For accounting firms and in-house bookkeeping teams we remove the retyping: one intake for source documents whatever channel they arrive on, details read off and posted to the ledger, a review queue driven by confidence rather than by folder, and bank reconciliation that runs continuously instead of at the close. On our accounting project a client month closed in four hours instead of three days, and a document took two minutes of manual work instead of twelve.
Counterparty, amount, tax and date are read off by eye and keyed in again. Around twelve minutes a document, and half of it is not accounting.
Email, messengers, a batch of scans at month end, sometimes paper. Between an inbox and a folder a document goes missing and resurfaces a quarter later.
The statement on one side, the ledger on the other, line by line. A discrepancy surfaces on day three and finding its cause takes another day.
Every client closes in the same week, and in that week the firm cannot take on another client because there is nobody to give it to.
One intake for every channel, details read off and validated against the counterparty card and the contract, and posting without retyping.
A queue built from what the model is unsure about or what disagrees with the contract, instead of a person opening everything in turn.
Statement lines matched as they arrive, with breaks raised on the day they appear rather than at the close.
The system knows what is missing for each client and asks for it, so nobody spends the last week of the month writing reminders.
per incoming document
Client under NDA. The figures are the client's own, comparing the periods before and after launch.
This is the project behind the case above, described the way it happened rather than as a method. An outsourced accounting firm, 10–15 people, 30–40 clients on the books. The bottleneck was not the accounting and not the reconciliation — it was that qualified people spent half the day retyping. The order of the work was set by the money: first the loop that removes the manual entry, then reconciliation, then reporting.
Email, messengers and uploads converge on a single queue. A document stops falling between channels, and whoever chases a client for a missing one can see exactly what is missing.
Details are read off automatically and the document reaches the accounting system without being retyped. Twelve minutes became two.
A person looks at what the model is unsure about, or at what disagrees with the contract, rather than at everything in turn. Low confidence is not an error, it is a route.
Lines match as statements arrive, so a discrepancy surfaces on the day it appears rather than on the third day of the close.
Statements are assembled and sent from inside the system, and the replies land in the same exception queue instead of in somebody's inbox.
Recurring reports assemble themselves with an audit trail: for every figure you can see which documents it was built from.
Three months, of which most went into two unglamorous things: the exchange with the accounting system, and the source-document formats, which differ for every client. What deliberately stayed with people: the decision on a disputed document, the methodology for a non-standard transaction, the conversation with the client, and the signature under the reporting.
One to two weeks. Fixed, and credited in full against the build.
A fixed quote against the scope the audit produced.
On this project. Two to four months is the range across our cases.
Model and infrastructure usage plus maintenance. Moves with volume.
Automation prepares, validates and explains an entry. Responsibility for what is posted and for the reporting stays with a named person, and nothing in the design blurs that.
Where a rule can be written as a rule — thresholds, tax treatment, coding — it stays a rule. Models are used for reading documents, not for deciding treatment.
A report line can be opened down to the documents it was built from, because in this sector reconstructing a number is a routine request rather than an exception.
For a firm holding several clients' books, access, retention and processing region are set per client rather than per firm, and the intake keeps them apart from the first minute.
The single largest block of manual time in the sector, with a saving that can be measured per document on the first day.
Matching as statements arrive rather than at the close, which moves the discrepancy hunt off the critical path of the month.
Knowing what each client still owes you and asking for it automatically, which is the quiet reason closes slip.
We do not decide accounting methodology or give tax advice. The system applies the rules your accountants set.
We will not replace a working accounting system to make an automation easier — the platform sits alongside it.
If counterparty cards and the chart of accounts disagree between clients, that is a data project first, and extraction on top of it will inherit the disagreement.
Each page states the workflow, the systems it integrates with, what it costs and when it is the wrong choice.
Reads incoming invoices, checks totals, tax and supplier against the ERP, and posts only what clears the rules.
Pulls figures from the accounting stack into a recurring report, with variance explained in plain language.
Sorts mixed incoming documents by type and routes each one to the queue, folder or system that owns it.
No. On our accounting project it stayed the system of record and reporting, and the platform exchanges data with it. What we replace is the retyping in front of it, which is where the hours were.
We don’t propose an off-the-shelf product either. The audit is there precisely to see where your process differs from the ordinary one, and the build is shaped around that difference. And if it turns out a standard tool covers you well enough, we’ll say so plainly — the process map stays with you in any case.
It’s a fair thing to ask, and responsibility matters more here than accuracy. A person answers for it, and the system is built so that they can: a doubtful case is never posted quietly, it goes to a review queue. Low confidence isn’t an error for us, it’s a route. Around 70% clears straight through and a person looks at the rest — and where exactly that line sits is yours to decide.
That’s a common requirement, and a reasonable one. We fit the setup to your data rules: if nothing may leave, the model runs locally inside your own perimeter, and the documents never cross it.
That’s not unusual, and it’s fine. Integrating without an API is the most underestimated line in a quote, which is why integration surface comes first among the four cost drivers, and why we don’t name a fixed price before the audit. Working without an API is perfectly possible — files, exports, email — it simply costs more, and it’s better to know that early. Which is why, fairly often, we build the API for you.
It happens often, and usually for one reason: the pilot was measured on the happy path and the exceptions were left for later — though the exceptions are where most of the work turns out to be. So we look at the share that clears without a person rather than at extraction accuracy, and we agree in advance who handles the rest, and how.
No, that isn’t what this is about. What goes is the retyping, not the people: decisions stay with a person — the disputed document, the non-standard transaction, the conversation with the client, the signature under the reporting. On one accounting project, closing a client month went from three days to four hours, not because anyone was let go, but because a qualified specialist stopped keying in details by hand. The time that frees up most often goes into growth: more clients with the same team, without the costs rising alongside.
It’s a fair question, and sometimes doing it yourselves really is the right answer. We say so when the volume doesn’t justify a build: below roughly 300 documents a month the arithmetic usually doesn’t work. The calculator on this site includes running costs, so you can weigh that up before you ever talk to us.
They do change, and that’s exactly what we build for: the model is a replaceable part here, not the foundation. The source of truth is our own database — the document, the counterparty, the entry. The model is attached to the side, and we update it whenever something better appears; for you that’s planned maintenance, not a rebuild.
We use cookies for analytics — to see which pages bring enquiries. Nothing else, and nothing before you agree. Cookie Policy