What this is
You run the restaurant. NO.2 runs the numbers. It turns your sales history and your cost targets into the three numbers you decide with every day: how much you need to sell, how many labor hours you can afford, and how much you can spend on food and beverage this week.
It is deliberately not accounting software, not a BI dashboard, and not a scheduling system. It sits on top of what you already use and tells you where you stand while you can still do something about it.
The problem it solves
Most operators can get these answers eventually, from a spreadsheet or a P&L that lands three weeks late. This answers them before service:
- How much are we likely to sell this week, and are we ahead or behind right now?
- How many labor hours can today's sales actually pay for?
- How much can I spend on food and beverage this week without breaking the month?
- Did last week's forecast actually hold, and by how much?
- Which parts of the day make money, and which quietly lose it?
How it works
- Set your targets once. Labor percentage, loaded wage, and food and beverage cost targets.
- It forecasts what you will sell from your own trading history, day by day.
- That converts into guardrails — labor hours per day and purchasing budgets per week.
- You operate against pace and get a text or email only when something is worth acting on.
- The week closes against the forecast it was given — including the weeks it got wrong.
- It checks itself against what actually happened, and changes what it uses when something works better. It does not learn; it measures.
What you would actually receive
A short morning brief, alerts during service only when a rule you set fires, a nightly recap, and a weekly plan and review. Every message carries sales and labor against what they should be, and the gap in dollars and hours — never a bare percentage like "labor variance +1.6%", and never an instruction about your floor.
Open the Alerts tab to see all of them, exactly as they would arrive by text and email.
How to use this demo
Seven screens. Everything is live — change an input and every number moves with it.
BriefToday's goal, what is left to spend and schedule, and at most three signals. Try: read only this screen and see if you know what to do today.
WeekThis week's labor hours and purchasing budget, and the seven days. Tap a day to go period by period. Try: enter this week's food purchases and watch the Brief react.
PlanningThree weeks at once — what last week actually did, where this week stands against its plan, how accurate past forecasts were, and next week's labor and purchasing before you commit to them. Try: push next week’s sales up and watch its hours and budgets move with it.
SignalsEverything open, with HEARD ✓ to clear it, plus what the last four closed weeks prove.
AskThe questions operators ask daily, answered from your own numbers.
MoreTargets (the engine), loans and debt, weekly review, history, alert previews, service periods. Try: set the hourly labor target to 26% and watch labor move everywhere.
What is real and what is not
Copper Cactus Kitchen is not a real restaurant, but its numbers came from one. The weekday sales curve, the Lunch/Dinner/Brunch day-part mix and the labor-to-sales ratio are taken from a real single-location full-service restaurant and bar’s own trading history — rescaled and re-dated so nothing here is the real tenant’s actual dollars or calendar, and recomputed against today’s date on every load rather than frozen to a fixed week. NO.2 is the product; Copper Cactus is the demo tenant. A three-location version of this same demo, modelled the same way on a real multi-location coffee brand, is at ?mode=demo&kind=multi.
The math is real. Every calculation is deterministic and would run the same way on your data — no estimates generated by AI.
Not yet connected. In this demo you enter sales, labor, and purchases. In the product, sales and labor cost come from your POS and payroll; purchases stay manual at first.
What comes later
POS, payroll, and accounting integrations; scheduling systems so recommended hours can be compared to what is actually scheduled; invoice ingestion for purchasing; then multi-location roll-ups.
What I would love your feedback on
- Which single number here would you check first every morning?
- What would you need to see before you would trust it enough to schedule against?
- What is missing that you cannot run your restaurant without?
Reply to the message that sent you this link with anything that felt confusing, wrong, or missing. Blunt is useful.