13-week cash flow forecasting software keeps a rolling weekly view of the quarter ahead by reading your accounting ledger, so nobody has to type the forecast in. Each week it does six jobs a spreadsheet leaves to the finance team: it brings in reconciled balances, invoices and bills, places each receipt in the week the cash is expected, fills the closed weeks with what happened, shows the date the balance crosses a minimum you set, holds a what-if version on top of the forecast, and exports the 13-week view. The judgement stays with the team: when a slow customer will pay, what has been agreed but not yet invoiced, and what to do about the tight week.

This page follows one working week and takes those six jobs in order, using Float as the example because its help centre documents each one. It does not cover how to build the forecast, which is in our guide to implementing a 13-week cash flow forecast, how far to trust the later weeks, which is in how accurate 13-week cash flow forecasts are, or the spreadsheet itself, which is in the 13-week cash flow forecast template.

What "13-week forecast software" covers

Thirteen weeks is a convention. The Association of Corporate Treasurers describes operational cash forecasts as short term, "which generally covers the next 13 weeks on a rolling basis". ICAEW's guidance for finance professionals describes rolling weekly forecasts over 13 weeks as common in turnarounds and sometimes asked for in loan applications. No accounting standard or statute requires a solvent private company to keep one. Finance teams keep one because cash moves by the week and a monthly view hides the week that matters.

Three kinds of product are sold under the name. Treasury systems connect to bank accounts and are built for groups managing payments, borrowing and currency risk. Planning platforms forecast the profit and loss account and balance sheet and derive cash from them. Ledger-connected cash forecasting tools read the accounting platform a business already keeps its books in and forecast cash directly from invoices, bills and bank transactions. This page is about the third kind, which is the one a finance team of three to six running Xero or QuickBooks Online is choosing between when it weighs software against its spreadsheet.

The difference between a ledger connection and a bank connection decides what the forecast can see. A tool that reads the ledger sees what the bookkeeping has recorded, as at its last import. A tool that connects to bank accounts as a regulated account information service sees balances and transactions as the bank reports them, and in the UK it may refresh without the user present "no more than four times in a 24-hour period" unless the bank agrees to more, with the user reconfirming consent every 90 days. Neither a daily import nor four refreshes a day is real-time, so the useful question for any tool is its schedule.

Float is the ledger-connected kind. It reads Xero or QuickBooks Online, it does not connect to the bank, and it forecasts cash only.

The six jobs, in the order the week runs

Before Monday: the data arrives

In a spreadsheet, the week starts with exports: bank lines, the aged debtors list, the bills awaiting payment, and the bank balance keyed in as the new opening position.

Float imports from the accounting platform once a day, at an hour set in company settings, with a sync button for the moment you have just finished reconciling. The connection is one-way. Float reads the ledger and sends nothing back, so nothing done in the forecast changes the books. What arrives is reconciled bank transactions, open invoices and open bills. The opening position is the reconciled balance of the accounts you choose to include, which is the "Balance in Xero" or "In QuickBooks" figure and not the bank's own balance.

That last point is the one to hold on to. A bank line nobody has reconciled is a cash movement the forecast cannot see, so the forecast is as current as the last reconciliation. The full list of what a connection takes over, step by step and platform by platform, is in our guide to how much manual data entry forecasting software removes.

Monday: each receipt sits in the week you expect the cash

A due date is what the contract says. An expected date is what the customer does. ICAEW's eighth principle for cash flow forecasting is to "assign expected, worst case and best-case dates for receipts and payments", and the British Business Bank's guidance is to "put the figures in for when you know clients will pay invoices". No professional body prescribes how the expected date should be worked out. The Association of Corporate Treasurers points to "past trends or patterns of behaviour" and to working with whoever runs receivables.

In Float every open invoice and bill carries an expected date. It comes from the accounting platform where one is set there, and falls back to the due date where it is not. From there the team can move one item by seven or thirty days or to a chosen date, re-date a batch in one action, or split an invoice into part payments on separate dates. A change made to a due or expected date in the accounting platform overwrites the date set in Float at the next import.

Two behaviours are worth knowing before the first Monday. An overdue invoice with no new expected date is assumed to be paid today and is marked with a warning triangle in the app, so the overdue list is the first thing to clear each week. And for Xero users, Smart Expected Dates applies each customer's average days late, taken from three to twelve months of your own payment history, to new invoices and bills as they import; the date is set once, is never recalculated, and can be overridden. On QuickBooks Online that feature is not available, so the team sets expected dates in Float for anything the platform does not already carry.

Through the week: closed weeks fill with what happened

In a spreadsheet, someone overwrites last week's forecast cells with actual figures, or adds a second row for them, and then extends every recurring line by a week.

In Float the estimates for cash that has no invoice or bill yet are budgets, and a budget is a placeholder. As invoices, bills and reconciled transactions arrive they fill it. Where the actual figure is larger than the budget, the actual is used, and at the end of the month actuals replace budgets altogether. A repeating budget for rent or payroll carries on at its set frequency until its end date, or indefinitely if it has none, so the far end of the view does not need rebuilding by hand.

What this does not give you is a record of what last Monday's forecast said. Float's budget variance report compares budget with actual for the previous month or the last three months, by account. A week-by-week comparison of forecast against outturn, the habit our accuracy guide describes, needs a saved copy of each week's forecast, and the weekly export described below is the way to keep one.

Any day: the forecast is read against a floor

A column of closing balances does not tell anyone what to do. The Association of Corporate Treasurers puts the requirement as establishing "the levels of excess or shortfalls at various points in time, versus plan/baseline/zero balance". A spreadsheet does this with a floor row and a headroom row beneath the closing balance.

Float does it with a threshold. By default it is zero. Set it to the balance the business has decided not to go below, or to a negative figure where an overdraft is the real limit, and the side panel shows the date on which the forecast balance is due to cross it. This is a figure on the screen. Float does not send an alert, an email or a notification when the date moves, so someone has to look. How to size the floor is a separate question, covered in how much cash a business should keep in reserve.

When a decision is on the table: a what-if sits on top

ICAEW's fourth principle is to produce scenarios, including a worst case, a most likely case and a best case. In a spreadsheet that usually means a copy of the file, and from that point two forecasts to keep in step.

In Float a scenario is a layer. It holds budgets that are added to, changed in or removed from the base forecast, and it stacks on top of the base so the graph shows both lines at once. A layer can be duplicated as the starting point for another, compared with others by week in the scenario comparison report, and its budgets can be moved into the base once the decision is made. Each plan includes a set number of scenario layers, listed on our pricing page.

The limit matters in a 13-week view. A scenario layer changes budgets. It does not change when an invoice or a bill is expected to be paid, because those dates belong to the forecast itself. So a planned facility draw, a delayed hire or a pulled-forward purchase is a scenario; a customer paying three weeks late is an edit to the expected date, with a comment on the invoice recording why.

Friday: the 13-week view leaves as a file

Somebody asked for the forecast: the finance director, the board, a lender. From the weekly view, Float exports the full 13 weeks as a PDF, with the graph, the table, a summary of cash in, cash out and net movement, and your logo, or as a CSV. Each cell in the export shows the higher of the budget and the actual. A person starts every export. Nothing is scheduled and nothing is sent on a timer. Where the reader only needs to look, a view-only user role does the same job without a file.

One week, worked

The template page carries a full quarter for a fictional 32-person consultancy. Its week 4 is the one to borrow. The business has set a floor of £70,000. In week 4 a VAT payment of £68,300 leaves, receipts are thin, and the week closes at £35,080, which is £34,920 below the floor. The cause is a £58,000 council invoice due on 4 October that the customer's own history says will arrive on 21 October. The figures are that page's illustration, not data from a real business.

In the spreadsheet, a person found that week by reading the headroom row. Here is the same week with the six jobs done by the tool.

The council invoice arrived from the ledger with its due date. Its expected date is 21 October, set by hand on QuickBooks Online or applied from the customer's payment history on Xero. That one date is why the gap is visible three weeks ahead. With the threshold set at £70,000, the side panel shows the date in week 4 on which the balance is due to cross it.

The template page offers three ways out, and they are three different actions. Chasing the council to pay on its due date is a change to that invoice's expected date, which is an edit to the base forecast and worth a comment on the invoice. Moving the week 4 supplier run of £10,630 into week 5 is a batch change to the expected dates on those bills. Drawing £40,000 on a facility in week 4 and repaying it in week 5 is two budgets in a scenario layer, which leaves the base untouched and shows the week closing at £75,080 beside the original £35,080.

The tool placed the invoice, held the floor, layered the draw and will export whichever version the finance director chooses. Deciding that the council will be late, that £70,000 is the right floor, and that a draw is better than a difficult call to a supplier was done by a person.

What still needs a person

No professional body draws a line between what forecasting software should do and what a finance team should keep. The Association of Corporate Treasurers says only that forecasting "is also partly an art form, since the future is rarely a carbon copy of the past". The division below is ours, and it is specific to the 13-week view.

Reconciling. Only reconciled transactions reach the forecast. Float's own guidance is to reconcile daily, or at least weekly.

The expected dates that carry the low week. A tool can apply an average. It cannot know that a customer's payment run has moved or that an invoice is in dispute. The five or six receipts that decide the lowest week are worth a phone call.

Cash that the ledger has not seen. Contracted work not yet invoiced, a tax bill not yet assessed, a hire agreed for next month. These go in as budgets, and somebody retires them when the invoice or bill appears or the plan changes.

The floor. The tool shows the date the balance crosses it. Choosing the figure is a decision for the finance director.

The weekly snapshot. If you want to know how good the forecast was, export the 13-week view on the same day each week and keep it.

The choice. The forecast shows the size of the gap and the week. Which lever to pull is the reason the review happens. Our guide to running a weekly cash review in a finance team of three covers that meeting.

None of this is an argument that software removes spreadsheet risk by itself. A study of operational business spreadsheets by Powell, Baker and Lawson found errors in a small share of formula cells but in most of the workbooks it examined. A connection removes the cells where figures are pasted and re-keyed each week. The assumptions are still yours, and a wrong expected date is wrong in any tool.

How to check a 13-week tool before you buy

No professional or government body publishes selection criteria for cash flow forecasting software. General guidance covers whether a tool fits the systems you already use and whether you can get your data out. The five checks below are ours, one for each point where a 13-week view is made or broken. A fuller list for the stage before a shortlist is in what a finance team should require from cash flow forecasting software.

Step 1: Ask where the data comes from. Ledger, bank, or both, and exactly which records. If the answer is the ledger, ask whether unreconciled transactions are included and what the opening balance is. Float reads the ledger only, imports reconciled transactions, and opens on the reconciled balance.

Step 2: Ask for the refresh schedule. A time of day and a manual option is a checkable answer, and "live" is not. Float imports once a day at an hour you choose, with a sync on demand.

Step 3: Ask how a receipt gets its date. Due date, a date you set, or a date learned from payment history, and on which accounting platform. In Float it is the platform's expected date or the due date, changed by hand, with history-based dates for Xero users only.

Step 4: Ask what a what-if can change. Budgets only, or invoice and bill dates as well, and how many versions you can hold. In Float a scenario layer changes budgets, and the number of layers is set by plan.

Step 5: Ask what leaves the tool, and who starts it. File formats, whether the weekly export carries all 13 weeks, whether anything is sent automatically, and whether the tool writes anything back to your books. Float exports PDF and CSV on request, sends nothing by itself, and writes nothing to the ledger.

How Float fits

Float is a cash flow forecasting tool for finance teams that keep their books in Xero or QuickBooks Online. Sage Intacct support is coming soon; you can join the waitlist. It offers monthly and 13-week forecasting options, built by the direct method from the ledger's own invoices, bills and reconciled bank transactions.

In the weekly view it does the six jobs on this page: a daily one-way import at an hour you choose, an expected date on every invoice and bill, budgets that fill with actuals as they reconcile, a threshold with the date the balance is due to cross it, scenario layers on the base forecast, and a PDF or CSV of the full 13 weeks. For groups, companies on the dashboard consolidate into one weekly or monthly view in a display currency of your choice, on the plan that includes consolidation.

It also has limits, and they are worth stating beside the features. Float does not connect to the bank, so it has no intraday view. It does not send alerts or scheduled reports. Smart Expected Dates and automatic VAT and GST forecasting are for Xero users; QuickBooks Online users set expected dates in Float and carry the tax payment as a budget on its due date. And a spreadsheet forecast does not import as a file: budgets are entered directly or pasted in from a spreadsheet grid.

If you are still weighing the spreadsheet you have, what a spreadsheet cash forecast costs to maintain gives you a model for costing it.

Frequently asked questions

Is there 13-week forecast software for better cash management?

Yes. 13-week forecast software for cash management falls into three kinds: treasury systems that connect to banks, planning platforms that forecast the full accounts, and ledger-connected tools that build a weekly cash view from the invoices, bills and bank transactions in Xero or QuickBooks Online. Float is the third kind. It offers a 13-week weekly view alongside a monthly one, fed by a daily import from the accounting platform.

What does 13-week cash flow forecasting software do that a spreadsheet leaves to you?

It takes over six weekly jobs: bringing in balances, invoices and bills from the ledger, placing each receipt and payment in the week it is expected, filling closed weeks with what happened, showing the date the balance crosses a minimum, holding a what-if version on top of the forecast, and exporting the 13-week view. A spreadsheet can do all six, but a person has to do each one by hand every week. The judgement about payment timing and what action to take stays with the finance team in both cases.

Does 13-week forecasting software connect to my bank, and is the data real-time?

It depends on the kind of tool. Float connects to Xero or QuickBooks Online, not to the bank, and imports reconciled transactions once a day at an hour you choose, with a manual sync on demand, so the forecast is as current as your last reconciliation. Tools that connect to UK bank accounts as account information services may refresh without the user present no more than four times in 24 hours unless the bank agrees to more. Ask any tool for its schedule.

Can forecasting software predict when customers will pay?

For Xero users, Float's Smart Expected Dates applies each customer's average days late, from three to twelve months of your own payment history, to new invoices and bills as they import. The date is applied once and can be overridden. On QuickBooks Online the team sets expected dates in Float. On either platform an overdue invoice with no new expected date is assumed to be paid today and flagged in the app.

Can I test a what-if in a 13-week forecast without changing the forecast itself?

Yes, for anything entered as a budget. A scenario layer in Float holds added, changed or removed budgets on top of the base forecast and shows both lines on the graph, so a facility draw or a delayed purchase can be tested and then discarded or moved into the base. A change to when an invoice or bill will be paid is not a scenario; it is an edit to the expected date in the forecast itself.

Does 13-week forecasting software send an alert when cash is going to run short?

Float does not. It shows, on the side panel, the date the forecast balance is due to drop below a threshold you set, which is zero by default and can be negative for an overdraft. That date is displayed, not sent. Nothing is emailed or notified, and exports are started by a person, so the weekly look at the forecast remains a job for the team.

Who can see and change the forecast?

Float has four user roles. The owner is the user who connected the company; admins can edit the forecast and manage users; editors can edit the forecast but not users; viewers have read-only access. The connection to the accounting platform is one-way, so nothing done in Float changes the books, and two-factor authentication is mandatory for customers connecting Float to Xero.

How much does 13-week cash flow forecasting software cost?

Float's pricing depends on the size of the business and whether you need to consolidate more than one entity. Current plans, and what each includes, are on our pricing page. Every plan includes the 13-week view and starts with a 14-day free trial.

The quickest way to see what the six jobs look like on your own ledger is to connect it. Start a 14-day free trial of Float and Float will import your Xero or QuickBooks Online data in about three minutes, ready for you to set the expected dates, the floor and the first what-if.