> "Most of how we actually work is unwritten — and we cannot tell how much it is costing us."
^pain
A collective-track pain where the dominant operating mode is tribal knowledge: unwritten practice known inside a group, opaque to outsiders. Onboarding is slow, cross-team collaboration is brittle, and the cost is invisible because the gap is never inventoried — missing things are not listed as missing. Distinct from standards-in-heads: tribal knowledge is *group-held* practice, while standards-in-heads is *individual-held* expertise. Both fail when the boundary of the holding group is crossed.
## Discovery questions
- "What does your team do differently from any documented standard — and where does that knowledge live?"
- "How would you onboard someone if the entire team that built this product went on sabbatical?"
- "What are the things 'everyone knows' that aren't actually written anywhere?"
^discovery-questions
## Examples
- NASA OIG: loss of Saturn V and Apollo-era engineers' tribal knowledge made aspects of modern heavy-lift development harder because many design decisions were never fully documented.[^1]
- NTSB Aircraft Safety Report on 737 MAX: reliance on veteran engineers' informal understanding (rather than fully documented system behaviour) contributed to MCAS knowledge gaps.[^2]
[^1]: https://oig.nasa.gov/docs/IG-18-019.pdf
[^2]: https://www.ntsb.gov/investigations/AccidentReports/Reports/ASR1901.pdf