Eduardo Salinas has run a restaurant in San Antonio, Texas, for 8 years. Contemporary Mexican cuisine, 60 seats, a strong local reputation. But in December 2025, he almost ran out of cash to make payroll at year's end.
Not because the business was struggling. November sales had been the best of the year. The problem was that Eduardo didn't see the crisis coming because he was making financial decisions with data that was 6 weeks old.
His process: the accountant sent the P&L and balance sheet around the 20th of each month, covering the previous month. If something went wrong in October, Eduardo found out on November 20th. By then, the damage was already done.
December 2025 had been an unusually strong year in sales, which led Eduardo to make three investments in Q4: new kitchen equipment, a bathroom renovation, and an advance payment to a wine supplier to lock in prices. All reasonable on paper. The problem: the three large payments landed during weeks of high pre-holiday inventory — and two corporate clients paid their event invoices late. The result: $23,000 less in the account than expected the week before payroll.
"If I could have seen the cash flow in real time, I would have staggered those payments. But all I had was October's P&L."
Why the monthly P&L isn't enough for operational decisions
The income statement (P&L) is an accounting tool — it's designed to measure the performance of a past period. It's excellent for talking to the bank, filing taxes, and comparing years. But it's nearly useless for day-to-day or week-to-week operational decisions.
What a restaurant owner needs to operate well isn't knowing how much they made last month. It's knowing:
- How much cash is available today, and what payment obligations are coming in the next 14 days?
- Am I on track to beat or miss projected sales for this week?
- What's the actual cost of my menu items this month vs. last month, given current ingredient prices?
- When should I pay which suppliers to avoid compromising payroll?
None of those questions are answered by last month's P&L.
The real-time financial dashboard system
We built a three-layer system that converts the restaurant's operational data into a financial view updated every 24 hours.
Layer 1: Real-time sales data (from the POS)
The restaurant uses Toast POS — one of the most common systems in U.S. restaurants. Toast has an API that lets you pull sales by day, by shift, by menu category, and by payment method. n8n extracts this data every night at 11 PM and loads it into the Power BI data model.
What this lets you see: Month-to-date sales vs. budget, sales trend by day of the week, average ticket by shift (lunch vs. dinner), and best and worst performing menu categories.
Layer 2: Payment obligations (from QuickBooks + payment calendar)
The most critical side of the dashboard is the cash outflow projection. We connected QuickBooks to capture outstanding payables and built a fixed commitment calendar (payroll on the 1st and 15th, rent on the 5th, estimated utility costs, vendor payment cycles).
The result is a visual timeline showing — for the next 30 days — what payments are going out and when, compared against projected available cash based on the current sales pace.
The question this view answers: Is there any point in the next 30 days where projected cash drops below $X? If yes, Eduardo sees it 3–4 weeks in advance, not 6 weeks after the fact.
Layer 3: Menu cost and margin per dish
This layer required upfront work: building a recipe table with the cost of each ingredient, connected to current supplier purchase prices (updated manually by the executive chef once a week in a Google Sheet).
Power BI crosses this information with POS sales data to calculate the actual food cost in real time, and the margin per dish. When an ingredient price goes up (like avocado in summer), the dashboard automatically reflects that in the margin of every dish that uses it.
What this lets you see: Which dishes have the best margin this week, what's happening with the overall food cost vs. the 28–32% target, and which dishes to promote this week to improve the overall margin.
The main dashboard views
Daily view (Eduardo checks on his phone every morning):
- Yesterday's sales vs. same day last week
- Cash in account today
- Next large payment in the next 7 days
- Food cost for the last 7 days vs. target
Weekly view (reviewed on Mondays with the general manager):
- Month-to-date sales vs. budget
- Month-end projection (if current pace continues)
- Top 5 dishes by sales and by margin (not always the same list)
- Alerts: any KPI more than 10% outside its target range
Monthly view (for the accountant meeting):
- P&L vs. Power BI projection comparison (to validate data accuracy)
- 12-month trends in sales, food cost, and margin
- Seasonality analysis to plan the next quarter
Technical implementation
Tools used
- Power BI Desktop + Power BI Service: Dashboard build and publishing (accessible from phone and computer)
- Toast POS API: Real-time sales and operational data
- QuickBooks Online API: Accounts payable, expense history, bank account balances
- Google Sheets: Recipes and ingredient costs (manual weekly update)
- n8n: Nightly data pipeline connecting all sources and refreshing the data model
Implementation time: 3 weeks
Week 1: Connecting Toast and QuickBooks, cleaning 12 months of historical data, defining KPIs with Eduardo.
Week 2: Building the dashboard, creating the recipe table with the chef, configuring the n8n pipeline.
Week 3: Validating data with the accountant, adjusting QuickBooks expense categories to match dashboard categories, training Eduardo and his general manager.
Results at 5 months
- Liquidity crises: 0 since the dashboard went live (vs. the December 2025 near-miss that almost ended in payroll default)
- Average food cost: from 34% to 29% — With weekly visibility of cost per dish, the chef adjusted three low-margin dishes and Eduardo made promotion decisions based on margin, not just popularity
- Menu items removed for low margin: 4 — Replaced by more profitable alternatives the chef already had in mind but had never had numeric justification to propose
- Meetings with the accountant: shifted from "what happened last month?" to "what do we do next month?" — The accountant now arrives at the meeting to do strategic analysis, not explain numbers Eduardo already knows
Eduardo summarized it in the 5-month review: "Before, I was driving my restaurant in the fog. I could feel whether things were going well or badly, but I couldn't see how close I was to the edge. Now I see it weeks ahead."
What we learned
1. A 30-day cash flow projection is worth more than last month's P&L.
The P&L tells you where you were. The cash flow projection tells you where you're going. For daily operations, you need the second one.
2. Real-time food cost changes how the chef thinks.
When the chef can see the margin of each dish in real time, they start thinking like a business partner, not just a cook. Menu decisions stop being purely creative and become financial as well.
3. The recipe table is the most valuable asset — and the hardest to build.
Documenting every recipe with ingredient costs takes time and discipline. But once built, that table becomes the foundation for all menu intelligence. It's worth every hour invested.
4. The dashboard should be designed for the worst moment, not the best.
The question that guided the design was: what would Eduardo need to see if the business were at risk and he had 5 minutes to decide what to do? The answers to that question are the most important metrics in the dashboard.
Do you know today how much cash your business will have in 3 weeks?
If the answer is "roughly" or "I'd have to ask my accountant," there's a gap between what you see and what you need to see to operate well. A real-time cash flow dashboard can close that gap.
Schedule a 30-minute call and we'll review what data you already have and how to turn it into real financial visibility.