Rejection ledger
A record kept beside the active AI portfolio of serious ideas that did not proceed, with the binding constraint, the evidence date and what would have to change to reopen them.
A rejection ledger sits beside the active AI portfolio. For every serious opportunity that does not proceed, it records the main constraint that made the decision fail at that time, and the date of the evidence behind it.
Most organizations track what they build and forget what they declined. That discards some of their most valuable information, because AI capability changes quickly and many earlier rejections are no longer true. A ledger tells you what would have to become true for each "no" to turn into a "yes".
What goes in it
- The idea and who raised it
- The binding constraint that blocked it
- The evidence date — the date belongs next to the no
- The condition for reopening
Keep the no, not a score. When a model release or price change lands, review only the entries whose recorded constraint actually moved, rather than rescoring the whole backlog or treating every release as a new strategy cycle. Some rejections even become assets, such as decisions about what the organization will not automate.
Read more in Keep the AI ideas you rejected.
Related terms
Binding constraint
The main reason an AI idea was rejected at a given time, recorded so the decision can be reopened only when that specific constraint actually changes.
AI pilot
A limited trial of an AI use case to test whether it works in practice. A pilot proves one team can succeed; capability starts when others can reproduce the result.