NORM.INV in Excel: Convert Lower-Tail Probabilities to Normal Cutoffs

=NORM.INV(probability, mean, standard_dev)

All three arguments are required. probability lies strictly between zero and one. mean is the arithmetic mean, and standard_dev is positive. The output has the same measurement units as the mean and standard deviation.

NORM.INV inverts the cumulative distribution, not the probability density. It finds x for which NORM.DIST(x,mean,standard_dev,TRUE) equals the input probability, subject to numerical precision.

Mathematically the cutoff is mean + standard_dev*z, where z is the standard normal quantile at the same probability. This relation provides an independent calculation, without asserting how Excel implements its inversion internally.

Put 0.908789 in A2, 40 in A3 and 1.5 in A4. Enter =NORM.INV(A2,A3,A4) in B2.

The independent target is 42.0000020096, consistent with the source’s displayed 42.000002. It is slightly above 42 because the supplied probability is rounded. The more precise normal cumulative probability at 42 is approximately 0.908788780274132.

For a round-trip check, enter =NORM.INV(NORM.DIST(42,40,1.5,TRUE),40,1.5). The mathematical target is 42, but native floating-point agreement was not measured. Do not require literal bit-for-bit equality merely because the functions are mathematical inverses.

For a normal model with mean 100 and standard deviation 15, calculate =NORM.INV(0.025,100,15) and =NORM.INV(0.975,100,15).

Independent cutoffs are 70.60054023 and 129.39945977. Each outer tail contains 2.5%, leaving 95% between the limits. Using 0.05 and 0.95 would instead leave 90% in the central interval.

This interval describes the distribution of individual modeled values with those parameters. It is not automatically a confidence interval for an estimated mean; that requires a different sampling argument and an appropriate standard error.

If an upper tail should contain 5%, use lower-tail probability 0.95: =NORM.INV(1-0.05,100,15). The independent target is 124.6728044.

Using 0.05 directly would return a low cutoff, approximately 75.3271956. Place “upper-tail proportion” beside the input and explicitly show the 1-tail conversion so later editors do not reverse the intended event.

With probability 0.5, the independent result equals the mean, 100. That is a useful central sanity check for any parameter table.

Keep probability fixed at 0.975. Increasing the mean by ten shifts the cutoff upward by ten. Doubling standard deviation from 15 to 30 doubles the cutoff’s distance above the mean, from about 29.39945977 to 58.79891954.

This follows from the linear location-scale relation. It makes a simple audit table possible: vary one parameter at a time and verify the direction of the change. It does not measure uncertainty in estimated parameters or certify that the normal model describes observed data.

Microsoft documents #VALUE! for nonnumeric arguments, #NUM! for probability≤0 or probability≥1, and #NUM! for standard_dev≤0. Exact zero and one correspond to unbounded mathematical normal tails rather than finite cutoffs, so they are not supported probability inputs here.

Enter 97.5% or 0.975, not 97.5. A displayed 100% may conceal an interior value rounded by formatting, while a literal 1 is an excluded endpoint. Inspect the stored value before diagnosing an error.

The reference states that inverse precision depends on NORM.DIST precision. Retain full calculated probability values when chaining functions and avoid overstating extreme-tail accuracy. No native Excel execution, convergence profiling or platform comparison was performed.

• Current documentation: the dedicated Microsoft reference lists Excel for Microsoft 365, Excel for Microsoft 365 for Mac, Excel 2024, Excel 2024 for Mac, Excel 2021, Excel 2021 for Mac, Excel 2019, Excel 2016.
• Introduced: Microsoft’s alphabetical catalog marks NORM.INV as introduced in Excel 2010. Exact first build and platform-specific introduction dates remain unverified.
• Last functioning version: unknown; a current help listing is not a last-working-build test.
• Evidence checked: October 15, 2026. Official documentation was freshly retrieved; examples were independently calculated. No native Excel, Mac or web-client execution was performed.

• Microsoft NORM.INV reference.
• Microsoft function catalog and introduction-marker explanation.
• Editorial testing standards.

Examples use English function names and comma separators. Expected numbers are rounded independent mathematical calculations; documented error expectations are labeled as such. They are not captured Excel output. All illustrative data and policies are fictional.