Daily bank position
Totals by currency, company, and account, with the unclassified items always in view.
orbit position takes the bank statement in whatever format each bank delivers, extracts the movements, and classifies them with an AI auditor that states its confidence level. What falls short of confidence lands in a review queue; the rest goes through. On top sits the consolidated position by currency, company, and account.
The finance team of a group with several companies and several banks builds the daily position by hand: download each bank statement, paste it into a sheet, classify every movement by company, account, cost center, and concept, and split income from expenses and transfers. Every bank delivers a different format. The sheet becomes the source of truth and nobody knows, mid-morning, how much is actually available.
A platform where the bank statement comes in whatever format and every movement classifies itself. The AI auditor proposes cost center, operation concept, and accounting concept, each with its confidence percentage, and whatever falls short of the threshold goes to a queue where a person confirms or corrects. Every movement carries a fingerprint that prevents duplicates if the same statement is uploaded again.
From the bank file to the consolidated position, with nothing pasted into a sheet.
CSV, PDF, or Excel. CSV is processed natively; PDF and Excel go through the document extraction service.
It proposes cost center, operation concept, and concept, each with its confidence percentage.
An editable grid shows what was extracted and proposed, and whatever misses the threshold goes to the auditor's queue.
The dashboard totals by currency, company, and account, splitting income, expenses, transfers, investments, loans, and fees.
The modules a finance team uses daily in production today.
Totals by currency, company, and account, with the unclassified items always in view.
Each bank's statement comes in as it is and comes out as classified movements.
Predictions with their confidence, so a person approves or corrects only the doubtful ones.
Tax receipt and credit note blocks per movement, with their net payable.
Companies, accounts, banks, cost centers, concepts, and projects managed from the platform.
The filtered table exports to the bank position format the team already uses.
Five differences from building the position in a spreadsheet.
Request demoCSV, PDF, or Excel come in the same way and come out as classified movements.
Cost center, operation concept, and concept, each with its confidence level.
What falls short of confidence lands in a queue; the rest goes through untouched.
Every movement carries a fingerprint of account, date, amount, and description: re-ingesting the same file duplicates nothing.
Totals by currency, company, and account on the dashboard, not in a file someone has open.
An honest comparison between building the position by hand and building it with orbit position.
It runs with the finance team of a Mexican financial group with several companies. The architecture serves any multi-bank treasury.
It is where it runs today, every day.
Applicable by architecture. No case yet.
The daily position by currency and company, the already-classified movements, and the AI auditor's queue with its confidence level. Recorded in the platform, with the client's data blurred.
In 30 minutes we tell you whether orbit position fits your process, what it integrates and what it doesn't, and what a pilot looks like. If it doesn't fit, we tell you that too.
Run the discovery with rocky