Abstract
A problem description usually says what's wrong. It rarely says what's fine, even though that comparison is where the cause tends to show up. Is / Is Not analysis puts the two side by side: where the problem is, and where you'd expect it but it isn't.
If invoices from one office have the wrong tax rate and invoices from the office next door are correct, the cause is probably something that differs between the two. Look for what changed around that difference. A candidate cause only counts if it explains every IS and every IS NOT.
Method
- 01
State the deviation
Name the object and what's wrong with it in one line. For example: invoices from the new system show the wrong tax rate.
- 02
Fill in IS and IS NOT
For what, where, when and how much, describe where the problem is and where it could reasonably be but isn't.
- 03
Find the distinctions
For each pair, ask what's different or unusual about the IS side.
- 04
Look for changes
Ask what changed in or around each distinction, and when.
- 05
Test the possible causes
Check each candidate cause against every IS and IS NOT. Keep the one that explains them all.
- 06
Verify before you fix
Confirm the cause with a test or an observation before committing to a fix.
When to use it
- A problem started at a particular time and someone needs to know why.
- There are several plausible causes and people feel strongly about them.
- A fix worked for a while and then the problem came back.
Common pitfalls
- A vague IS NOT column. The comparison only works if the IS NOT is a close, realistic neighbour.
- Jumping to causes before you've filled in the table.
- Accepting a cause that explains most of the facts and ignoring the ones it doesn't.
- Skipping verification because the answer feels obvious.
Origin and references
Attributed to Charles Kepner and Benjamin Tregoe (1965).
- Kepner, C. H., & Tregoe, B. B. (1965). The Rational Manager. McGraw-Hill.
- Kepner, C. H., & Tregoe, B. B. (1981). The New Rational Manager. Princeton Research Press.