Financial decision modeling

Most margin problems are decision problems.

Prices, promotions, customer terms and product mix are decided long before the P&L reports the result. The tools and worked analyses here let you test those decisions before you make them.

Browse the tools Talk through your plan
Plan: profit at average demand
$150,000
Average of 100 futures
$110,200
−$39,800 vs plan
Futures that beat the plan
0 of 100
Monthly profit in 100 plausible months, against the planMonthly profit against units demanded. Each unit contributes $50 and fixed costs are $350,000, so profit rises with demand up to capacity of 10,000 units a month, then stays at $150,000. 100 plausible months, with demand averaging 10,000 units at ±20% uncertainty, sit on that line, and in 50 of them demand is above capacity. The plan, profit at average demand, is $150,000. The average of the 100 months is $110,200, and 0 of 100 beat the plan.4,0008,00012,00016,000−$300k−$200k−$100k$0$100k$200k$300k$400kMonthly profitUnits demanded in the monthAverage outcomePlanCapacityMonthly profit in 100 plausible months, against the planMonthly profit against units demanded. Each unit contributes $50 and fixed costs are $350,000, so profit rises with demand up to capacity of 10,000 units a month, then stays at $150,000. 100 plausible months, with demand averaging 10,000 units at ±20% uncertainty, sit on that line, and in 50 of them demand is above capacity. The plan, profit at average demand, is $150,000. The average of the 100 months is $110,200, and 0 of 100 beat the plan.6,00012,00018,000−$300k−$200k−$100k$0$100k$200k$300k$400kMonthly profitUnits demanded in the monthAverageoutcomePlanCapacity
  • One plausible month
  • Demand above capacity
  • The plan
  • Average outcome

Average demand 10,000 units · $50 contribution a unit · $350,000 fixed costs

Demand averages exactly what the plan assumed. With capacity sized to that average and ±20% uncertainty, capacity caps the good months but not the bad ones, so no month beats the plan and the average month falls about $40,000 short.Illustrative figures.
Start with the decision

What are you deciding?

Each question links to a free tool or a worked analysis. Use them as they are, or bring the decision to a conversation.

The method

How each analysis is built

  1. Measure what the P&L leaves out

    Costs booked to logistics, operations and receivables are pulled down to the customer, product or decision they belong to. Much of the answer usually sits here.

  2. Model how the business responds

    Volume responds to price, sales respond to promotions, and forecasts carry error. Each response is estimated from the data, and its uncertainty is stated rather than assumed away.

  3. Choose within real constraints

    Capacity, cash and time are limited. The recommendation is the best option inside those limits, with the trade-off shown, so the decision stays with the people who own it.

AI does the work that doesn't need judgment: assembling data, checking a forecast against its record, drafting the first read. Deciding what belongs in the model, and what the result means, is finance work.

Toolkits

Run the model on your own data

The calculators answer one case at a time. Toolkits cover the whole portfolio, with the model, the method and a guide.

Available now

CFO Monthly Close Toolkit

Free

Drop in your raw monthly financials and get the full board pack back in one run: an interactive dashboard, a 9-slide deck, a formula-driven Excel workbook, and written management commentary. The mechanical assembly is done before you start reviewing, so your time goes to the analysis instead. Runs as a Claude skill, no integration to maintain.

Download skill Excel template User guide (PDF)
Free

Forecast Review Toolkit

Weekly actuals against plan and prior year become a driver-based forecast for the rest of the year, with the assumptions on screen and the maths reconciled. Get it →

 Free

Pricing & Optimization Toolkit

Willingness to pay, cost to serve and constrained mix solved together, returning a ranked decision set where every recommendation carries what it is worth. Get it →

Get notified when new toolkits launch

Insights

Recent analysis

Worked examples, one decision at a time.

Pricing · Statistics Framework

A Promotion Spike Is Not a Demand Curve

A 20% promotion lifted volume 80%. Read one way, the data says hold price. Read correctly, it says raise it.

Read →
Reporting · Variance Framework

Which Variances Deserve Your Attention

40 variances, 40 comments, and a CFO who learns nothing. Two filters turn that into 3 signals worth acting on.

Read →
Forecasting · Planning Free model

Scenarios Are Not Sensitivities

The same downturn modelled both ways: $825,000 of profit as a sensitivity, $556,000 once the linked drivers move together.

Read →

All insights

Behind the work

Yehuda Slater

I'm Yehuda Slater, and I'm based in Toronto. My PhD in economics at the Hebrew University of Jerusalem was built on optimization models that allocate scarce resources under real constraints. My corporate work has been in FP&A: forecasting, budgeting, capital planning and margin analysis. The tools on this site come from putting those two together.

PhD Economics, Hebrew University of Jerusalem · Corporate FP&A · Published research in resource economics

More about my background

Bring a decision you're weighing

A price increase, a customer that may cost more than it earns, a forecast the business has stopped trusting. Send a few lines about it, and I'll reply with how I would think it through.

Start a conversation