# Retain failed attempts after a retry passes

Illustrative planning brief; no automatic product import.

Synthetic planning case: A supposedly repeatable external check runs three times on the same declared artifact and setup: fail, fail, pass. A report keeps only the final pass.

## Decision

Retain two failures and one pass with the artifact/setup context. The last result does not establish stable success or justify erasing contradictory attempts.

## Owned work

- Preserve all attempts
  - Owner role: Test recorder
  - Acceptance evidence: The two failures remain accessible with the third pass.
- Check the sameness claim
  - Owner role: Investigator
  - Acceptance evidence: Artifact, setup and unrecorded conditions are examined before calling inputs identical.
- Resolve repeatability evidence
  - Owner role: Acceptance reviewer
  - Acceptance evidence: Any waiver or investigation is explicit; a final pass alone does not satisfy a repeatability requirement.

## Workflow

1. Preserve all attempts. Check: The two failures remain accessible with the third pass.
2. Check the sameness claim. Check: Artifact, setup and unrecorded conditions are examined before calling inputs identical.
3. Resolve repeatability evidence. Check: Any waiver or investigation is explicit; a final pass alone does not satisfy a repeatability requirement.

## Judgment

Three invented attempts cannot establish a calibrated failure rate. The claimed sameness of conditions requires independent evidence.


## Filled manual planning note

Retain two failures and one pass with the artifact/setup context. The last result does not establish stable success or justify erasing contradictory attempts. The recorded attempts include 2/3 failures and 1/3 passes. Those descriptive counts do not estimate a population flake probability or explain the cause. Three invented attempts cannot establish a calibrated failure rate. The claimed sameness of conditions requires independent evidence.


## Working artifact

- Two failed attempts / Recorded before the final pass / Retain fixtures and context
- Same-input claim / Declared, not fully verified / Inspect artifacts and setup changes
- Final pass / One observed result / Review repeatability criteria before closure


## Workflow questions

### Does the log prove nondeterministic software?

No. Unrecorded setup changes, timing or evidence errors may explain the disagreement.

### Should every intermittent failure block everything?

Apply the actual acceptance policy; this worksheet preserves evidence and does not supply a universal release rule.

## Product connection

Use the owned checks and downloaded brief to discuss this planning decision alongside your TeamBoostAI tasks. Confirm available fields, roles and account features separately. The example is manual; it does not calculate live analytics, create work or run an experiment in the product.

Confirm account availability before adopting this manual outline.

## Original worked case

Synthetic records, manual planning only. No account import or live analytics.

### Inspect the invented case records

Attempt | Declared artifact/setup | Observed result
--- | --- | ---
1 | Same declared inputs | Fail
2 | Same declared inputs | Fail
3 | Same declared inputs | Pass

### Reasoning

The recorded attempts include 2/3 failures and 1/3 passes. Those descriptive counts do not estimate a population flake probability or explain the cause.

### Bounded result

Retain two failures and one pass with the artifact/setup context. The last result does not establish stable success or justify erasing contradictory attempts.

### Distinct decision

The guide preserves attempt instability, distinct from counting multiple reviews per accepted proof.

### Limits

Three invented attempts cannot establish a calibrated failure rate. The claimed sameness of conditions requires independent evidence.

### Definitions and method context

- Original worked-case definitions — Definitions, policy choices, records and calculations are authored for this worksheet. No external standard, statistical validation or live product measurement is claimed.