Invoicing and Accounting for a One-Person Business: One Tool or Two

Solo operators tend to buy software in the order they encounter problems: a time tracker first, then an invoicing tool, then a spreadsheet for expenses, then a panicked subscription to accounting software in March. The result is four systems, three of which disagree about how much money came in last quarter.

The useful split is not invoicing versus accounting. It is between records that have to reconcile to a bank statement and records that describe work. The first category has to live in one place. The second can live anywhere you will keep it up to date, which is usually not the same place.

The records that must agree with the bank

Three things have to be true at the end of the year, and only one system should be responsible for all of them.

Every invoice you issued exists, with a number, a date, and a payment status that matches what actually landed in the account. Every business expense exists, categorized, with a receipt attached. And the difference between money in and money out matches the bank statement to the penny.

That reconciliation is the whole job. Everything else in accounting software is a report on top of it.

This is why invoicing and expenses belong together even though they feel like different tasks. If invoices live in one tool and expenses in another, the reconciliation happens in a third place, usually a spreadsheet, usually in a hurry, usually wrong. The tax return is a by-product of a reconciled ledger, not a document you assemble separately.

Deposits are the standard place this breaks. A client pays half up front, and a naive setup records it as a paid invoice when it is actually a liability until the work is delivered. In a single system the deposit sits against the client and is applied to the final invoice, so the date money arrived and the date it became revenue stay separate. Split across two systems, the deposit lands in the bank, matches nothing, and gets counted as income in the month it arrived, which is both wrong and the kind of wrong that is tedious to unwind later.

What can live somewhere else

Time tracking, project notes, client history, and the pipeline of who might hire you next are all legitimate software needs. None of them are tax records.

Time tracking earns its own tool because the data is granular and the interface has to be frictionless: a timer you forget to start is worse than no timer. Notes and client context earn their own tool because they are read by you, not by an accountant, and because a document that is pleasant to write in gets updated.

The mistake is trying to make the accounting tool do this work. Invoicing software with a built-in project tracker is usually worse at tracking than a dedicated tool, and the data it captures is shaped by billing rather than by what you need to remember.

The reverse mistake is worse: pushing billing data into a notes tool and calling it a system of record. A Notion database of invoices is not a ledger. It has no immutable numbering, no audit trail, and no way to prove that an invoice was issued on the date it claims.

Receipts deserve one specific habit: capture at the point of purchase, not at the end of the month. A photo taken in the shop takes eight seconds and survives; a shoebox of paper in January takes an afternoon and half of it has faded. Whatever you use for expenses should make that capture a two-tap action on a phone, because the friction at that moment decides whether the records are complete enough to be useful.

The integration tax

Connecting two tools is cheap to set up and expensive to own. The setup takes twenty minutes. The ownership takes a few minutes a month, forever, plus an unbounded amount of time on the day the sync silently stops.

Silent failure is the specific danger. A broken integration rarely throws an error you will see. It stops importing, and you find out when a client asks why they were never invoiced for February, or when the accountant asks about a gap in the expense records.

There is a rule that prevents most of this: any number that appears in two systems must have exactly one owner. If hours live in the timer and money lives in the accounting tool, then the invoice total is owned by accounting and the timer is only an input at the moment of invoicing. Once the invoice is issued, the timer's number stops mattering. If you find yourself reconciling the two, the ownership is unclear and the integration will eventually produce a discrepancy you cannot explain.

The quarterly close, priced in hours

Suppose a one-person consultancy with about sixty transactions a quarter: fifteen invoices, forty-five expenses.

In a two-tool setup, the close means exporting invoices from the invoicing tool, exporting expenses from wherever they were captured, importing both into a spreadsheet, and matching sixty lines against the bank statement. Call it three hours, plus an hour of fixing the mismatches that always exist: a duplicate receipt, a payment recorded against the wrong invoice, a bank fee nobody categorized. Four hours a quarter, sixteen a year. At an illustrative 50 units an hour, 800 units.

In a single tool with a bank feed, the matching is proposed automatically and the exceptions are the only thing you touch. Call it forty-five minutes a quarter, three hours a year, 150 units.

The gap is 650 units of your time against a price difference that is usually well under 200 units a year. The single tool wins on arithmetic before you count the reduced chance of an error that costs an afternoon with an accountant.

The exception is when the second tool does something the first genuinely cannot, and the honest list is short: time tracking with client-facing reports, multi-currency invoicing at volume, or a jurisdiction-specific tax filing the accounting tool does not support.

The tax rules that actually constrain the choice

Three questions decide which tool is even eligible, and they are worth answering before comparing interfaces.

Does it handle the sales tax or VAT regime you are registered under, including the filing format your authority expects? A tool that calculates tax but cannot produce the return is half a tool, and thresholds for registration vary by country and by what you sell.

Does it produce the export your accountant wants? Most accountants have a preferred format and a preferred tool, and their preference is worth more than yours. Ask before you buy, not in March. A bookkeeper who can log in directly will charge less than one who reconstructs your year from PDFs.

Does it match your legal structure? A sole trader, a single-member company, and a partnership have different filing obligations, and some tools handle only one shape cleanly. This is the constraint that most often forces a change of tool after the first year, when the structure changes.

The rule of thumb

Put invoicing, expenses, and the bank feed in one tool, and accept that it will be the least pleasant software you pay for. Put time tracking and project notes wherever you will actually use them, and do not connect them unless the connection removes a manual step you do more than once a week.

If you are choosing today, pick the tool your accountant already knows how to read, and buy the smallest plan that includes bank reconciliation. Everything else is a feature list.