13-Week Cash Flow: The Minimum Dashboard for Owners Who Hate Surprises

1. Why does a profitable restaurant go under?

The owner goes broke from illiquidity, not from a lack of profitability. 60% of restaurants that close showed a positive margin on their last P&L:

a venue with a 12% operating margin can run out of cash by the third week of the month because of the mismatch between collection and payment. Payroll is paid every 15 days, VAT falls due on a fixed date, rent doesn't wait; but delivery settles at 21 days and corporate accounts at 30 or 45. Accrual accounting records the sale the day it happens and hides that timing trap. Cash flow exposes it. I've seen it in dozens of restaurants: the P&L says you earned $8,000 for the month, and the bank shows $1,200 on the Thursday you owe $14,000 in payroll. It isn't a margin problem; it's a calendar problem. The 13-week board orders that calendar before the bank balance decides for you.

2. Accrual versus cash: two different truths

The P&L tells you if you're earning; the 13-week board tells you if you can pay. They are two different questions and 80% of owners only look at the first. Accrual recognizes income and expenses when they are generated, not when the money moves: it records the catering sale on the event day even if you collect at 30 days, and it spreads the annual policy into twelfths even though you paid it in full in January. That logic is right for measuring profitability and terrible for measuring survival. The cash board flips the focus: it counts only the money that enters and leaves the account that week, with its real date. A restaurant with 30% food cost and 28% labor —a healthy 58% prime cost— can have 4 straight weeks of negative cash if it clustered three big payments into the same stretch. The board measures exactly that: not whether the business is good, but whether it survives next Tuesday.

3. The rolling 13-week horizon

Thirteen weeks is the horizon that combines precision and useful anticipation: a full quarter, enough to see a tax payment or a slow season coming, and short enough for every line to stay credible. The board is rolling: each Monday the finished week drops off and a new one is added at the end, so you always have 13 ahead. This turns forecasting into a continuous muscle, not an annual budget filed away in February and forgotten. Each week fills a column with three blocks: opening balance, cash inflows (sales collected by channel), outflows (payroll, suppliers, rent, taxes, debt) and closing balance, which starts the next week. Diego F. Parra at Masterestaurant sums it up like this: the annual exercise predicts poorly and late; the rolling 13-week board misses little because it corrects every 7 days. The discipline of updating it on Mondays, 30 minutes, is worth more than the sophistication of the model.

4. Prime cost granularity by week

Traditional control reports aggregated monthly food cost; the board breaks food cost and labor down week by week and detects the leak in the exact line item and the exact week. A monthly food cost of 31% can hide a week at 38% offset by another at 25%: the average lies, and by the time you close the month you've already lost three weeks of margin. Broken out by week, the deviation jumps the following Tuesday, when you can still renegotiate with the protein supplier or adjust the weekend menu. The same with labor: a week with a private event pushes overtime to 33% labor cost, and the board isolates it instead of diluting it. The hard Masterestaurant rule: food cost per plate never above 32%, and prime cost under 60%. Measured by week, that ceiling stops being an end-of-month aspiration and becomes an actionable alert within 7 days.

5. From data to decision: the board's architecture

The projected balance stops being a data point and becomes a decision architecture. When you see week 6 dropping to $900 in cash and week 7 demanding $11,000 in payroll plus VAT, you have five weeks to act, not five days. That margin changes everything: you negotiate a 45-day term with a supplier instead of 30, you split the $6,000 oven purchase into two tranches, you pull forward corporate invoicing, or you freeze a hire. The board doesn't make the decision; it enables it in advance. Without it, the bank balance dictates by urgency: you pay whoever shouts loudest and defer whoever suits you most. With it, every week in the red lights an alert and every alert has three known levers: collect sooner, pay later, or cut the outflow. The mistake I see again and again is owners brilliant in the kitchen who discover the cash problem on payment day, with no room to maneuver.

6. How to build your first 13-week board

Start with today's real bank balance and thirteen weekly columns; in under an hour you have version 1. Load the inflows by channel with their real collection term: cash and card the same day, delivery at 21 days, corporate at 30-45. Then the fixed outflows with their exact date: payroll on the 15th and 30th, rent on the 1st, VAT on its due date, debt installments. After that the variables: suppliers per your purchasing cycle, typically 60-70% of sales in prime cost. Each week's closing balance feeds the opening of the next. The first time you'll discover two or three weeks in the red you didn't suspect; that finding alone pays for the exercise. Update it every Monday with what was actually collected and paid, correct the projection and add week 14. At Masterestaurant we set it up on a simple sheet: the discipline of 30 weekly minutes weighs more than any $200-a-month software.

7. The three errors that drain the cash without you noticing

The costliest error isn't overspending, it's failing to see the mismatch in time, and it shows up in three forms. First, confusing sales with collection: you record $40,000 in sales for the month but $9,000 are still in the delivery gateway at 21 days, and you plan expenses against money that hasn't arrived. Second, ignoring payments that don't appear in the weekly P&L: debt installments, VAT, bonuses, the annual policy; these are cash outflows that aren't that month's expense and they throw off the projection if you don't load them. Third, buying inventory against the sales peak instead of against available cash: you fill the cooler before the big weekend and come up short for Monday's payroll. The 13-week board neutralizes all three because it forces every dollar to carry its real movement date. Diego F. Parra repeats it in every diagnostic: cash doesn't lie, the P&L sometimes does.

8. The 4 differences that change the survival equation

Accrual vs. cash: the P&L tells you if you earn; the 13-week dashboard tells you if you can pay. A restaurant can post a 12% operating margin and still run out of cash in the third week of the month due to the mismatch between collection and payment. The dashboard measures the latter. Rolling horizon: each week the oldest one drops off and a new one is added. You always have 13 weeks ahead. This turns projection into a continuous muscle, not an annual exercise that gets filed and forgotten. Prime cost granularity: traditional control reports aggregated monthly food cost; the dashboard breaks food cost and labor down by week, catching capital leakage in the exact line item and the exact week it spikes. Decision architecture: the projected balance stops being a data point and becomes a traffic light. Green buys CapEx; yellow phases payments; red triggers the containment protocol. The decision gets standardized and stops depending on the owner's mood.