, Excel, OFX, QFX or QIF all work — and import it, with a preview of every row before anything is saved.

How long does this actually take each day?

About four seconds per transaction, which for most people is under a minute a day.

A typical entry is an amount, a category and a tap. The date defaults to today, the description is optional, and the form remembers nothing you have to re-choose. On a phone the whole thing is a sheet that opens over whatever you were doing; on a laptop the keyboard shortcut opens it from any screen without touching the mouse.

The realistic weekly total is a few minutes, not the half-hour people brace for. What takes longer is the monthly look — sitting with the calendar and the category breakdown for ten minutes and noticing what actually happened. That part is optional in the sense that nothing forces it, and it is also the entire point, because logging without ever looking is just data entry.

The four seconds is a feature rather than a cost, which is the part that sounds like a rationalisation until you have done it for a month. An app that fills itself in produces a categorised list you scroll past. Typing "$8.40, lunch" is the moment you notice you bought lunch, and noticing is the mechanism that changes anything.

What happens on the first day — is there a setup process?

There is no setup step: you sign up, a sensible set of spending categories is already there, and the next screen you see can take a transaction.

No accounts to name, no budget to design, no connection wizard, no onboarding tour that has to be dismissed. A new account arrives with its categories already created, and you can rename, recolour, delete or add to them at any point without anything downstream breaking.

The one decision worth making on day one is a small one: do not set any budget limits yet. A cap chosen before you have measured anything is a guess, and missing a guess in week two is the single most common reason somebody abandons a budgeting app. Record a month first, then set two or three limits on the categories that surprised you.

If you are arriving with history, the import is the first thing to do rather than the last — a dashboard with a month of real data on it is worth looking at, and an empty one is not.

What happens if I forget to log for a week?

Nothing breaks and nothing is lost — you back-date the entries from your bank app or import the week from a statement file, and the calendar fills itself back in.

Every entry carries its own date, and that date is editable, so catching up is not a special mode or a repair operation. Adding Tuesday’s coffee on Friday puts it on Tuesday, and the calendar, the dashboard and every total move accordingly. The dashboard does keep a quiet count of days logged in a row, and a gap simply ends it — there is no penalty attached, and after a few days away the same line turns into an offer to help you catch up.

For a longer gap the importer is faster than typing. Download whatever your bank gives you for those dates, drop it in, and the preview shows every row before anything is written — including which rows duplicate what you already logged, which is the thing that makes catching up safe rather than a source of double-counted weeks.

And it is worth saying plainly that a week missed is not the failure people treat it as. The reason manual tracking works is the attention, not the completeness, and a month with a gap in it still tells you more than no month at all. The people this does not suit are the ones who will not come back after the gap — which is a real limit and is covered below.

Does it work on a phone, and does it work offline?

Yes to both: it installs to a phone home screen as an app, and the shell keeps loading with no connection, though new entries need one to save.

It is a web app that installs. On iOS you add it to the home screen from the share sheet; on Android the browser offers to install it. Either way you get an icon, a full screen with no browser chrome, and the same app — there is no separate mobile version to be behind the desktop one, because there is only one codebase.

A service worker caches the shell and the fonts, so opening it on a train with no signal gives you the interface rather than a browser error page. What it cannot do offline is save: a transaction has to reach the database, so an entry made with no connection will not stick. The honest workflow with no signal is to note the amount somewhere and enter it later, back-dated.

The layout below 640 pixels is genuinely designed rather than shrunk. The calendar cell height, the overflow menus in place of button rows, the search that collapses to a single control — each of those exists because the full-width version did not fit on a 375-pixel screen, which is the screen most of this gets read on.

What happens to my data if I stop paying, or want to leave?

Everything you logged stays readable and exportable for ever, subscribed or not, and the CSV export is one click with no fee attached to it.

This is worth being specific about, because a lapsed subscription means very different things across this category. Here it means the account goes read-only: every transaction, category, income source and recurring rule stays exactly where it was, the Dashboard and Activity keep showing them, and the export keeps working. What stops is writing — adding and editing — along with the views built on top. Your own records are never the leverage.

That is enforced rather than promised. The database policies deciding what a lapsed account may do gate inserts and updates, and deliberately do not gate reads or deletes — so getting your data out, or deleting it for good, cannot be broken by a billing problem or by an edit to a pricing page.

The export is a plain CSV of everything: transactions, categories, income sources and recurring rules. It is the same format the importer reads, so it moves to another Accountable account cleanly, and it is ordinary enough that a spreadsheet, a competitor, or an accountant can open it with no conversion.

Deleting the account deletes the data. That is a real deletion rather than a flag, which means it cannot be undone and is worth exporting before doing.

How do I bring two years of history with me?

Download whatever files your bank or old app will give you and import them one at a time — the reader takes CSV, Excel, OFX, QFX, QBO, OFC, QIF and text-bearing PDF statements.

Exports from budgeting apps are the easy case: Mint, YNAB, Monzo and an Accountable export are recognised by their column layout, categories and all, so there is nothing to map. A raw bank file is the harder one, because a bank supplies no categories at all — the merchant is guessed from the description, which lands roughly 95% of a real statement, and everything else arrives as Uncategorised rather than as a confident guess that is wrong.

Two years is usually more than one file. Most banks cap the transaction export at somewhere between twelve and twenty-four months while keeping PDF statements for far longer, so the practical route is a spreadsheet export for the recent stretch and PDF statements for the rest. Both end up in the same preview.

Nothing is written until you confirm the preview, and duplicates are found on the date and amount rather than on the description — because your bank calls the same deposit something entirely different from what you called it. Rows that look like something you already have are flagged and can be imported anyway, since two genuine $4.12 purchases on one day is a real thing that happens.

What can Accountable not do?

It cannot fetch your transactions, track investment balances automatically, or work for somebody who will not log — and if any of those is what you need, another app is the right answer.

There is no bank connection and there never will be, which is the whole design and also its cost. If the appeal of a budgeting app is that it fills itself in, this is the opposite of that, and a month spent finding out is a month wasted. Something that syncs is the honest recommendation for thirty card transactions a week across four accounts.

Investment balances are the second real gap. A market value changes daily and nobody is going to re-type it, so what this offers is a dated balance adjustment — a real transaction on the day the value moved, which keeps the net-worth chart honest without pretending to a live feed.

And it is not built for a shared household budget where several people all have to remember to log. The weakest link sets the accuracy of the whole ledger, and two people with one shared account and different habits is exactly the situation manual entry handles worst.

What is left is what it is genuinely good at: fewer, more deliberate transactions, a record you believe because you made it, and no third party holding a credential to your bank.