AIIMPACT-001™

Written By Samantha Avril-Andreassen

The SAFECHAIN™ Accountability Integrity Impact & Effectiveness Framework™

Establishing the Governance Standard for Measuring Whether Accountability Produces Meaningful, Demonstrable and Sustainable Institutional Change Across AI1™–AI5™

Framework Reference: AIIMPACT-001™
Framework Type: Impact, Effectiveness, Outcome Measurement & Longitudinal Accountability Framework
Parent Framework: ACCOUNTABILITY-001™ — The SAFECHAIN™ Governance Answerability, Consequence & Institutional Accountability Framework™
Classification Architecture: AI1™–AI5™
Framework Series: SAFECHAIN™ Accountability Integrity Series
Version: 1.0
Year: 2026

1. Framework Purpose

The SAFECHAIN™ Accountability Integrity Impact & Effectiveness Framework™ (AIIMPACT-001™) establishes the governance methodology for determining whether accountability activity has produced meaningful, measurable and sustainable change.

It addresses the final question that every accountability system should be capable of answering:

What difference did accountability actually make?

Institutions frequently measure accountability through activity.

They count:

  • Investigations completed;

  • recommendations accepted;

  • policies revised;

  • training sessions delivered;

  • complaints processed;

  • audits undertaken;

  • actions closed;

  • reports published.

Those measures may demonstrate activity.

They do not necessarily demonstrate impact.

An institution may complete every recommendation while the underlying behaviour remains unchanged.

It may introduce new safeguarding procedures while people remain exposed to the same risks.

It may deliver extensive training while decision-making remains poor.

It may close hundreds of corrective actions while the same failures continue to recur.

It may provide a remedy without improving the affected person's position.

It may create new governance structures without changing who actually exercises power.

AIIMPACT-001™ therefore moves accountability evaluation beyond:

What did the institution do?

to:

What changed because the institution did it?

The framework establishes the sequence:

Baseline → Intention → Intervention → Outcome → Impact → Verification → Comparison → Learning → Sustainability

2. Central Question

After the investigations, findings, recommendations, consequences, remedies and reforms, can the institution demonstrate that accountability materially improved how it behaves, governs, protects people and responds to failure?

3. Governing Principle

Accountability should ultimately be judged by the measurable and sustainable difference it produces in institutional behaviour, governance integrity, safeguarding, affected-person outcomes and recurrence risk—not merely by the volume of accountability activity undertaken.

4. Accountability Impact™

For AIIMPACT-001™, Accountability Impact™ means:

The demonstrable difference attributable, wholly or materially, to accountability intervention in institutional behaviour, governance capability, safeguarding, decision quality, affected-person outcomes, recurrence risk and institutional learning.

Impact should be distinguished from:

Activity — what the institution did.

Output — what the activity produced.

Outcome — what changed following the intervention.

Impact — whether that change materially improved accountability conditions.

5. SAFECHAIN™ Accountability Impact Architecture™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Accountability Impact Architecture™

The architecture comprises nine stages.

AIA1 — Baseline

Establish the accountability condition before intervention.

AIA2 — Intended Impact

Define what the intervention is expected to change.

AIA3 — Intervention

Identify what accountability action occurred.

AIA4 — Outcome Measurement

Measure what changed.

AIA5 — Impact Assessment

Determine whether the change is meaningful.

AIA6 — Attribution

Assess whether the accountability intervention materially contributed to the change.

AIA7 — Sustainability

Determine whether improvement persisted.

AIA8 — Learning

Identify what worked, what failed and why.

AIA9 — Reassessment

Use impact findings to inform future AI1™–AI5™ classification and governance action.

6. SAFECHAIN™ Accountability Impact Chain™

Material accountability interventions should be traceable through:

Failure → Finding → Intervention → Intended Impact → Indicator → Outcome → Impact → Sustainability → Learning

A break in this chain may prevent the institution from demonstrating accountability effectiveness.

7. Accountability Effectiveness Baseline™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Accountability Effectiveness Baseline™

Before significant intervention, the institution should establish the best available baseline against which subsequent change can be assessed.

The baseline may include:

  • Failure frequency;

  • complaint patterns;

  • safeguarding incidents;

  • recurrence rates;

  • decision quality;

  • control effectiveness;

  • escalation performance;

  • challenge experience;

  • affected-person outcomes;

  • assurance findings;

  • accountability classification;

  • behavioural indicators;

  • leadership response.

8. Baseline Integrity

A baseline should be:

  • Relevant;

  • evidence-based;

  • proportionate;

  • sufficiently documented;

  • capable of later comparison.

Where historical data is incomplete, limitations should be recorded rather than concealed.

9. SAFECHAIN™ Baseline Integrity Principle™

An institution cannot credibly demonstrate improvement if it has no reliable understanding of the condition from which improvement is being claimed.

10. Retrospective Baselines

Where no formal baseline existed before intervention, institutions may construct a retrospective baseline using reliable historical evidence.

The methodology and limitations should be documented.

11. Intended Impact Map™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Intended Impact Map™

Every material accountability intervention should identify the change it is intended to produce.

12. Intended Impact Categories

The Intended Impact Map™ may include:

Behavioural Impact

What behaviour should change?

Governance Impact

What governance weakness should improve?

Safeguarding Impact

What risk to people should reduce?

Decision Impact

What should improve about institutional decisions?

Accountability Impact

What should improve about answerability, consequence or ownership?

Affected-Person Impact

What should change for people experiencing the consequences of institutional failure?

Recurrence Impact

What failure should become less likely to happen again?

13. SAFECHAIN™ Intended Impact Principle™

An institution should define what success is expected to look like before declaring that success has occurred.

14. Impact Assumptions

The Intended Impact Map™ should identify material assumptions.

For example:

Training will improve knowledge.

Improved knowledge will influence behaviour.

Improved behaviour will reduce the identified failure.

Each assumption should remain capable of testing.

15. Outcome Indicator Set™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Outcome Indicator Set™

Institutions should use indicators capable of measuring whether intended change occurred.

Indicators may be:

Quantitative

and/or

Qualitative.

16. Quantitative Indicators

Examples include:

  • Recurrence rate;

  • complaint rate;

  • escalation time;

  • remediation time;

  • safeguarding incident frequency;

  • action completion;

  • decision reversal rate;

  • control failure rate;

  • response time.

17. Qualitative Indicators

Examples include:

  • Affected-person experience;

  • staff confidence;

  • quality of decision reasoning;

  • leadership behaviour;

  • challenge culture;

  • safeguarding quality;

  • evidence quality;

  • independent review findings.

18. SAFECHAIN™ Indicator Integrity Principle™

Indicators should measure the intended accountability outcome rather than merely what is easiest for the institution to count.

19. Indicator Manipulation Safeguard™

Institutions should guard against selecting indicators primarily because they produce favourable results.

Where indicators change during the assessment period, the reason should be documented.

20. Activity–Impact Separation™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Activity–Impact Separation™

Examples:

Activity: 500 staff trained.

Impact question: Did relevant behaviour improve?

Activity: Policy revised.

Impact question: Did practice change?

Activity: Investigation completed.

Impact question: Did recurrence reduce?

Activity: Remedy issued.

Impact question: Did the affected person's position improve?

21. Behavioural Change Test™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Behavioural Change Test™

The test determines whether accountability intervention changed relevant institutional behaviour.

Ask:

What behaviour required change?

Who needed to change it?

What intervention occurred?

What evidence demonstrates changed behaviour?

Has the change persisted?

22. Behavioural Evidence

Evidence may include:

  • Observed practice;

  • case sampling;

  • decision records;

  • supervision records;

  • complaints;

  • audit findings;

  • safeguarding records;

  • staff feedback;

  • affected-person feedback.

23. SAFECHAIN™ Behaviour-over-Training Principle™

Training completion should not automatically be treated as evidence of behavioural change.

24. Behavioural Regression Alert™

A Behavioural Regression Alert™ should activate where initially improved behaviour subsequently deteriorates.

This should trigger consideration under AIMON-001™ and AIGR-001™.

25. Governance Change Test™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Governance Change Test™

This determines whether accountability intervention materially improved governance architecture.

It should examine:

  • Ownership;

  • authority;

  • escalation;

  • oversight;

  • independence;

  • evidence integrity;

  • challenge;

  • monitoring;

  • assurance;

  • reporting.

26. Governance Change Questions

Ask:

Are responsibilities clearer?

Are decisions more traceable?

Are conflicts better managed?

Can challenge reach independent authority?

Does the board receive better information?

Are failures detected earlier?

Is remediation more reliable?

27. SAFECHAIN™ Structural Change Principle™

Creating a new committee, policy or reporting line should not automatically be treated as governance improvement unless the change materially improves how accountability operates.

28. Safeguarding Impact Test™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Safeguarding Impact Test™

Where accountability concerns safeguarding, the institution should determine whether intervention materially improved protection.

29. Safeguarding Impact Questions

Ask:

Has risk reduced?

Has continuing harm stopped?

Are concerns recognised earlier?

Are escalation routes faster?

Are affected persons safer?

Are repeat safeguarding failures reducing?

Have barriers to reporting been addressed?

30. SAFECHAIN™ Safeguarding Impact Principle™

Safeguarding reform should ultimately be assessed by whether protection improved, not merely by whether safeguarding documentation increased.

31. Safeguarding Impact Override™

Where aggregate indicators suggest improvement but serious safeguarding failure persists, the serious failure should not be concealed by overall positive performance.

32. Affected-Person Impact Test™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Affected-Person Impact Test™

This determines whether accountability produced meaningful change for those directly affected by institutional failure.

33. Affected-Person Impact Areas

Where appropriate, assess:

  • Harm stopped;

  • risk reduced;

  • records corrected;

  • decisions reconsidered;

  • services restored;

  • remedy delivered;

  • participation improved;

  • communication improved;

  • acknowledgement provided;

  • continuing consequences addressed.

34. SAFECHAIN™ Affected-Person Reality Principle™

An institution should not claim successful accountability solely because its internal process improved if the people harmed by the original failure remain materially affected by unresolved consequences.

35. Participation

Affected-person participation should be proportionate, safe and meaningful.

Institutions should avoid treating participation as a requirement for affected persons to repeatedly relive harmful experiences merely to prove institutional improvement.

36. SAFECHAIN™ Participation Safeguard™

Impact evaluation should seek evidence from affected persons where appropriate without transferring the institution's burden of proving improvement onto those who experienced the failure.

37. Remedy Effectiveness

AIRESP-001™ determines what remedy is required.

AIIMPACT-001™ determines whether that remedy produced the intended outcome.

38. Recurrence Reduction Measure™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Recurrence Reduction Measure™

The measure determines whether similar accountability failures have reduced following intervention.

39. Recurrence Comparison

Compare, where possible:

Pre-Intervention Frequency

Post-Intervention Frequency

Severity

Affected Domains

Common Causes

Failed Controls

Time to Detection

Time to Response

40. SAFECHAIN™ Recurrence Impact Principle™

One of the strongest indicators of effective accountability is that the same preventable failure becomes materially less likely to happen again.

41. Zero Recurrence Caution

Absence of recorded recurrence does not automatically prove success.

Consider whether:

  • Reporting declined;

  • detection weakened;

  • records changed;

  • definitions narrowed;

  • failures moved elsewhere.

42. Risk Displacement™

AIIMPACT-001™ establishes the concept of:

SAFECHAIN™ Risk Displacement™

Risk displacement occurs where an intervention appears to solve a problem in one area while moving substantially the same risk into another area, process or population.

43. Risk Displacement Test™

Ask:

Did the risk actually reduce, or did it move?

44. Institutional Learning Effectiveness Test™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Institutional Learning Effectiveness Test™

This determines whether claimed learning changed institutional behaviour or governance.

45. Learning Effectiveness Questions

Ask:

What was learned?

How was it identified?

Who received the learning?

What changed because of it?

Was the change tested?

Did recurrence reduce?

Was learning shared across relevant domains?

46. SAFECHAIN™ Learning Evidence Principle™

"Lessons learned" is not an accountability outcome unless the institution can identify the lesson, the change it produced and evidence that the change affected future practice.

47. Learning Transfer™

Where a failure has implications beyond one department, learning should be considered across comparable:

  • Teams;

  • locations;

  • services;

  • decisions;

  • risk domains.

48. SAFECHAIN™ Learning Transfer Principle™

Institutional learning should travel further than the individual case where the underlying cause is capable of recurring elsewhere.

49. Decision Quality Impact™

Impact assessment should examine whether accountability improved institutional decision-making.

Indicators may include:

  • Better evidence use;

  • clearer reasoning;

  • fewer unsupported assumptions;

  • improved challenge;

  • improved documentation;

  • better conflict management;

  • fewer avoidable reversals.

50. Evidence Integrity Impact™

Institutions should assess whether accountability improved:

  • Evidence preservation;

  • evidence testing;

  • record quality;

  • traceability;

  • contradictory evidence handling;

  • decision-evidence linkage.

51. Independence Impact™

AIIND-001™ findings should inform whether:

  • Conflicts are identified earlier;

  • recusal improves;

  • investigation independence strengthens;

  • external review is used appropriately;

  • self-review reduces.

52. Challenge Impact™

AICHAL-001™ findings should inform whether:

  • People speak up earlier;

  • retaliation reduces;

  • challenge reaches appropriate authority;

  • concerns receive substantive responses;

  • challenge influences decisions.

53. Leadership Impact™

AIIMPACT-001™ should examine whether leadership accountability materially changed.

This may include:

  • Ownership;

  • response to adverse findings;

  • escalation behaviour;

  • consequence consistency;

  • board visibility;

  • willingness to reconsider decisions.

54. SAFECHAIN™ Leadership Impact Principle™

Leadership accountability should be evidenced through decisions and behaviour, not solely through statements of commitment.

55. Implementation Impact™

AIFU-001™ establishes whether actions were implemented.

AIIMPACT-001™ establishes whether those implemented actions produced meaningful change.

56. SAFECHAIN™ Implementation–Impact Boundary™

Implementation answers "Did we do it?" Impact answers "Did doing it make the required difference?"

57. Unintended Consequence Review™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Unintended Consequence Review™

Every material accountability intervention should consider whether it created new risks or adverse effects.

58. Potential Unintended Consequences

These may include:

  • Reduced reporting;

  • defensive decision-making;

  • excessive bureaucracy;

  • risk displacement;

  • exclusion;

  • new safeguarding barriers;

  • inappropriate blame;

  • unintended inequality;

  • suppression of legitimate discretion.

59. SAFECHAIN™ Unintended Consequence Principle™

An intervention should not be judged effective solely because it solved the problem it was designed to address if it created another material accountability failure elsewhere.

60. Counterfactual Consideration™

Where feasible, institutions should consider:

What would likely have happened without the intervention?

This should be proportionate and should not create false claims of scientific causation where evidence cannot support them.

61. Attribution Assessment™

AIIMPACT-001™ distinguishes:

AT1 — Strong Attribution

Evidence strongly supports the intervention as a material cause of improvement.

AT2 — Probable Contribution

The intervention likely contributed materially.

AT3 — Plausible Contribution

Contribution is plausible but uncertain.

AT4 — Weak Attribution

Little evidence connects intervention and outcome.

AT5 — No Demonstrated Attribution

No reliable connection has been established.

62. SAFECHAIN™ Attribution Integrity Principle™

Institutions should distinguish between improvement occurring after an intervention and improvement occurring because of the intervention.

63. Accountability Impact Score™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Accountability Impact Score™

The Accountability Impact Score™ provides a structured impact assessment across core domains.

Suggested domains include:

D1 — Behavioural Change

D2 — Governance Change

D3 — Safeguarding Impact

D4 — Affected-Person Impact

D5 — Recurrence Reduction

D6 — Decision Quality

D7 — Evidence Integrity

D8 — Independence

D9 — Challenge & Speak-Up

D10 — Leadership Accountability

D11 — Implementation Sustainability

D12 — Institutional Learning

64. Domain Scoring

Each domain may be assessed:

0 — No Evidence of Impact

1 — Limited Impact

2 — Partial Impact

3 — Material Positive Impact

4 — Strong and Sustained Impact

The maximum indicative score across twelve domains is 48.

65. Accountability Impact Score™ Bands

AIS1 — 40–48

Strong & Sustainable Accountability Impact

AIS2 — 31–39

Material Positive Impact

AIS3 — 21–30

Partial/Mixed Impact

AIS4 — 11–20

Weak Accountability Impact

AIS5 — 0–10

No Demonstrated or Materially Ineffective Impact

Scoring should never override critical-domain evidence.

66. SAFECHAIN™ Critical Domain Override™

A serious unresolved safeguarding, independence, affected-person or recurrence failure may prevent an institution from relying solely upon a favourable aggregate Accountability Impact Score™.

67. Score Evidence Requirement

Every material score should identify:

Evidence

Baseline

Indicator

Outcome

Attribution

Limitations

Reviewer

68. SAFECHAIN™ Score Integrity Principle™

The Accountability Impact Score™ is an evidence-structuring mechanism, not a substitute for professional judgment, critical-domain analysis or transparent reasoning.

69. Impact Confidence Rating™

AIIMPACT-001™ establishes:

IC1 — High Confidence

Strong and convergent evidence.

IC2 — Moderate Confidence

Adequate evidence with limited uncertainty.

IC3 — Limited Confidence

Material evidence gaps exist.

IC4 — Low Confidence

Evidence is substantially insufficient.

Impact scores should be accompanied by an appropriate confidence assessment.

70. Longitudinal Effectiveness Review™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Longitudinal Effectiveness Review™

Impact should be reassessed over time where the significance of the intervention warrants continued review.

71. Review Horizons

Proportionate review points may include:

  • Immediate;

  • short-term;

  • medium-term;

  • long-term.

The appropriate interval depends upon:

  • Risk;

  • seriousness;

  • safeguarding;

  • expected behavioural change;

  • recurrence cycle;

  • institutional complexity.

72. SAFECHAIN™ Longitudinal Integrity Principle™

An intervention that appears effective immediately after scrutiny should not automatically be assumed to remain effective after institutional attention moves elsewhere.

73. Impact Deterioration Trigger™

A SAFECHAIN™ Impact Deterioration Trigger™ should activate where previously demonstrated improvement materially weakens.

This should inform:

  • AIMON-001™;

  • AIGR-001™;

  • AIREC-001™;

  • AIP-001™.

74. Impact Plateau™

An institution may improve initially and then cease progressing.

A plateau should prompt consideration of whether:

  • Further improvement is reasonably required;

  • structural barriers remain;

  • intervention has reached its practical limit.

75. Impact Reversal™

Impact reversal occurs where institutional conditions return substantially toward the original failure state.

This should trigger governance reassessment.

76. Comparative Analysis

Where appropriate, impact may be compared across:

  • Time periods;

  • departments;

  • services;

  • locations;

  • classifications;

  • intervention types.

Comparison should account for material differences in context.

77. SAFECHAIN™ Comparison Integrity Principle™

Comparative impact analysis should illuminate meaningful differences rather than create artificial league tables from incomparable accountability conditions.

78. Impact and AI1™–AI5™

AIIMPACT-001™ should inform the wider accountability classification.

AI1™ — Effective Accountability

Strong evidence demonstrates sustainable positive impact.

AI2™ — Effective with Improvement

Positive impact exists with identifiable improvement opportunities.

AI3™ — Material Accountability Gap

Impact is partial, inconsistent or insufficient in material domains.

AI4™ — Serious Accountability Failure

Major interventions fail to produce required change.

AI5™ — Systemic Accountability Breakdown

Accountability activity repeatedly fails to produce meaningful institutional correction.

79. Impact Classification™

AIIMPACT-001™ establishes five impact conditions.

IM1 — Strong Sustainable Impact

Material and sustained improvement demonstrated.

IM2 — Positive Impact with Improvement

Meaningful improvement exists with residual weaknesses.

IM3 — Partial/Mixed Impact

Some improvement exists but material gaps remain.

IM4 — Weak/Ineffective Impact

Accountability activity has produced limited meaningful improvement.

IM5 — Failed/Systemic Impact

Repeated accountability intervention fails to produce sustainable institutional change.

80. Impact Classification and Score

The Accountability Impact Score™ informs but does not automatically determine IM1™–IM5™ classification.

Critical-domain findings, confidence and contextual evidence remain relevant.

81. Impact Reporting

AIR-001™ reporting should distinguish:

Activity

Output

Outcome

Impact

Sustainability

Confidence

82. SAFECHAIN™ Impact Reporting Principle™

Institutions should not report activity statistics as evidence of impact without demonstrating the relationship between the activity and the claimed improvement.

83. Executive Impact Dashboard™

Material programmes may be reported through a SAFECHAIN™ Executive Impact Dashboard™ showing:

  • Baseline;

  • intended impact;

  • current indicators;

  • Accountability Impact Score™;

  • impact classification;

  • confidence;

  • recurrence;

  • safeguarding;

  • affected-person outcomes;

  • deterioration alerts.

84. Board Visibility

AIO-001™ should provide governing-body visibility of:

  • IM4™/IM5™ conditions;

  • serious safeguarding impact failure;

  • persistent recurrence;

  • failed affected-person outcomes;

  • impact deterioration;

  • ineffective major remediation.

85. Board Impact Question™

Boards should ask:

What evidence shows that the actions we approved changed the conditions that produced the failure?

86. Impact Assurance

AIA-001™ should provide proportionate independent assurance over significant impact claims.

Assurance should test:

  • Baseline reliability;

  • indicator quality;

  • evidence;

  • attribution;

  • scoring;

  • critical-domain overrides;

  • sustainability.

87. Independence

AIIND-001™ should determine appropriate independence for significant impact assessment.

Those responsible for an intervention should not necessarily be the sole judges of its success.

88. Challenge

AICHAL-001™ should permit challenge to:

  • Impact claims;

  • favourable indicators;

  • excluded evidence;

  • attribution assumptions;

  • premature declarations of success.

89. Monitoring

AIMON-001™ should monitor indicators capable of identifying impact deterioration or recurrence.

90. Governance Review

AIGR-001™ should incorporate impact evidence when determining whether accountability classification remains current.

91. Closure

AICL-001™ should consider whether required impact evidence exists before final closure where the nature of the matter requires demonstrated outcomes.

92. Restoration

AIP-001™ restoration should not rely solely upon completed remediation.

Where appropriate, it should consider whether remediation has produced demonstrable impact.

93. Recurrence

AIREC-001™ recurrence evidence should be a significant component of impact assessment.

94. Root Cause

AIROOT-001™ should be revisited where interventions repeatedly fail to produce intended impact.

The failure may indicate that the original root cause analysis was incomplete.

95. Consequence

AICONS-001™ should be reviewed where weak accountability impact results from persistent consequence avoidance.

96. Remedy

AIRESP-001™ should inform affected-person impact and remedy effectiveness.

97. Ownership

AIOWN-001™ should identify responsibility for:

  • Impact measurement;

  • data quality;

  • reporting;

  • escalation;

  • reassessment.

98. Follow-Up

AIFU-001™ determines whether commitments were implemented.

AIIMPACT-001™ determines whether they worked.

Together:

Implementation without impact may demonstrate activity without effectiveness.

99. SAFECHAIN™ Accountability Effectiveness Equation™

AIIMPACT-001™ establishes the conceptual:

SAFECHAIN™ Accountability Effectiveness Equation™

Accountability Effectiveness = Corrective Action + Demonstrable Outcome + Reduced Risk + Sustainable Change + Institutional Learning

No mathematical precision is implied.

The equation expresses the governance components required for meaningful accountability effectiveness.

100. Evidence Hierarchy

Impact claims should draw upon multiple evidence sources where appropriate.

Higher-confidence assessment may combine:

  • Operational data;

  • case sampling;

  • audit evidence;

  • assurance;

  • recurrence data;

  • safeguarding evidence;

  • affected-person evidence;

  • independent review.

101. SAFECHAIN™ Evidence Convergence Principle™

Confidence in accountability impact increases where different reliable evidence sources independently point toward the same conclusion.

102. Negative Evidence

Institutions should actively consider evidence inconsistent with claimed improvement.

103. SAFECHAIN™ Impact Challenge Principle™

An impact assessment should seek evidence capable of disproving success as well as evidence capable of confirming it.

104. Data Quality

Weak data should reduce confidence rather than be silently treated as reliable.

105. Missing Data

Missing data should not automatically be interpreted favourably.

Institutions should determine:

  • Why data is missing;

  • whether absence is material;

  • whether alternative evidence exists;

  • how confidence is affected.

106. Impact Integrity Alert™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Impact Integrity Alert™

This should activate where:

  • Activity is misrepresented as impact;

  • indicators are selectively reported;

  • adverse evidence is excluded;

  • baselines are manipulated;

  • critical failures are hidden by averages;

  • attribution is overstated;

  • sustainability is assumed without evidence.

107. Impact Misrepresentation

Materially misleading impact reporting should itself be capable of accountability review.

108. SAFECHAIN™ Impact Transparency Principle™

Institutions should be as transparent about where accountability failed to produce change as they are about where improvement succeeded.

109. Continuous Improvement

Impact evaluation should feed future:

  • Framework design;

  • governance;

  • training;

  • safeguarding;

  • assurance;

  • leadership development;

  • policy;

  • remediation.

110. Accountability Impact Learning Loop™

AIIMPACT-001™ establishes the:

SAFECHAIN™ Accountability Impact Learning Loop™

Measure → Compare → Challenge → Learn → Adapt → Reassess

111. AIIMPACT-001™ Impact & Effectiveness Integrity Test™

Before an institution can demonstrate accountability impact and effectiveness, ask:

1. Does the Accountability Impact Architecture™ operate?

2. Has an Accountability Effectiveness Baseline™ been established?

3. Is the baseline evidence-based?

4. Are baseline limitations recorded?

5. Has retrospective baseline construction been documented where necessary?

6. Has an Intended Impact Map™ been established?

7. Is intended behavioural change defined?

8. Is intended governance change defined?

9. Is intended safeguarding change defined?

10. Is intended affected-person impact defined?

11. Is intended recurrence reduction defined?

12. Are material assumptions identified?

13. Does an Outcome Indicator Set™ exist?

14. Are quantitative indicators appropriate?

15. Are qualitative indicators appropriate?

16. Do indicators measure outcomes rather than merely activity?

17. Have indicator changes been documented?

18. Has Activity–Impact Separation™ been applied?

19. Is training distinguished from behavioural impact?

20. Is policy change distinguished from governance impact?

21. Is investigation completion distinguished from recurrence reduction?

22. Is remedy delivery distinguished from remedy effectiveness?

23. Has the Behavioural Change Test™ been applied?

24. Is the behaviour requiring change identified?

25. Is behavioural evidence available?

26. Has behavioural change persisted?

27. Has a Behavioural Regression Alert™ arisen?

28. Has the Governance Change Test™ been applied?

29. Has accountability ownership improved?

30. Has decision traceability improved?

31. Has conflict management improved?

32. Has challenge improved?

33. Has board visibility improved?

34. Has early detection improved?

35. Has the Safeguarding Impact Test™ been applied where relevant?

36. Has safeguarding risk reduced?

37. Has continuing harm stopped?

38. Has safeguarding escalation improved?

39. Are affected persons safer?

40. Has serious safeguarding failure been prevented from being hidden by aggregate results?

41. Has the Affected-Person Impact Test™ been applied?

42. Has harm been reduced?

43. Have records been corrected where required?

44. Have decisions been reconsidered where required?

45. Has remedy been delivered?

46. Has continuing impact been addressed?

47. Is affected-person participation proportionate?

48. Has the institution retained responsibility for proving improvement?

49. Has remedy effectiveness been assessed?

50. Has the Recurrence Reduction Measure™ been applied?

51. Has pre-intervention recurrence been compared with post-intervention recurrence?

52. Has recurrence severity changed?

53. Has time to detection improved?

54. Has time to response improved?

55. Has absence of reported recurrence been tested rather than automatically accepted?

56. Has Risk Displacement™ been considered?

57. Did risk genuinely reduce rather than move?

58. Has the Institutional Learning Effectiveness Test™ been applied?

59. Is the claimed lesson identifiable?

60. Did learning produce a change?

61. Was the change tested?

62. Was learning transferred across relevant domains?

63. Has decision quality improved?

64. Has evidence integrity improved?

65. Has independence improved?

66. Has challenge and speak-up effectiveness improved?

67. Has leadership accountability improved?

68. Has implementation produced meaningful impact?

69. Has the Implementation–Impact Boundary™ been respected?

70. Has an Unintended Consequence Review™ been undertaken?

71. Did the intervention create new risks?

72. Did bureaucracy increase without corresponding benefit?

73. Did reporting decrease because people became less willing to report?

74. Did the intervention create inappropriate blame?

75. Has counterfactual consideration been undertaken where appropriate?

76. Has attribution been assessed?

77. Has attribution been classified AT1–AT5 where appropriate?

78. Has correlation been distinguished from causation?

79. Has an Accountability Impact Score™ been calculated where appropriate?

80. Have all relevant domains been assessed?

81. Is every score supported by evidence?

82. Has the Critical Domain Override™ been considered?

83. Has safeguarding been prevented from disappearing within aggregate scoring?

84. Has affected-person failure been prevented from disappearing within aggregate scoring?

85. Has independence failure been prevented from disappearing within aggregate scoring?

86. Has recurrence failure been prevented from disappearing within aggregate scoring?

87. Has an Impact Confidence Rating™ been assigned?

88. Does confidence reflect evidence quality?

89. Has the Longitudinal Effectiveness Review™ been undertaken where appropriate?

90. Has short-term impact been assessed?

91. Has medium-term impact been assessed where required?

92. Has long-term impact been assessed where required?

93. Has an Impact Deterioration Trigger™ arisen?

94. Has impact plateaued?

95. Has impact reversed?

96. Has comparative analysis been appropriately contextualised?

97. Has impact informed AI1™–AI5™ classification?

98. Has impact been classified IM1–IM5 where appropriate?

99. Has the Accountability Impact Score™ been prevented from automatically determining classification?

100. Does AIR-001™ distinguish activity, output, outcome and impact?

101. Does executive reporting show impact rather than activity alone?

102. Does the governing body receive IM4™ and IM5™ conditions?

103. Does the board receive serious safeguarding impact failures?

104. Does the board receive recurrence evidence?

105. Can the board answer what changed because of its interventions?

106. Has AIA-001™ provided appropriate assurance over material impact claims?

107. Has baseline reliability been assured?

108. Has indicator quality been assured?

109. Has attribution been challenged?

110. Has AIIND-001™ protected impact-assessment independence?

111. Can AICHAL-001™ challenge favourable impact claims?

112. Does AIMON-001™ monitor deterioration?

113. Does AIGR-001™ incorporate impact evidence?

114. Does AICL-001™ consider impact before closure where appropriate?

115. Does AIP-001™ consider demonstrated impact before restoration where required?

116. Does AIREC-001™ provide recurrence evidence?

117. Is AIROOT-001™ reconsidered when interventions repeatedly fail?

118. Is AICONS-001™ considered where consequence avoidance undermines impact?

119. Does AIRESP-001™ inform affected-person impact?

120. Does AIOWN-001™ assign impact-measurement responsibility?

121. Does AIFU-001™ provide reliable implementation evidence?

122. Has the Accountability Effectiveness Equation™ been considered?

123. Do multiple evidence sources converge?

124. Has negative evidence been actively considered?

125. Has evidence capable of disproving success been examined?

126. Has poor data quality reduced confidence appropriately?

127. Has missing data been assessed?

128. Has an Impact Integrity Alert™ arisen?

129. Has activity been misrepresented as impact?

130. Have indicators been selectively reported?

131. Has adverse evidence been excluded?

132. Has the baseline been manipulated?

133. Have averages concealed critical failure?

134. Has attribution been overstated?

135. Has sustainability been assumed without evidence?

136. Has materially misleading impact reporting been escalated?

137. Is the institution transparent about ineffective interventions?

138. Does impact evidence feed continuous improvement?

139. Does the Accountability Impact Learning Loop™ operate?

140. Can the institution demonstrate what conditions existed before intervention?

141. Can it demonstrate what it intended to change?

142. Can it demonstrate what intervention occurred?

143. Can it demonstrate what outcome followed?

144. Can it demonstrate whether the intervention materially contributed to that outcome?

145. Can it demonstrate whether the improvement persisted?

146. Can it demonstrate whether recurrence reduced?

147. Can it demonstrate whether affected persons experienced meaningful improvement where relevant?

148. Can it demonstrate whether safeguarding materially improved?

149. Can it demonstrate whether institutional behaviour changed?

150. Can it demonstrate whether governance changed in substance rather than form?

151. Can it demonstrate whether leadership behaviour changed?

152. Can it demonstrate that lessons learned became lessons applied?

153. Can it identify where accountability intervention failed to work?

154. Can it explain why?

155. Can it demonstrate that failed interventions were reconsidered rather than simply re-labelled successful?

156. Can an independent reviewer reproduce the Accountability Impact Score™ from the evidence?

157. Can an independent reviewer understand the limitations and confidence level?

158. Can an independent reviewer identify contradictory evidence?

159. Can the institution demonstrate that impact survived after immediate scrutiny ended?

160. Ultimately, can the institution answer:

What became demonstrably better because accountability occurred?

If yes, the institution has passed the:

SAFECHAIN™ AIIMPACT-001 Impact & Effectiveness Integrity Test™

112. Framework Outcomes

Implementation of AIIMPACT-001™ is intended to provide:

✓ Accountability Impact Architecture
✓ Accountability Effectiveness Baseline
✓ Intended Impact Mapping
✓ Outcome Indicator Sets
✓ Activity–Impact Separation
✓ Behavioural Change Testing
✓ Behavioural Regression Alerts
✓ Governance Change Testing
✓ Safeguarding Impact Testing
✓ Affected-Person Impact Testing
✓ Participation Safeguards
✓ Recurrence Reduction Measurement
✓ Risk Displacement Detection
✓ Institutional Learning Effectiveness Testing
✓ Learning Transfer
✓ Decision Quality Impact Assessment
✓ Evidence Integrity Impact Assessment
✓ Independence Impact Assessment
✓ Challenge Impact Assessment
✓ Leadership Impact Assessment
✓ Implementation–Impact Separation
✓ Unintended Consequence Reviews
✓ AT1™–AT5™ Attribution Assessment
✓ Accountability Impact Score™
✓ AIS1™–AIS5™ Score Bands
✓ Critical Domain Override
✓ IC1™–IC4™ Impact Confidence Ratings
✓ Longitudinal Effectiveness Reviews
✓ Impact Deterioration Triggers
✓ Impact Plateau and Reversal Detection
✓ IM1™–IM5™ Impact Classification
✓ Impact Reporting
✓ Executive Impact Dashboard
✓ Accountability Effectiveness Equation™
✓ Evidence Convergence
✓ Impact Integrity Alerts
✓ Accountability Impact Learning Loop
✓ AI1™–AI5™ integration

113. Governing Statement

Accountability is often measured at the point where institutional work appears to finish.

The investigation closes.

The recommendation is accepted.

The action plan is completed.

The policy changes.

The training occurs.

The board receives the final report.

The matter is marked closed.

AIIMPACT-001™ asks what happens after that point.

Did behaviour actually change?

Did safeguarding improve?

Did the affected person experience meaningful redress?

Did decision-making become better?

Did institutional power become more accountable?

Did people become safer when they challenged failure?

Did conflicts become easier to identify?

Did evidence become more reliable?

Did leaders behave differently?

Did the same failure happen again?

And, months or years later:

Did the change last?

This distinction is fundamental.

An institution can become exceptionally efficient at processing accountability without becoming more accountable.

It can investigate quickly.

It can produce sophisticated reports.

It can accept recommendations.

It can maintain dashboards.

It can close actions.

It can publish improvement plans.

And yet the conditions that created the original failure can remain substantially unchanged.

AIIMPACT-001™ therefore establishes:

Baseline → Intention → Intervention → Outcome → Impact → Verification → Comparison → Learning → Sustainability

The framework requires institutions to distinguish:

Activity from outcome.

Output from impact.

Implementation from effectiveness.

Short-term compliance from sustainable change.

Institutional assurance from affected-person reality.

Recorded learning from applied learning.

Absence of reports from absence of failure.

Correlation from attributable improvement.

And ultimately:

Accountability performed from accountability achieved.

The ultimate success of an accountability architecture is not the quantity of accountability work it generates.

It is whether the institution becomes demonstrably better at preventing failure, detecting failure, responding to failure, protecting people, correcting harm and learning from what went wrong.

The principle at the centre of AIIMPACT-001™ is therefore:

Accountability has not demonstrated its effectiveness merely because the institution completed the process. Its effectiveness is demonstrated when the institution can show that something materially changed—and that the change endured.

Copyright and Intellectual Property Notice

© 2026 Samantha Avril-Andreassen. All Rights Reserved.

AIIMPACT-001™ — The SAFECHAIN™ Accountability Integrity Impact & Effectiveness Framework™ is an original governance impact, accountability-effectiveness, outcome-measurement, behavioural-change, safeguarding-impact, affected-person-impact, recurrence-reduction, institutional-learning and longitudinal-review framework developed and authored by Samantha Avril-Andreassen, LLB (Hons), LLM, LPC, FRSA, Founder of SAFECHAIN™.

AIIMPACT-001™ forms part of the SAFECHAIN™ Accountability Integrity Series and operates within the wider SAFECHAIN™ governance architecture, including ACCOUNTABILITY-001™, AI1™–AI5™, AIP-001™, AIA-001™, AIO-001™, AIR-001™, AIMON-001™, AIGR-001™, AICL-001™, AIREC-001™, AIROOT-001™, AICONS-001™, AIRESP-001™, AIOWN-001™, AICHAL-001™, AIIND-001™ and AIFU-001™.

The original expression, selection, arrangement, architecture, terminology, impact methodology, baseline methodology, intended-impact architecture, outcome-measurement structures, behavioural and governance tests, safeguarding and affected-person impact mechanisms, recurrence-reduction methodology, institutional-learning methodology, attribution structures, impact scoring, confidence assessment, longitudinal-review architecture, classifications, tests, triggers, alerts and associated implementation materials contained within this publication constitute proprietary intellectual property.

This includes, where original to AIIMPACT-001™, the SAFECHAIN™ Accountability Impact Architecture™, AIA1™–AIA9™ Impact Stages, Accountability Impact Chain™, Accountability Effectiveness Baseline™, Baseline Integrity Principle™, Intended Impact Map™, Intended Impact Principle™, Outcome Indicator Set™, Indicator Integrity Principle™, Activity–Impact Separation™, Behavioural Change Test™, Behaviour-over-Training Principle™, Behavioural Regression Alert™, Governance Change Test™, Structural Change Principle™, Safeguarding Impact Test™, Safeguarding Impact Principle™, Safeguarding Impact Override™, Affected-Person Impact Test™, Affected-Person Reality Principle™, Participation Safeguard™, Recurrence Reduction Measure™, Recurrence Impact Principle™, Risk Displacement™, Risk Displacement Test™, Institutional Learning Effectiveness Test™, Learning Evidence Principle™, Learning Transfer Principle™, Leadership Impact Principle™, Implementation–Impact Boundary™, Unintended Consequence Review™, Unintended Consequence Principle™, AT1™–AT5™ Attribution Assessment, Attribution Integrity Principle™, Accountability Impact Score™, AIS1™–AIS5™ Score Bands, Critical Domain Override™, Score Integrity Principle™, IC1™–IC4™ Impact Confidence Ratings, Longitudinal Effectiveness Review™, Longitudinal Integrity Principle™, Impact Deterioration Trigger™, IM1™–IM5™ Impact Conditions, Impact Reporting Principle™, Executive Impact Dashboard™, Accountability Effectiveness Equation™, Evidence Convergence Principle™, Impact Challenge Principle™, Impact Integrity Alert™, Impact Transparency Principle™, Accountability Impact Learning Loop™ and AIIMPACT-001™ Impact & Effectiveness Integrity Test™, together with associated framework materials.

No part of this publication may be reproduced, copied, republished, adapted, translated, distributed, licensed, sublicensed, sold, commercially exploited or incorporated into another governance framework, impact methodology, accountability measurement model, benchmarking system, audit methodology, assurance system, certification scheme, accreditation programme, consultancy methodology, training product, artificial-intelligence system, analytics platform, scoring model, software product or derivative commercial offering without prior written permission from the applicable rights holder, except to the extent otherwise permitted by applicable law.

Publication or public accessibility of AIIMPACT-001™ does not grant authority to issue or represent any SAFECHAIN™ Accountability Impact Score™, AIS1™–AIS5™ score, IM1™–IM5™ impact classification, IC1™–IC4™ confidence rating, AT1™–AT5™ attribution assessment, AI1™–AI5™ classification, assurance opinion, certification, accreditation, governance rating, SAFECHAIN™ Seal or other credential as officially authorised, approved, verified, certified or accredited by SAFECHAIN™.

No unauthorised person or organisation may issue official SAFECHAIN™ impact assessments, scores, classifications, assurance opinions, certificates, seals, credentials or accreditation claims, or represent itself as a SAFECHAIN™ authorised assessor, auditor, verifier, certification body, accreditation body, implementation partner, training provider or assurance authority without express authorisation under applicable SAFECHAIN™ governance and licensing arrangements.

References within AIIMPACT-001™ to generally established concepts including impact assessment, effectiveness, outcomes, baselines, indicators, behavioural change, safeguarding, recurrence, attribution, longitudinal review, evaluation, monitoring and continuous improvement do not constitute claims of exclusive ownership over those underlying concepts.

The proprietary claim relates to the original SAFECHAIN™ expression, selection, arrangement, architecture, terminology, methodologies, classifications, tests, scoring structures, standards, triggers, alerts and framework materials developed by the author.

The Accountability Impact Score™ is a SAFECHAIN™ governance assessment mechanism and should not be represented as a scientifically validated psychometric, actuarial, statistical or regulatory instrument unless and until separately validated for such a purpose. Scores should be interpreted alongside evidence quality, critical-domain findings, professional judgment and the applicable Impact Confidence Rating™.

The SAFECHAIN™ Accountability Effectiveness Equation™ is a conceptual governance expression and is not intended to represent a mathematical, statistical or scientific formula.

The use of the ™ symbol identifies names, concepts, methodologies and framework identifiers being asserted as proprietary brand or framework designations. It does not, by itself, constitute a representation that any particular designation has been registered as a trade mark in any jurisdiction.

Nothing within AIIMPACT-001™ should be interpreted as legal advice, statutory guidance, regulatory approval, governmental accreditation, scientific validation, judicial determination or determination of legal liability.

An AIIMPACT-001™ finding, Accountability Impact Score™, AIS1™–AIS5™ score, IM1™–IM5™ impact condition, IC1™–IC4™ confidence rating, AT1™–AT5™ attribution assessment or related AI1™–AI5™ classification does not, by itself, establish negligence, causation in law, misconduct, breach of statutory duty, regulatory breach, contractual breach, professional misconduct, criminal responsibility or other legal liability.

AIIMPACT-001™ is a governance impact and accountability-effectiveness framework. Its mechanisms should be applied proportionately, independently and consistently with applicable law, regulatory requirements, safeguarding obligations, authorised governance arrangements, ethical research and evaluation principles, data-protection requirements, procedural rights and the evidential circumstances concerned.

Author and Framework Developer:
Samantha Avril-Andreassen, LLB (Hons), LLM, LPC, FRSA
Founder — SAFECHAIN™

Framework: The SAFECHAIN™ Accountability Integrity Impact & Effectiveness Framework™
Framework Reference: AIIMPACT-001™
Parent Framework: ACCOUNTABILITY-001™
Classification Architecture: AI1™–AI5™
Framework Series: SAFECHAIN™ Accountability Integrity Series
Version: 1.0
Year: 2026

© 2026 Samantha Avril-Andreassen. All Rights Reserved.

Samantha Avril-Andreassen https://www.themagicccircle.com
Next

AIFU-001™