# The 13-Week Cash Flow Forecast: The Practical Model That Keeps You Out of a Liquidity Crisis *The standard every CFO in the world runs on — simplified for a Saudi business owner with no finance background* > **In short:** Build a 13-week cash flow forecast the way CFOs do — why 13 weeks, the exact template, and a worked Saudi example that prevents a liquidity crunch. - **URL:** https://www.snad.io/en/blog/tanabu-tadafuq-naqdi-13-usbu - **Arabic original:** https://www.snad.io/blog/tanabu-tadafuq-naqdi-13-usbu - **Category:** Guides — Core Accounting - **Tags:** cash flow, financial planning, liquidity management, accounting, financial reporting, ERP - **Published:** 2026-05-18 - **Updated:** 2026-08-02 - **Publisher:** Snad (snad.io) A Saudi business owner opens his bank account on Thursday morning and sees a balance of SAR 28,000. Payroll is due in six days and comes to SAR 47,000. He has no idea exactly when the invoices sitting with his customers will actually land. He swings between panic and denial. This scenario is not the exception — it is the rule for a great many Saudi companies that lose control of their cash while being profitable on paper. The standard model for managing that risk is called the 13-Week Cash Flow Forecast. Every finance chief at a large company uses it, and it works perfectly well in a business with three employees. This guide builds the model with you step by step, using Saudi examples and realistic numbers. ## Why 13 weeks specifically, and not a month or a year Thirteen weeks is not an arbitrary number. It is a global finance standard that CFOs at large companies adopted for very specific reasons: **1. Three months = one full financial quarter** 13 weeks is 91 days ≈ 3 months. That is the window in which a business owner can realistically secure bank financing, defer a supplier payment, collect overdue receivables, or restructure costs. Anything shorter and there is no time to act; anything longer and the forecast fog doubles. **2. A weekly update catches changes fast** A monthly budget spots a cash problem 30 days after it starts. A weekly model spots it after 7 days. In a crisis, that gap can be the difference between a company that survives and one that does not. **3. Wide enough to reveal seasonality** Many Saudi companies have a cycle *within* the month — a peak at the start of the month when salaries land, and a slump at month-end. A 13-week model captures 3 monthly cycles and exposes the pattern. **4. Long enough to activate financing options** A Saudi bank needs 4–8 weeks to approve a new credit facility. If you see the problem 13 weeks out, you have room to move. If you see it 4 weeks out, it is already too late. The methodology is known internationally as the **13-Week Cash Flow Forecast**, or **TWCF**, and turnaround managers use it almost universally when restructuring companies. ## Cash flow vs. profit vs. revenue Three terms that business owners routinely confuse — and confusing them is a leading cause of insolvency despite a healthy 'profit on paper': **Revenue**: the value of sales invoices issued in the period. Whether or not the money was collected is irrelevant. **Profit**: revenue minus recognised expenses (the accrual principle). It appears on the income statement and does not reflect actual cash. **Cash flow**: what actually entered and left your bank account and your cash drawer during the period, regardless of invoices. **An embarrassing but extremely common example**: a wholesale grocery business issues SAR 450,000 of invoices in April (strong revenue). Its direct costs plus expenses come to SAR 380,000 — a paper profit of SAR 70,000. But its customers pay on 45-day terms (SAR 240,000 uncollected), while its suppliers demand cash (SAR 380,000 paid immediately). The cash result for April: **SAR -210,000** — a liquidity crisis in a month it reported a profit. That is exactly what the 13-week model exposes. It does not care when an invoice falls due; it cares when money actually hits the bank and when it actually leaves. ## Building a 13-week model from scratch **Step 1: Establish your opening cash balance** Add up your bank accounts, your cash floats, and any card-terminal (POS) cash not yet settled or deposited. That figure is 'Week 0'. **Step 2: Build the structure (a spreadsheet)** Across the top: 13 columns (W1 through W13). Down the side: - Opening cash - Cash In line items — 6 to 10 rows - Total cash in - Cash Out line items — 8 to 15 rows - Total cash out - Net cash flow = cash in − cash out - Closing cash = opening + net **Step 3: Start with one week, at very high accuracy** Week 1 should be close to certain. Pull the customer invoices due this week, the payroll run, and the supplier invoices you have received. These are hard numbers, not guesses. **Step 4: Extend outward with declining precision** Weeks 2–4: base it on actual due amounts and signed contracts. Weeks 5–8: base it on historical patterns (average weekly sales). Weeks 9–13: base it on plans and expectations (a marketing campaign, a contract you expect to sign). **Step 5: Drop in the irregular items on their real dates** Zakat and tax payments to the Zakat, Tax and Customs Authority (ZATCA) on the 15th of the following month, month-end payroll, quarterly rent, annual vehicle insurance. These large, infrequent items are the landmines of liquidity. **Step 6: Colour-code the closing balance** Closing cash: green (safe — above your floor of 30 days of payroll), amber (caution — 15 to 30 days), red (danger — under 15 days). If red appears in any week, act now, not then. ## How to forecast receipts accurately (Cash In) Receipts are the hardest side to forecast, because customers do not reliably honour their payment dates. Four techniques will sharpen your accuracy: **1. Grade receivables by probability of payment** Not every customer is the same. Sort them: - Excellent payers (on time 95% of the time): count the full amount. - Good payers (80–95%): count 90% of the amount. - Erratic payers (50–80%): count 70%. - Distressed payers: 30–50%, or park them under 'unconfirmed receipts'. **2. Analyse each customer's payment history** An accounting system like Snad gives you a 'days sales outstanding (DSO) by customer' report. A customer with a DSO of 38 days means that if you invoice today, you should expect the cash in 38 days — not 30, whatever the invoice terms say. **3. Treat forecast sales conservatively** For the distant weeks, use: average weekly sales over the last 13 weeks × a seasonal factor (if Ramadan or the holiday season falls inside the window). Do not plug in the growth number you *want*. **4. Separate immediate cash from receivables** A cash sale at the POS lands the same day. A credit sale lands X days later. Your model has to split them, not lump them together. ## How to forecast payments accurately (Cash Out) Payments are easier to forecast because most of them are known in advance. But the details matter: **Fixed and known items** - Payroll: headcount × average salary, plus General Organization for Social Insurance (GOSI) contributions at 9.75% employee share + 11.85% employer share. - Rent: on its fixed dates (monthly or quarterly). - Fixed subscriptions: SaaS, internet line, management allowances. - Loan and financing instalments: on their published schedule. **Variable but estimable items** - Cost of goods sold (COGS): as a percentage of forecast sales. - Electricity and water bills: the average of the last 3 months. - Fuel and maintenance: based on your vehicle fleet. **Items you must not forget** - Zakat and tax (reminder: the return is due by the 15th of the following month). - End-of-service award, if you plan to release an employee. - Annual vehicle insurance. - Municipality licence and Commercial Registration (CR) renewal. **One important trick**: create a line called 'unforeseen items' worth 5–8% of total expenses. No model ever survives without surprises. That buffer line is what protects the accuracy of everything else in the model. ## Updating the model weekly and learning from variances A model that is never updated is just paper. Every Monday morning, close out the week that just ended: **1. Replace the forecast with the actual** For every line: what was actually received or paid? Keep the forecast figure in an adjacent column so you can measure the variance. **2. Calculate the variance** The gap between actual and forecast, in riyals and as a percentage. Anything that moved more than 15% needs an explanation. **3. Ask 'why' for every material variance** Did customer C pay late? Did a supplier demand an advance you had not planned for? Did you overspend on a particular line? Every answer adjusts your assumptions for the weeks ahead. **4. Add a new week on the end (W14)** You swap the week that just closed for a fresh one, so you always keep a full 13 weeks of visibility ahead of you. This is what is called a **rolling forecast**. **5. Write the lessons down** A short file of plain observations: 'individual retail customers run 5–8 days past terms', 'summer electricity bills are 35% higher than winter'. After 3 months you have institutional knowledge that transforms your accuracy. ## A worked example: a Saudi store selling retail and wholesale A building-materials store in Riyadh, opening balance SAR 180,000 on 1 June 2026. Its sales: - Retail, cash: SAR 22,000/week on average (lands the same day). - Wholesale on 30-day credit: SAR 60,000/week on average (lands roughly 30 days later). Fixed monthly costs: - Payroll: SAR 38,000 (month-end). - Rent: SAR 12,000 (1st of the month). - Utilities and internet: SAR 3,500. - Fuel and maintenance on the delivery vehicles: SAR 4,800. - GOSI: SAR 4,600 (15th of the month). - Inventory purchases: SAR 55,000/week on average. Week 4 of the model reveals the problem. Cash in for W4 = 22 + 60 = SAR 82,000. Cash out for W4 = 38 (payroll) + 12 (rent) + 55 (inventory) = SAR 105,000. Net for W4 = **SAR -23,000**. The closing balance drops by SAR 23,000. Three weeks like that in a row eat SAR 70,000 out of an opening balance of SAR 180,000, leaving SAR 110,000 after a month. Had he not noticed, a critical cash squeeze would have ambushed him 60 days later. **The decision the model unlocks**: negotiate with three of his largest wholesale customers to shorten payment terms from 30 days to 21, in exchange for a 1.5% discount. The result: an extra SAR 36,000 landing every month. That is the value of the 13-week model — he saw the problem 60 days before it arrived, which left him room to negotiate rather than to panic. ## How Snad builds the forecast automatically from your data Snad generates the cash flow forecast automatically by connecting your live data to the forecasting model: - **Receipts pulled automatically from sales invoices**: every outstanding invoice is placed in the week it is expected to be collected, using that customer's historical average collection days. - **Payments pulled automatically from purchase invoices**: supplier bills appear in the week they are due under their agreed terms. - **Saved recurring items**: payroll, rent, subscriptions — defined once, then repeated automatically. - **Proactive alerts**: when a closing balance is forecast to fall below your safety floor, you are told which week and by how much. - **Variance analysis**: at the end of each week, actual is compared against forecast and every line that moved more than 15% is flagged. - **What-if scenarios**: what if sales rise 10%? What if a major customer slips another 30 days? Snad generates the scenario in minutes. The 30-day free trial is long enough to build a complete 13-week model and watch your forecast accuracy improve after just two weeks of weekly updates. ## The practical takeaway for a business owner Five rules for a 13-week model that will not let you down: 1. **Start this week, not next month**: a crude version in Excel beats a 'perfect model, coming later'. Update it for four weeks and the payoff will surprise you. 2. **Set your cash safety floor explicitly**: a common rule of thumb in the Saudi market is never to let your balance fall below 30 days of fixed costs. For a company with SAR 150,000 in monthly fixed costs, the floor is at least SAR 150,000. 3. **Put zakat, tax and GOSI in on their real dates**, not as a 'monthly average'. Returns have to be settled by a fixed deadline, and late payment carries financial penalties. 4. **Share the model with a partner or a finance manager**: two sets of eyes are better than one at catching a bad assumption. 5. **Treat the model as a decision tool, not a reporting tool**: every week it should push you to a specific decision — negotiate with a customer, defer a payment, apply for a bank facility, postpone a hire. A model that produces no decisions is just paper. A Saudi company that runs the 13-week model consistently for six months ends up able to see crises before they land, capture early-payment discounts from suppliers, and negotiate with banks from a position of strength. The biggest benefit of all: you sleep soundly, because you know exactly what is going to happen in your bank account over the next 91 days. ## Frequently asked questions ### What is the difference between a 13-week model and a cash flow statement? A cash flow statement is a historical report that summarises what happened over the past 3, 6 or 12 months. The 13-week model is predictive: it tells you what will happen over the next 91 days. The first is for accountants and investors; the second is for the person making operating decisions. ### Do I need an accounting qualification to build the model myself? No. Mathematically the model is trivial — addition and subtraction only. The difficulty is weekly discipline, not accounting theory. If you know basic Excel, you can build the model in two working days. ### What do I do if I see a red balance in week 7? Escalate in stages: (1) review your overdue customers and call them to accelerate payment. (2) Negotiate with two or three major suppliers to push a payment out by two weeks at no cost. (3) If the red signal persists, apply for a bank credit facility — approval takes 4–8 weeks, so start early. (4) In critical cases, postpone a hire or defer non-essential spending. ### Does the model work for a services business with no inventory? Yes, and sometimes it is simpler. Services firms typically have a longer collection cycle (45–60 days from the client) against weekly or monthly payroll. The model exposes that timing gap clearly and forces the owner to manage collection terms firmly. ### How accurate should I expect the model to be? At the start, a 20–30% variance is normal. After three months of weekly updates and learning from the misses: 8–12% variance for weeks 1–4 and 15–20% for weeks 8–13. That is more than accurate enough to make decisions ahead of events. ### Should I update the model if a customer pays a few days late? For a day or two, no need. For five days or more, yes — update it. A customer who slipped once will probably slip again. Adjust the collection assumption for the coming weeks and add a 'customer payment delay buffer' line at 8–12% of forecast receipts. --- ## About the publisher **Snad (سند)** — a private Saudi software company based in Riyadh, founded 2025. Legal form: Sole proprietorship. Commercial registration: 7038154642 VAT number: 310959226500003 Only official domain: snad.io > Snad is a private commercial business-management platform. It is not a > government body, not a bank, and not a government services portal, and it > is not affiliated with any government entity. Any site or app with a > similar name is unrelated to Snad.