b2KIT

Sensitivity Analysis Table Builder

Create two-variable data tables showing how outputs change as two key assumptions vary with conditional formatting heat maps.

Tested tool guide Tested browser tools Checked August 16, 2026

What Sensitivity Analysis Table Builder does, with a checked example

Enter one formula and a range for each of two inputs, and this tool evaluates the formula at every pairing of the two ranges, laying the results out as a grid with one assumption across the columns and one down the rows. Every cell is the output for that exact combination, so the grid shows interaction, not two independent effects. The heat map colors each cell relative to the table's own lowest and highest values, which is what people misread first: the deepest color marks the strongest cell of this table, not a universal verdict of good or bad.

Worked example

A concrete input and expected output from the current implementation.

Input

Output formula: Profit = (Price - Cost) * Volume - FixedCosts
Constants: Cost = 8, FixedCosts = 10000
Variable 1 (columns): Price from 10 to 18, step 2
Variable 2 (rows): Volume from 500 to 2000, step 500

Expected output

Columns: Price 10, 12, 14, 16, 18. Rows: Volume 500, 1000, 1500, 2000.

          P=10    P=12    P=14    P=16    P=18
V=500    -9000   -8000   -7000   -6000   -5000
V=1000   -8000   -6000   -4000   -2000       0
V=1500   -7000   -4000   -1000    2000    5000
V=2000   -6000   -2000    2000    6000   10000

Negative cells render red, positive cells green, with deepest shades at the grid's extremes.

Every cell is (Price - 8) * Volume - 10,000 for its row and column, so each row isolates price's effect at fixed volume and each column isolates volume's effect at fixed price. Only one combination lands exactly on zero; the red-to-green boundary shows where profit turns positive between grid points.

How the result is produced

1

Grid evaluation

You give an output formula written in terms of two named variables, plus a start, end, and step for each. The tool evaluates the formula at every combination of the two ranges and lays the results in a matrix, one variable across the columns and one down the rows. Nothing is interpolated: every cell is an exact evaluation at that pair of values.

2

Heat map scaling

Each cell is colored by its position within the table's own value range, usually with a sign split so negatives take one hue and positives another, and the crossing between them becomes a visible boundary. The scale is relative to this table's minimum and maximum, so the same numeric value can sit at opposite ends of the palette in two different grids.

Good uses

  • Set a price sheet: sweep selling price against unit volume and see which combinations turn gross profit negative before the price list goes out.
  • Compare loan options: grid interest rate against term to see how much a longer term cuts the monthly payment and what it adds in total interest.
  • Stress a sales forecast: pair units sold with average selling price so the plan shows which two factors must both hold to stay above breakeven.

Limits and checks

  • Additive formulas show no interaction: if the output is A + B, every row is the same shape shifted, and the grid adds nothing over two one-variable sweeps.
  • The palette is relative: the deepest hue always marks this table's own maximum, so an all-profitable grid still shows strong color contrast. Judge the numbers, not the colors.
  • A coarse step hides the crossing: if break-even falls between two grid values, no cell shows it. Refine the range when the boundary between colors matters.

Common questions

Can I get more than one output, like revenue and profit, from a single table?

Usually one formula per table, so build a separate grid for each output you need. If you need several at once, check whether the tool accepts multiple formulas; if not, compare the grids side by side, since their patterns rarely agree.

Why is my darkest cell not the one I expected?

Because the colors rank cells inside this table only. The strongest color marks the largest computed value, whatever it is; if every combination is a loss, the deepest color marks the smallest loss. Set your ranges around the realistic region and confirm the winning cell is a combination you would actually choose.

References and verification

The example and behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.

Related Tools