Cash flow forecasting

A forecast that starts from how your money actually moves

Morgonplex builds your forecast from twelve to twenty-four months of real transactions, then keeps it current every day. No growth-rate guess typed into a cell.

Actuals to forecast
Forecast from today
₹0₹24L₹48L₹72LMaySepJanMaySepJanTodayCash-out May 2027
How forecasts are built

Three layers of pattern, one projection

Every forecast is the sum of separate models for money coming in and money going out. Each is visible, and each can be overridden.

Recurring commitments

Payroll, rent, subscriptions, loan EMIs and statutory payments are detected from history and scheduled on the dates they actually clear.

Collection behaviour

Invoices are projected on when each customer tends to pay, not when the invoice is due. A client who pays in 52 days is modelled that way.

Variable and seasonal spend

Marketing, cloud, travel and inventory are forecast with seasonality, so festive-quarter peaks and year-end lulls show up ahead of time.

Worked example

One forecast, end to end

An illustrative B2B software company with ₹50L in the bank. Historical data goes in on the left; the projection comes out on the right.

In: last 6 months of history

Customer receipts (avg)₹18.4L / mo
Payroll, 14 people₹15.2L / mo
Cloud and tools₹2.9L / mo
Marketing₹3.4L / mo
Rent and office₹1.6L / mo
Other operating₹0.9L / mo
Avg. days to collect47 days

Out: 12-month projection

Net burn, next 3 months₹5.6L / mo
Burn drift+1.2% / mo
Lowest monthMarch (annual renewals)
Projected runway8.4 months
Confidence band± 0.9 months
Largest driverPayroll, 64% of outflow
RefreshDaily, on new transactions
Always current

Forecasts that update themselves

When a customer pays early, a vendor raises prices or a new subscription appears, the forecast absorbs it on the next sync. You see what changed and by how much.

  • Daily re-forecast on new bank and ledger data
  • Change log explaining every movement
  • Manual overrides that persist across refreshes
  • Weekly, monthly and quarterly views
What changed since yesterday3 changes
Northwind paid 9 days early+₹4.2L
New tool detected: Figma-₹18K / mo
AWS bill above trend-₹62K
Net effect on runway+0.1 months
Questions

Forecasting FAQ

How much history do I need?

Six months gives a usable forecast. Twelve to twenty-four months lets the engine pick up seasonality and annual payments properly.

How accurate are the forecasts?

Accuracy depends on how regular your business is. Every forecast month shows a confidence band so you can see where the engine is sure and where it is not. Forecasts are decision support, not a guarantee.

Can I override a projection?

Yes. You can pin any line item, add known future events like a new contract, and your overrides persist through daily refreshes.

Get your first forecast today

Connect your ledger and bank. The engine reads your history and projects twelve months forward in minutes.