Change one cell. Watch the whole plan move.
Three tools for the model you have already built: name your scenarios and switch between them, backsolve for the input that hits a number, or sweep a whole grid of inputs and read the results as a table.
Method guide for Google Sheets Three what-if tools free, up to 32 changing cells a scenario
What-if analysis in Google Sheets, three tools
Scenarios saves a set of input cells under a name. Base, Bear, Bull, or the three versions of the plan the room keeps arguing about. Switch between them in a click and write a summary report that puts them side by side.
Goal Seek runs the model backwards. You know the answer you need and you want the input that gets you there: the price that breaks even, the grade that saves the semester, the day rate that pays for the year. If it cannot reach the target it tells you how close it got, rather than quietly returning something wrong.
Data Table sweeps. A one-variable table walks one input across a list of values and records every result you asked it to watch; a two-variable table crosses two inputs and gives you the grid, which is what the room actually wants to look at when growth and margin are both up for debate.
Which one answers your question
All three are in the What-If panel, all three are free, and all three work on the model already in your sheet. Nothing is exported and nothing is rebuilt.
Compare named versions
- Scenarios save input sets under a name, switch between them, and print the summary.
Work backwards from an answer
- Goal Seek find the input that makes a formula hit the number you need.
Sweep a range and read the grid
- Data Table (one-variable) one input across many values, tracking several results.
- Data Table (two-variable) two inputs crossed, one result, drawn as the grid.
When the input is a range rather than a list
- Monte Carlo risk simulation every value at once, weighted by how likely it is, with odds on the answer. A Pro engine, on the Monte Carlo page.
Scenario planning without a second copy of the model
Scenario planning usually means three saved copies of the same file, and by Thursday nobody is sure which one the board saw. Scenarios keeps the base, the bear and the bull case as named sets of input cells in one spreadsheet, so switching case is a click and the model underneath every case is the same model.
The summary report puts the cases in columns next to each other, so the meeting argues about the assumptions rather than about which file is current. When somebody then asks how likely the bear case actually is, that is the question a Monte Carlo run answers.
A sweep is not the same as a simulation
A Data Table answers “what happens at each of these values”. A simulation answers “how likely is each of them, and what does that make the answer”. The table is the right tool when you have a shortlist and the room wants to see the grid; the simulation is the right tool when the input is genuinely uncertain and nobody can name the value.
They live one panel apart, and a model built for one runs in the other. Start with the sweep, and when somebody asks “but how likely is the bad column?”, the Monte Carlo guide is the next page.
Free or Pro
Scenarios, Goal Seek and Data Tables are free, and no plan buys a bigger one. There is no cap on the number of scenarios you keep, on how many result cells you read back, or on how often you run. The engines carry two ceilings of their own: one scenario holds at most 32 changing cells, and one axis of a data table takes at most 1,000 values.
Pro covers the five simulation engines and the AI readings. Pro is $199/year, and a $9 Day Pass covers the week of a single decision.
Worked examples to start from
Each one loads with the sweep or the target already set up, so the first thing you see is the answer and the second is how it was built.
- The planning meeting, as one grid of growth against marginHow Do Growth and Margin Move EBITDA Together?
- Exactly what you need on the finalWhat Do You Need on the Final?
- Raise prices, and see how much of the rise survivesIf We Raise Prices, Do We Come Out Ahead?
The other 45 models in the library that use one of these:
- Pay a little extra each month. How much sooner is it gone?How Much Sooner Does an Extra Payment End the Loan?
- What should your day rate actually be?What Day Rate Does a Freelancer Need to Charge?
- Your test came back positive. What are the odds it is real?Positive Test, Real Risk
- Does the launch still pay after it eats the flagship?Launch With Cannibalization
- What does that broken process actually cost per year?What a Broken Process Costs per Year
- What CPA will your ad budget actually buy?PPC Campaign Planner
- How many rentals until an item pays for itself?Rental Break-Even Turns
- Which risks stay red after controls?IT Risk: Still Red After Controls?
- What does skipping the salary negotiation really cost you?What Does Not Negotiating the Salary Cost Over a Career?
- How many survey responses do you actually need?Survey Sample Size Planner
- What's the most you should pay before you walk away?Walk-Away Price Calculator
- If you raise your bid, what does it actually buy?Ad Auction Simulator
- When the company sells, what do you take home?Exit Payout Reality Check
- Is the feasibility study worth it before you bid?Should We Pay for a Feasibility Study Before Bidding?
- How much gross profit does each inventory dollar earn?What Does Each Dollar of Stock Earn in a Year?
- Where does a 35% hurdle rate actually come from?What Return Should a Risky Project Have to Clear?
- Should you sell the hardware below cost to win the platform?Platform Launch Pricing
- If we build one 50% bigger, what will it actually cost?What Does Twice the Capacity Cost?
- Claim at 62, 67 or wait until 70?When Should You Take Social Security?
- Is an electric car actually cheaper to own?
- How much should you order each time?How Much to Order Each Time (EOQ)
- How many servers does the queue actually need?How Long Will the Line Get?
- Do the decision rules agree on what to build?Which Option Wins, and at What Odds?
- How far is the campaign from breaking even?The Conversion Rate That Breaks Even
- What will you weigh six months from now?How Much Weight Will I Lose?
- Pick a rate you can hold. What date does that give?How Long Until I Hit My Goal Weight?
- How far apart are your best year and your worst?Base, Better, Worse: Save the Three Cases
- What does each of the three budgets actually produce?Which of Three Budgets Should the Practice Plan On?
- What do the lean, normal and cautious bids each look like?Three Ways to Price This Bid
- What is the lowest rate that is not a subsidy?What Rate Do I Have to Charge?
- When does the next unit of stock stop paying?How Much Safety Stock Do We Actually Need?
- How much stock does a longer lead time cost you?When Should We Reorder?
- What does a 95% promise cost, in units?What Stock Level Holds a 95% Service Level?
- Does discounting to fill the room ever work?What Does the Course Have to Charge?
- What would have to change for this channel to pay?What Conversion Rate Makes This Channel Work?
- What does hiring now cost you next January?What Three Hiring Plans Cost
- Is your markup quietly costing you margin?What Markup Gets This Bid to Target Margin?
- What does an unreliable supplier cost you in stock?Safety Stock for a Plant That Cannot Stop
- How many tickets must you sell to break even?What Ticket Price Covers the Room?
- What reserve holds nineteen years out of twenty?How Much Should We Reserve Per Matter?
- How much cash does the bigger buy tie up?Which Season Plan Should the Chain Commit To?
- What does half a point of fee cost the investor?What Fee Does the Fund Need to Clear the Hurdle?
- Which media split buys you the most orders?Three Media Plans, Compared
- Is the Extra Shift Worth It?
- Overtime or Agency?
Try it in your own sheet
- Open Sortia in Google Sheets and choose Start from a template.
- Pick one of the models above, and it loads with the inputs filled in.
- Change the assumptions to fit your situation and press Run.
Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.
Other methods: Monte Carlo Decision trees Schedule risk Critical chain Optimization under uncertainty Statistics Machine learning Forecasting Optimization