A thirteen-week view of a company's cash, rebuilt every week, where every number can be traced back to the record it came from.
The problem.
A company closes its books weeks after the month ends. By then the decisions those numbers should have informed are already made.
A weekly cash view surfaces an overspend the week the payroll runs.
Most companies wait until something goes wrong, or hire a turnaround firm to build a spreadsheet only that firm can operate, at a cost that makes stopping harder than continuing. Itô replaces the second.
What it does.
Itô reads a company's own accounting records, works out what happened, and proposes what happens next.
- The result
- A thirteen-week picture of cash: what comes in, what goes out, what is left at each week's end.
- Rebuilt
- Every week, against what the company's books now say.
Every number says what kind of number it is.
On most finance screens a settled fact and someone's judgement look the same.
Settled
It happened, and the books have locked it.
Fixed by contract
A signed schedule says so. No range around it.
Estimated
Worked out from the company's own history; the assumption is named.
A person's figure
Someone replaced it. Their reason and name are attached.
Disputed
The system and a person disagree. Both readings are kept.
Not known
Nothing is known. It is not zero.
An unknown is never shown as a zero.
A week with nothing due and a week where the calculation failed are opposite statements.
A model that shows both as zero has told a lie nobody catches.
An unknown appears as a dash, down to the database, which cannot store a zero that means absence. A total missing part of its input is not drawn like a complete one.
When a person disagrees with it.
Finance teams override forecasts, and they should. They know things the records do not.
When the records later disagree with an override, Itô shows the person's number, marks the disagreement, and puts it in front of someone to settle.
Every week, the same discipline.
The weekly review has one job: decide what to do about cash.
The controller prepares the data. The CFO chairs and owns the decisions.
What moved
Which line, how much, which direction.
What drove it
Open the detail. Which vendor, which customer.
One-off or structural
A cause that has passed, or the new normal?
Does the baseline move
One-off: record it and move on. Structural: change the assumption today.
A variance report that does not change the next forecast is a history lesson.
What it will never do.
Not now, not later.
It does not connect to a bank.
It reads bank activity from the accounting records. Reconciliation is work the company already does.
It never moves money.
It models and schedules payments. It cannot approve one.
It never sends anything on its own.
It produces reports. A person sends them.
It does not write the company's forecast.
It proposes. The finance team owns what they publish.
It will not publish a figure that disagrees with the books.
The report is refused by name, not issued with a caveat.
Where the intelligence sits.
No number in Itô is produced by a model.
How the figures are made
Arithmetic over the company's own records, run the same way every time; the method can be handed to a lender. Forecasting is statistics over the company's own history: collection patterns, payment timing, seasonality.
Where a model is used
Neither place can write a number. At setup it proposes how each account maps into the cash view; a person reviews every proposal before it takes effect. Day to day it answers questions in plain language, from figures and records the person asking may see.
Who runs it.
Implementation is a service. Operation is not.
- Set-up
- A team stands it up against the company's systems, trains the finance team, and hands it over.
- After that
- The only recurring manual work is categorising new accounts and vendors; the controller does that in minutes.
Why it matters.
Cash is where financial discipline is forced. A collections problem shows up in the bank before the income statement.
Conversations with lenders and boards start from a number nobody has to defend.
Noise resolves to signal.
Itô is named for Kiyosi Itô, whose calculus on a process that jitters underlies modern derivatives pricing.
Thirteen weeks, every week, for the finance team rebuilding the same forecast by hand.
Start a conversation