Problem guide
The cash forecast is a guess
The cash forecast becomes a guess when the numbers behind it, work done but not billed, invoices due, bills coming and payroll, live in different places and arrive late. Pull those few inputs together on a fixed schedule, and compare each forecast against what actually happened so it gets better over time.
What it usually looks like
The owner wants to buy a new truck and asks whether cash will be tight next month. The controller opens a spreadsheet, pastes in receivables and payables from accounting, and adds payroll from memory. Then she calls three project managers to ask what they expect to bill. One is not sure because two tickets are still in trucks. Another gives a number that includes work the customer has not approved yet. The forecast says cash is fine. Three weeks later, the company draws on its credit line to make payroll.
Signs you have this problem
- The forecast is rebuilt from scratch each time someone asks for it
- Cash surprises come from work that was finished but not yet billed
- The controller calls project managers to ask when invoices will go out
- Nobody checks last month's forecast against what actually came in and went out
- Big purchases are approved without anyone looking at cash for the coming weeks
What usually causes it
- Work finished in the field is not billed yet, and the office cannot see how much is waitingThe invisible queue
- Receivables, payables, payroll and open jobs sit in separate systems and are retyped into one spreadsheetThe re-entry trap
- Expected billing dates come from project managers by phone, late and from memoryThe field-to-office split
What to try this week
- List the inputs you actually useWrite down every number that goes into the forecast and where it comes from. Mark which ones are typed in by hand and which are estimates.
- Compare the last few forecasts to realityTake recent forecasts and put actual cash in and out next to them, week by week. Note which lines were furthest off.
- Count finished but unbilled workOnce a week, list jobs where work is done and no invoice has gone out, with an expected billing date for each.
- Update on a fixed dayRefresh the forecast the same day each week, even when nobody asked for it. A regular forecast gets checked. An occasional one gets trusted blindly.
Where do-it-yourself stops
These steps make the forecast regular and show you where it misses. They do not remove the retyping of receivables, payables, payroll and job status into one spreadsheet, or the late picture of unbilled work in the field. A lasting fix is a tool built around your own accounting and job records, so the forecast reads from the numbers you already keep. That is what a Workflow Spec designs.
Questions people ask
Why is our cash forecast always wrong?
Usually because its inputs arrive late or come from memory, especially the amount of work done but not yet billed.
How far ahead should we forecast?
Far enough to cover your next payroll cycles and large known payments. Start short and accurate, then extend as the numbers prove out.
Who should own the forecast?
Usually the controller or office lead, with project managers responsible for their expected billing dates.
Can our accounting system do the forecast?
It knows invoices and bills already entered. It usually does not know about finished work that has not been billed, which is often the biggest gap.
How do we know if the forecast is getting better?
Compare each forecast with what actually happened. If the gaps shrink over time, it is working.