Yes, for the parts of a client forecast that move data, and no, for the parts that need judgement. Connecting the forecast to the client's Xero or QuickBooks Online ledger removes the collecting and re-keying, and on Xero it reduces the work of dating when cash will arrive. It does not remove the conversation with the client, the assumptions, or the review before the forecast goes out. No published study measures how much time that saves a practice, so this page sets out where the hours go and gives a method for measuring the difference on one client over one month.
What "automated" means for a client forecast
The word covers less than it sounds. In a connected cash flow forecast, the import is automatic: balances, invoices, bills and transactions arrive from the client's ledger on a schedule, without anyone exporting or pasting them. The forecast itself is not automatic. Someone still decides when each customer will pay, which commitments the ledger does not hold yet, and whether the result makes sense for this client. Exports are not automatic either: a person runs them and sends them.
So the useful question for a practice is not whether forecasting can be automated. It is which stages of the work a ledger connection takes over, and how many minutes those stages were costing.
Before the forecast: the client's books
Every client forecast rests on the client's ledger, and a connected forecast rests on it more directly. A ledger-fed forecast reads reconciled transactions and the reconciled bank balance. Items that have not been reconciled are not in it. ICAEW's nine principles for cash flow forecasting make the same link from the other side: they ask finance professionals to reconcile cash flow reports to bank and creditor statements and to increase the frequency and rigour of balance sheet reconciliations in order to validate forecasts. No professional body sets a reconciliation frequency for small businesses.
In a practice that keeps the client's books, that reconciliation is often the practice's own work. It belongs in the measurement as a separate line. If it is folded into the forecast time, a connected forecast can look slower simply because it exposed books that were behind.
Where the hours go in a manual client forecast
A forecast built by hand in a spreadsheet goes through four stages each cycle. No professional body publishes a step list for a forecast a practice prepares for a client, so the grouping below is our own, drawn from the guidance that does exist on preparing a forecast.
Stage 1: Collect the client's ledger data. Export the bank transactions, the open invoices and the unpaid bills from the client's accounting platform, for each bank account and each entity.
Stage 2: Re-key and roll the forecast forward. Paste the exports into the workbook, repair the columns and formats that broke, key in the new opening balances, replace last period's forecast lines with what happened, and extend recurring items such as rent, payroll and subscriptions to the end of the horizon.
Stage 3: Date the cash. Decide when each receipt and payment will actually land, which is rarely the due date. This is where the practice asks the client what the ledger does not hold: a customer that has promised to pay on Friday, a contract signed but not yet invoiced, a hire starting next month, a loan repayment. ICAEW's principles ask for expected, worst-case and best-case dates on receipts and payments, which is a judgement about each one, not a calculation.
Stage 4: Review, explain and deliver. Read the forecast against what you know of the business, question anything that looks wrong, decide what the client needs to hear, and present it.
Who does each stage varies from practice to practice. A common arrangement is for a bookkeeper or junior to handle the first two and for a manager or partner to own the client conversation and the review, but no professional body prescribes that split for forecasting.
What a ledger connection removes, stage by stage
The verdicts below describe Float connected to Xero or QuickBooks Online. Other connected tools differ in the detail, so check each stage against the tool you are considering.
| Stage | With the ledger connected | Verdict |
|---|---|---|
| 1. Collect the client's ledger data | Balances, invoices, bills and reconciled transactions import once a day at an hour you choose, with a manual sync when the books have just been updated. Several client companies sit on one dashboard | Removed |
| 2. Re-key and roll forward | Nothing is pasted. Actuals fill the forecast as they reconcile, and repeating budgets roll forward on their own. On Xero, repeating invoices import too. On QuickBooks Online, recurring invoices and bills do not sync, so they are set once as repeating budgets | Removed for anything the ledger holds |
| 3. Date the cash | On Xero, Smart Expected Dates applies each customer's average days late, from the client's own payment history, to new invoices as they import. On QuickBooks Online, expected dates set in the platform import and the rest are set in Float by hand. On both, commitments the ledger has not seen are entered by a person | Reduced on Xero; largely manual on QuickBooks Online |
| 4. Review, explain and deliver | The review is the same job. The forecast is ready to present on screen or export, but someone still reads it, decides what it means and talks the client through it | Not removed |
Two details catch practices out in the first month. An overdue invoice with no expected date is assumed to be paid today, and is flagged, so the overdue list needs working through at the start of each cycle. And the connection runs one way: nothing set in the forecast changes the client's books, and a date changed in the client's ledger overwrites the one set in the forecast at the next sync.
What stays with the accountant
The stages a connection leaves alone are the ones clients pay an accountant for. Deciding when a slow customer will really pay, finding out which commitments the books do not show, choosing which scenarios are worth running, and reading the result for the client all stay with a person.
Where a forecast is part of a formal engagement, the professional standards say the same. ISAE 3400, the international standard for examining prospective financial information, places responsibility for a forecast and its assumptions with management. In Australia, APES 315 covers compilations of prospective financial information and has the engagement document record the basis of forecasting and the key assumptions the client provides, with any assumption the accountant makes brought to the client's attention. Both apply to formal engagements, not to a routine monthly forecast for an owner-managed client, but they describe where the judgement sits, and software does not move it.
What the published evidence says about time saved
There is no independent measure of the time a practice spends producing a client cash flow forecast, in total or by stage, in the UK, the US, Australia or New Zealand. Figures that circulate for automated forecasting come from software companies or from studies of large corporate finance teams, and none of them measures an accounting practice.
The closest evidence we can offer is one practitioner's account. Barry Hynd, a fractional CFO who uses Float with his clients, describes the spreadsheet years this way: "You weren't spending 15-20 minutes like you would now. It was 2, 3, 4 hours." And now: "Now it takes me 15, 20 minutes to look at a scenario and pull it out of Float." That is one person's experience, and the two sentences describe different tasks, so it is not a like-for-like measurement. It is the reason the method below exists: the number that matters is the one from your own client.
How to measure the difference on one client over one month
The method below measures working time on a single client's forecast before and after connecting it to the ledger. It takes one cycle of attention before the change and one after. The principle of taking a baseline before a change and the same measurement afterwards is standard in benefits measurement; the rest is our method, and you can adapt it.
Step 1: Choose one client and fix the scope. Pick a client whose forecast you produce every month, and write down what the deliverable is: the horizon, the weekly or monthly view, the scenarios, the meeting. Define the start as opening the client's files and the end as the forecast reaching the client.
Step 2: Set five time codes. Use the client's books, then the four stages: collect, re-key and roll forward, date the cash, and review and deliver. If your practice records time against task codes, add these to the client's job for the month. If it does not keep timesheets, keep a simple log beside the work: the code, the start time and the end time.
Step 3: Time one full cycle before the change. Record working minutes, not billed time and not elapsed calendar days. Include the chasing: the emails and calls to the client about payment dates and commitments belong to stage 3.
Step 4: Record set-up separately. Connecting the ledger, checking the first forecast against the old spreadsheet, arranging the layout and showing the team how to read it are one-off costs. Log them under their own code so they do not land in the monthly figure.
Step 5: Time the same scope after the change. Wait until the first full cycle is behind you, then time the next one against the same five codes, with the same deliverable and, where possible, the same people. Note anything unusual about either month, such as a new loan, a year end or a lost customer.
Step 6: Compare stage by stage and read the result for that client. Expect the collect and re-key codes to fall, the dating code to fall on Xero and move less on QuickBooks Online, and the review code to stay roughly where it was. If the books code rises, the connection has exposed reconciliation that was being skipped. The result is an estimate for that client; repeat it on a second client before treating it as the practice's figure.
People tend to work differently when they know they are being timed, so the first month's figures on both sides may run faster than usual. Timing a second cycle each side, where you can, evens that out.
When a spreadsheet is still the right answer for a client
A spreadsheet is a recognised way to produce a cash flow forecast. ICAEW's Business Finance Guide says a forecast can be created in a spreadsheet, an app or an online accounting system, and its spreadsheet principles ask users to decide whether a spreadsheet is a suitable tool for the job rather than assuming it is. No professional body sets a size or volume above which a client should move to a dedicated tool.
For a client with one entity, a handful of customers, steady receipts and a forecast you refresh a few times a year, the collect and re-key stages may take minutes in a spreadsheet, and there is little to measure. The case for a connection grows with the things that multiply those stages: more bank accounts, more entities, more invoices, a weekly forecast instead of a quarterly one, and clients who need a forward view every month. The measurement above will tell you which side of that line a client sits on.
How Float fits
Float is a cash flow forecasting tool that connects to Xero or QuickBooks Online, with Sage Intacct on the waitlist. It imports each client's balances, invoices, bills and reconciled transactions once a day at an hour you choose, with a manual sync on demand, and it reads the ledger without writing to it.
For a practice, multiple client companies sit on one dashboard, and companies can be combined into one consolidated view in a currency you choose. Each forecast has a weekly 13-week view and a monthly view. Comments and notes sit on invoices, bills and budgets, so the reason a payment date was moved stays with the item. A Viewer role gives a client read-only access to their own forecast, and presentation mode strips the screen back for walking a client through it.
The automation stops where this page says it does. Float sets expected dates from payment history for Xero clients only; on QuickBooks Online the team sets them. Float shows the date a balance will cross a threshold you set, but it sends no alerts or scheduled reports: exports to PDF and CSV are run by a person. Budgets can be pasted in from a spreadsheet, but a spreadsheet file cannot be imported. Scenarios are layers someone adds, not something the tool builds.
How Float is priced for practices is set out on the advisors page. For the decision of which tool to standardise on across a client base, see how an accounting practice should choose cash flow forecasting software; for what the spreadsheet costs an in-house finance team in pounds, see what a spreadsheet cash forecast costs to maintain.
Frequently asked questions
How much time does a ledger connection save on a client cash flow forecast?
No independent study measures it for accounting practices, so any single figure would be a guess. The saving comes from the stages that move data: collecting the client's ledger data and re-keying it into the forecast. Dating the cash is reduced on Xero and stays largely manual on QuickBooks Online, and the review is unchanged. Time one client's forecast before and after to get your own number.
Which parts of a client cash flow forecast can be automated?
The import of balances, open invoices, unpaid bills and reconciled transactions from the client's ledger can be automated, along with rolling the forecast forward as actuals arrive. On Xero, expected payment dates can be applied from each customer's payment history. Deciding when a slow customer will pay, adding commitments the books do not hold, choosing scenarios and reviewing the result cannot.
What still needs an accountant once the forecast is connected to the ledger?
The judgement does. Someone has to ask the client about contracts, hires, loans and promised payments the ledger does not show, decide which payment dates to trust, choose the scenarios worth running, and read the forecast before it goes out. Where a forecast is part of a formal engagement, standards such as ISAE 3400 place the assumptions with management and the review with the practitioner.
How can a practice measure forecast time if it does not keep timesheets?
Keep a simple time log for one client over one forecast cycle: a code for each stage, a start time and an end time. Use five codes: the client's books, collecting data, re-keying and rolling forward, dating the cash, and review and delivery. Repeat it for the same client after the change, and log one-off set-up time under its own code.
Does a connected forecast work if the client's books are not reconciled?
Only partly. A ledger-fed forecast reads reconciled transactions and the reconciled bank balance, so anything not yet reconciled is missing from it. ICAEW's cash flow principles link regular reconciliation to validating a forecast, and Float recommends reconciling every day or at least every week. In a practice that keeps the books, that reconciliation time is worth recording separately.
What is different for Xero and QuickBooks Online clients?
On Xero, Float imports repeating invoices and applies Smart Expected Dates, which set each new invoice's expected date from the customer's payment history, and VAT or GST can be forecast automatically in the UK, Australia and New Zealand. On QuickBooks Online, recurring invoices and bills do not sync and are set once as repeating budgets, expected dates are set by hand where the platform has none, and the tax payment is entered as a budget.
When is a spreadsheet still the right tool for a client forecast?
When the client is simple enough that collecting and re-keying take minutes: one entity, few customers, steady receipts and an occasional forecast. ICAEW's guidance accepts a spreadsheet as a way to produce a forecast and asks users to judge whether it suits the job. No professional body sets a threshold for moving a client to a dedicated tool.
Does Float change anything in the client's Xero or QuickBooks Online file?
No. The connection runs one way: Float reads the client's ledger and writes nothing back. Expected dates, budgets and scenarios set in Float stay in Float, and if a due date or expected date is changed in the client's ledger, that change overwrites the date in Float at the next sync.
To measure the difference on a real client, start a free trial, connect one client's Xero or QuickBooks Online ledger, and time the next forecast cycle against the last one.
