In short: The most valuable thing an organisation knows is what it tried and abandoned, and why. It is also the first thing lost when people move. A judgment system should make the graveyard first-class, and distinguish what was disproven from what was merely under-tried.
Key points
- Classify each abandoned approach as disproven, under-tried, overtaken or political, because each has different implications for revival.
- Many ideas organisations believe were disproven were actually under-tried.
- Each record should state what would make the idea worth trying again, turning a tombstone into a reopenable option.
- Match new proposals against graveyard records on the situation, not the wording, and surface the closest match.
- Count repeat failures, which should be zero, and successful revivals.
Every function has a graveyard. The channel that did not work. The pricing model withdrawn after a quarter. The partnership that looked perfect and was quietly ended. The analytics vendor dropped after the pilot. The reasons live in the heads of whoever was there.
When those people move on, a new owner arrives with fresh eyes and, very often, revives one of the abandoned ideas. Sometimes because a vendor pitched it, sometimes because a board member suggested it, sometimes because it simply looks obvious. Occasionally conditions have changed and it works. Often it fails again, for the same reason, at full cost, and nobody connects the two failures.
Not every death is the same
The most useful thing a graveyard record can do is say why something died, because the reasons have very different implications for revival.
Disproven: tried properly, with resources, and it did not work. Revive only with new evidence that the conditions have changed.
Under-tried: abandoned for lack of time, budget or attention, before it was really tested. These may be good ideas that were never given a fair chance.
Overtaken: worked, but was replaced by something better or made redundant by events.
Political: stopped because of who sponsored it or who opposed it, not because of its merits.
Many ideas that organisations believe were disproven were actually under-tried. That distinction alone can save a promising idea from permanent burial, and save the organisation from repeating a genuine failure.
The record
For each abandoned approach: what was tried, for how long, with what resources; what happened; why it was stopped, classified as above; and what would make it worth trying again. That last field turns a tombstone into a reopenable option with a stated condition.
Surfacing it at the right moment
A graveyard nobody consults is just storage. Match new proposals against graveyard records on the situation, not the wording, and surface the match when the proposal is made: "A similar approach was tried in 2025 and stopped after two quarters; classified as under-tried; revival condition: dedicated owner." That one message is often worth more than the rest of the system combined.
What to count
Repeat failures: an abandoned approach tried again without new evidence, and failing for the same reason. It should be zero.
Successful revivals: graveyard items reopened because their revival condition was met, that then worked. These prove the graveyard is not just a list of reasons to say no.
In practice
- Backfill the graveyard for one function: the ten most significant abandoned initiatives of the last three years.
- Classify each as disproven, under-tried, overtaken or political, with a revival condition.
- Match new proposals against it automatically and surface the closest record.
- Report repeat failures and successful revivals once a year.
What would prove this wrong. Organisations with a maintained graveyard showing no fewer repeat failures, and no more successful revivals, than those without. Then the past is less informative than this piece assumes.
Common questions
How do companies avoid repeating failed initiatives?
Keep a graveyard record of each abandoned approach: what was tried, for how long and with what resources, what happened, why it stopped, and what would make it worth trying again. Match new proposals against it and surface the closest record when the proposal is made.
What is the difference between a disproven idea and an under-tried one?
A disproven idea was tried properly, with resources, and did not work; revive it only with new evidence that conditions changed. An under-tried idea was abandoned for lack of time, budget or attention before it was really tested.
How do you start capturing institutional knowledge about abandoned projects?
Backfill the graveyard for one function with the ten most significant abandoned initiatives of the last three years, classify each, and give each a revival condition.