A zombie variant isn't dead and isn't alive. It sits in the price book, the configurator and every change impact analysis, costing a little effort everywhere, and almost never gets ordered. The term comes from Julian Weyer, and once you've heard it you start seeing zombies in every portfolio.
How they're made
- An option is added for one project and never removed when the project ends.
- A finish falls out of fashion but nobody owns the decision to stop selling it.
- A replacement part is released but the old variant stays orderable “just in case.”
What they cost
| Where | The hidden cost |
|---|---|
| Configurator | Rules and test configurations that must still pass |
| Pricing | A column in every price book and dealer price file |
| Engineering change | Another row on every where-used and impact analysis |
| Supply | Minimum order quantities for parts that rarely move |
| Sales | One more choice that makes the product harder to explain |
None of these is large on its own. Together, across a few hundred zombies, they are a meaningful share of a configuration team's year.
How to find them
Join order history to the option list and sort by volume. Anything with near-zero orders over two years is a suspect. Then ask who would complain if it disappeared. If the answer is “nobody” or “I'm not sure”, you've found one. The cure is SKU rationalization, done as a proper engineering change. There's a story about one in Nobody ordered the purple one.
Questions people ask
Are all low-volume variants zombies?
No. A low-volume option that wins large projects, or that is legally required in one market, earns its keep. A zombie is low-volume and unowned.
Who should kill a zombie?
Whoever owns the product line, with configuration and engineering change doing the retirement properly so nothing breaks on the way out.
Updated November 2026. Think I've got something wrong? Tell me.