Profitable businesses run out of cash all the time. The reason is simple and unfair: profit is an accounting opinion, while cash is a fact. A sale you booked in September pays your payroll only when the customer's money lands in your account, which may be in November. In between, salaries, rent, GST and suppliers all arrive on their own fixed dates, whether or not your customers are in a hurry.
Most owner-led firms respond to this with a monthly P&L and a quick glance at the bank balance. That tells you where you have been. It does not tell you that in week 8 your balance will fall below what you need to run payroll, which is the one thing you needed to know about six weeks earlier.
A 13-week cash-flow forecast fixes this. It is a week-by-week projection of every rupee expected in and out over the next quarter, rolled forward every week. It is the most useful financial model a small business can build, and it needs nothing more than a spreadsheet and thirty minutes. This guide walks through the full build, with a worked example, India-specific cash dates, scenario testing and the weekly rhythm that makes it work.
Why use a 13-week forecast, and why weekly?
Thirteen weeks is one quarter. It is long enough to see a cash squeeze coming and act on it, and short enough that your estimates are still grounded in invoices, contracts and known obligations rather than hope. Weekly granularity matters because cash problems live inside the month: a business can be comfortable on the 1st and in trouble by the 12th.
| Model | Time horizon | Granularity | Best question it answers | Weakness |
|---|---|---|---|---|
| Annual budget | 12 months | Monthly | Are we on track for the year's targets? | Stale by February; hides intra-month dips |
| Rolling 12-month forecast | 12 months | Monthly | What does the next year look like? | Too coarse to see a payroll-week shortfall |
| 13-week cash forecast | 13 weeks | Weekly | Will we have the cash to meet each obligation, when? | Short horizon; needs a weekly refresh |
| Daily cash position | 1 to 2 weeks | Daily | Can we pay today's bills? | Reactive; no time to fix anything |
The 13-week model sits in the useful middle. It is the only one of the four that gives you enough warning to chase a receivable, negotiate a supplier date or arrange a short credit line before you need it. The annual budget and the 13-week forecast do not compete: the budget sets direction, the forecast keeps you alive while you follow it.
Direct method or indirect method?
There are two ways to build a cash forecast. The indirect method starts from projected profit and adjusts for working-capital movements, which is how accountants build a cash-flow statement. The direct method simply lists actual cash receipts and payments. For a 13-week operational forecast, use the direct method.
| Factor | Direct method | Indirect method |
|---|---|---|
| Starting point | Bank balance and actual receipts and payments | Net profit, adjusted for non-cash items |
| Accuracy at weekly level | High: mirrors the bank statement | Low: hides timing |
| Effort to maintain | Moderate: needs AR and AP lists | Low once the P&L is closed |
| Reconciles to bank? | Yes, line by line | Only at period end |
| Best use | Operating decisions this quarter | Statutory and board reporting |
The direct method wins here because it forces the useful discipline: for every rupee, you ask which week it will actually move.
Step 1: Set up the skeleton
Open a spreadsheet and lay out thirteen columns, one per week, starting with the week beginning this Monday. Down the left, create four blocks: opening cash, cash in, cash out and closing cash. Closing cash for each week becomes the opening cash for the next. That chain is the whole engine of the model.
Under cash in, list your receipt streams: collections from existing invoices, expected collections from work in progress, new orders at realistic conversion, and any other inflows such as grants or refunds. Under cash out, group your payments into five buckets: people, suppliers, rent and fixed overheads, statutory dues, and financing. Resist the urge to add detail beyond that. A forecast with sixty lines is a forecast nobody updates.
Finally, add two cells above the grid: your minimum cash buffer and a trigger line. The buffer is the lowest balance you are willing to run, and a sensible starting point is two to four weeks of fixed costs. The trigger line is the balance at which you act. If the forecast crosses it in any week, a decision is due now, not when the week arrives.
Step 2: Forecast cash in from receivables, not from sales
The commonest forecasting error is to treat revenue as cash. Revenue is when you billed. Cash is when they paid. The gap between the two is your collection delay, and for most Indian MSMEs it is far longer than the contractual terms suggest.
Start with your accounts-receivable list. For each open invoice, put the amount in the week you actually expect it to be paid, based on how that specific customer behaves, not what the invoice says. A customer on 30-day terms who habitually pays at 55 days goes in at 55.
For revenue you have not yet invoiced, apply a collection curve: the share of an invoice typically collected after 0, 30, 60 and 90 days. Build it from your own history.
| Customer type | Stated terms | Typical actual pay time | Forecast treatment |
|---|---|---|---|
| Large corporate or PSU | 30 to 45 days | 50 to 90 days, with PO and approval delays | Add 3 to 5 weeks; assume month-end batch runs |
| Mid-market B2B | 15 to 30 days | 35 to 60 days | Add 2 to 3 weeks; confirm after first reminder |
| Small business customer | Advance or 7 days | Mostly on time; occasional short slips | Use terms plus 1 week |
| Retail and online | Immediate | Settlement lag of 1 to 3 days from gateways | Use settlement date, net of fees and returns |
| Export or Gulf (GCC) customer | 30 to 60 days | Variable; FX and bank processing add days | Add 1 to 2 weeks; model in rupees at a prudent rate |
These ranges are directional. Replace them with your own ageing data as soon as you have two or three quarters of history. The principle is what matters: forecast the behaviour of the payer, not the promise on the invoice. Where an order is not yet signed, include it only at a probability you would be comfortable defending, and keep it out of the base case if it is below about 50 percent.
Step 3: Forecast cash out with the obligations calendar
Cash out is easier to forecast than cash in because most of it is predictable. The risk is forgetting the lumpy items that arrive quarterly or annually. Build an obligations calendar once, and it will feed the forecast every week.
| Obligation | Usual timing in India | Forecasting note |
|---|---|---|
| Salaries | Month-end or first working days | The largest fixed outflow; place it in the exact week |
| PF and ESI | By the 15th of the following month | Follows payroll; do not forget employer share |
| TDS deposit | By the 7th of the following month | Includes TDS you deducted from vendors and staff |
| GST (GSTR-3B) | 20th of the following month for monthly filers; 22nd or 24th after the quarter for quarterly (QRMP) filers (GSTN) | Net of input credit; can be large in a strong sales month |
| Advance tax | 15 June, 15 Sept, 15 Dec, 15 March (section 408) | Easy to miss; a December and March dent is predictable |
| Rent and utilities | Monthly, often at the start | Check lease escalation dates |
| Loan EMIs and interest | Fixed monthly date | Include working-capital interest on overdrafts |
| Supplier payments | As per payment run | Schedule by actual due date, see the 45-day rule below |
Check due dates against current notifications and your tax adviser's calendar, since dates can be extended or changed. The point is to have each one sitting in the right week before it surprises you.
The 45-day rule changes how you pay micro and small suppliers
Since the 2023-24 financial year, Indian income tax law has tied your tax deduction for purchases from registered micro and small enterprises to payment timing. The rule was introduced as Section 43B(h) of the Income-tax Act, 1961, and continues as section 37(2)(g) of the Income-tax Act, 2025, which came into force on 1 April 2026. Broadly, if you do not pay such a supplier within the time limit in section 15 of the MSMED Act, 2006, which is the agreed period capped at 45 days (and 15 days where there is no written agreement), the expense is deductible only in the year you actually pay. Late payment therefore carries a tax cost on top of any relationship cost.
For forecasting, this means your payables schedule should flag every micro and small supplier with a payment-by-date, and treat that date as firm. Confirm the exact rules and your exposure with your chartered accountant, but build the dates into the model either way.
A worked example: a Rs 4 crore services firm
To make this concrete, here is a 13-week forecast for a services firm with about Rs 4 crore of annual revenue. All figures are in Rs lakh. It starts with Rs 18 lakh in the bank and a policy of never dropping below Rs 10 lakh. Cash in follows customer behaviour, cash out follows the obligations calendar: three payrolls of Rs 16 lakh, rent at the start of each month, three EMIs of Rs 1.5 lakh, three statutory payments of Rs 4 lakh, and weekly supplier payments.
| Week | Opening cash | Cash in | Cash out | Net | Closing cash |
|---|---|---|---|---|---|
| W1 | 18.0 | 9.0 | 6.0 | +3.0 | 21.0 |
| W2 | 21.0 | 7.0 | 4.5 | +2.5 | 23.5 |
| W3 | 23.5 | 12.0 | 9.0 | +3.0 | 26.5 |
| W4 | 26.5 | 6.0 | 19.0 | -13.0 | 13.5 |
| W5 | 13.5 | 8.0 | 6.0 | +2.0 | 15.5 |
| W6 | 15.5 | 14.0 | 6.5 | +7.5 | 23.0 |
| W7 | 23.0 | 5.0 | 7.0 | -2.0 | 21.0 |
| W8 | 21.0 | 6.0 | 20.0 | -14.0 | 7.0 (below buffer) |
| W9 | 7.0 | 11.0 | 7.0 | +4.0 | 11.0 |
| W10 | 11.0 | 7.0 | 4.5 | +2.5 | 13.5 |
| W11 | 13.5 | 13.0 | 9.0 | +4.0 | 17.5 |
| W12 | 17.5 | 6.0 | 3.0 | +3.0 | 20.5 |
| W13 | 20.5 | 10.0 | 22.0 | -12.0 | 8.5 (below buffer) |
Over the quarter, the firm collects Rs 114 lakh and pays out Rs 123.5 lakh, so cash falls by Rs 9.5 lakh to Rs 8.5 lakh. On a quarterly P&L this firm could look healthy. On the weekly view, two problems appear: the balance falls to Rs 7 lakh in week 8, when a customer payment slips and payroll lands in the same week, and it ends the quarter at Rs 8.5 lakh, below the buffer, after the third payroll in week 13.
This is exactly the information a monthly view hides. Armed with it in week 1, the owner has seven weeks to act: chase the large receivable due in week 8, ask two suppliers for a one-week extension, or arrange a small working-capital line before the squeeze, when banks are far more willing to lend.
Step 4: Stress-test with three scenarios
A single forecast gives a false sense of precision. Run three versions and compare them, changing only a few drivers.
| Scenario | Receipts assumption | Payments assumption | What you learn |
|---|---|---|---|
| Base | Customers pay at their historical pace | Obligations calendar as planned | Your expected path and lowest week |
| Downside | Top 3 customers pay 2 to 3 weeks late; new orders slip a month | Fixed costs unchanged | How much buffer or credit you really need |
| Upside | Pipeline converts; one large order pays an advance | Extra delivery cost and working capital | Whether growth itself will strain cash |
The downside case is the important one. If your lowest week in the downside is still above zero, you can run a calm business. If it dips below zero, you now know the size of the facility, deferral or cost action you should line up. Notice that the upside case matters too: fast growth consumes cash up front, since you pay people and suppliers before customers pay you, and many a successful quarter has ended in a cash crisis for exactly that reason.
Step 5: Turn the forecast into decisions
A forecast that does not change behaviour is just a nicer spreadsheet. When a week crosses the trigger line, you have a short menu of levers, ordered from cheapest to most expensive.
| Lever | Typical effect | Cost or risk | Lead time |
|---|---|---|---|
| Chase receivables, offer a small early-pay discount | Pulls cash forward 1 to 3 weeks | Discount cost; relationship tone | Days |
| Invoice sooner and in milestones | Starts the collection clock earlier | Needs contract discipline | Days to weeks |
| Re-sequence supplier payments | Shifts outflow by 1 to 2 weeks | Mind the 45-day rule and supplier goodwill | Days |
| Defer discretionary spend | Frees a predictable amount | Can delay growth work | Immediate |
| Bank working-capital line | Bridges a timing gap | Interest cost; needs approval time | Weeks |
| Invoice discounting or factoring | Converts receivables to cash now | Fee on the invoice value | Days to weeks |
Write the decision rule down in advance, for example: "If any week in the next eight falls below the trigger line, we chase the top five overdue invoices and review supplier timing within 48 hours." Deciding when you are calm beats deciding when you are short of cash.
How do you keep the forecast accurate every week?
The model is built once and kept alive weekly. Choose a fixed slot, typically Monday morning, and run the same five steps each time.
First, replace last week's forecast with the actual bank figures. Second, compare forecast to actual line by line and note the biggest misses. Third, learn from them: if three customers paid later than predicted, adjust their assumed pay times. Fourth, roll the window forward by dropping the week just closed and adding a new week 13. Fifth, decide: does any week now cross the trigger line, and if so what do you do about it?
Over a few months, the variance review improves your accuracy more than any formula. You discover that one customer is always late, that a certain supplier invoices early, that your sales team's "90 percent likely" means about 50. That learning is the real predictive power in predictive financial modeling: not a clever algorithm, but a model that gets calibrated to your business every seven days.
Which tool should you build the forecast in?
Start in a spreadsheet. The logic is simple, you will want to see and edit every assumption, and the structure matters more than the software. Move to accounting-software integrations only when manual updates become the bottleneck.
| Option | Best for | Strength | Limitation |
|---|---|---|---|
| Spreadsheet (Excel or Google Sheets) | First 6 to 12 months of forecasting | Full control; no cost; easy to audit | Manual data entry; version risk |
| Accounting software reports (Tally, Zoho Books and similar) | Pulling receivable and payable ageing | Source data already in one place | Built-in forecasting is usually basic |
| Spreadsheet plus accounting export | Most MSMEs after the first quarter | Automatic actuals, flexible assumptions | Needs a clean chart of accounts |
| Dedicated cash-forecasting platforms | Multi-entity or higher-volume businesses | Bank feeds, scenario tooling, dashboards | Subscription cost; set-up effort |
Check current pricing and bank-feed support for any platform before committing, since offerings change often. Whichever tool you pick, the discipline of the weekly loop is what produces the value.
Common mistakes to avoid
| Mistake | Why it hurts | Fix |
|---|---|---|
| Forecasting revenue, not cash | Ignores the collection delay that causes the squeeze | Forecast by expected payment week, per customer |
| Setting it and forgetting it | Goes stale within a fortnight | Fixed weekly slot, with an owner |
| Too much detail | Nobody updates a 60-line model | Keep to the five outflow buckets and a handful of inflow lines |
| Optimistic receipts | Hides the real low point | Use observed pay times; keep unsigned orders out of base case |
| Forgetting quarterly and annual dues | Advance tax and insurance arrive as shocks | Maintain the obligations calendar |
| No decision rule | Warnings are noticed but not acted on | Write the trigger and the response in advance |
How this connects to the rest of your operating system
A reliable cash forecast depends on reliable operations. Collections improve when invoicing is a documented process rather than an afterthought, which is the territory of business process improvement and the SOP blueprint. Cash is also tied to productivity: a business with higher revenue per employee tends to convert activity to cash more efficiently, and a team that runs on the Execution Grid makes fewer of the invoicing and follow-up misses that stretch collection time.
The same weekly-rhythm logic applies to growth. Predictable cash is what lets you invest in a steady marketing engine rather than stop and start. We cover that in Building B2B Content Authority, and the full system is the subject of our forthcoming pillar guide, The SEO and Content Flywheel, which shows how to build an organic growth engine that compounds without a matching jump in spend. This forecast also seeds our upcoming finance series on funding your growth, where a clear view of your cash gap is the starting point for choosing the right capital.
Your first 13-week forecast this week
You can have a working version by the end of the week.
| Day | Task | Output |
|---|---|---|
| Day 1 | Pull bank balance, receivable ageing and payable ageing | Opening position and the raw lists |
| Day 2 | Build the 13-column skeleton and obligations calendar | Cash-out grid with statutory dates placed |
| Day 3 | Place each receivable by expected pay week | Cash-in grid from real behaviour |
| Day 4 | Set buffer and trigger line; run the downside case | Lowest week and worst-case gap |
| Day 5 | Agree the decision rule and the Monday slot | A live, owned forecast |
Key takeaways
- Profit is not cash. A 13-week, weekly, direct-method forecast shows the squeezes that monthly accounts hide.
- Forecast receipts by how each customer actually pays, not by invoice terms, and keep unsigned orders out of the base case.
- Build an obligations calendar for payroll, PF, TDS, GST, advance tax, EMIs and rent, and treat the 45-day rule for micro and small suppliers as a firm date.
- Always run a downside scenario. The lowest week in that case tells you how much buffer or credit you need.
- Set a minimum buffer and a trigger line, and write down the response in advance so the forecast produces decisions.
- Update it weekly in a fixed slot. The variance review is what makes the forecast predictive.
This guide is part of the Stratisian Vault. Want a second pair of eyes on your cash model, or a forecast built into your operating rhythm? Book a strategy call and we will set it up with you.