Manifold Grace

Research & Knowledge · Systems governance · Published 12 November 2026

Systemic Review and Rebuilding the System

An exploratory analytical framework for examining organisational structure, operating rules, procedural integrity, cross-system consequences and preventive redesign. The objective is to understand not only whether a procedure was followed, but whether the procedure, safeguards and remedies were appropriate to the people and circumstances affected.

Scope and evidence status: This is a general framework and research direction, not a determination that any named organisation breached a law or procedure. Application requires the rules operative at the relevant time, reliable records, jurisdiction-specific legal analysis and consideration of competing explanations. Personal case histories and identifying details are not published here. No system implementation or empirical validation is claimed.

Research home · Related exploratory research question RQ-046 · Publication log

Three complementary levels of review

1. Structural integrity

Are governance, authority, responsibilities, controls and service architecture designed appropriately?

2. Procedural integrity

Did a decision follow applicable rules, evidence standards, lawful discretion, accessibility duties and fair procedure?

3. Systemic and preventive integrity

Could emerging failure be detected, corrected and prevented before it causes avoidable harm across systems?

Compliance with an internal procedure does not by itself prove that the procedure is lawful, fair, accessible or effective. Equally, an undesirable outcome does not on its own establish a procedural irregularity.

Three-tier operational architecture

The tiers organise the existing 40 examination areas without renumbering or replacing them. Each tier intersects with the structural, procedural and preventive integrity levels above; a dimension may have more than one analytical function.

TierAreasGroupingsDiagnostic question
Tier 1 — Core organisational and process foundations1–14Governance and authority: 1–4; rules and legal mandates: 6–9; execution and infrastructure: 5, 10–14Who has authority, under which rules, and through what process?
Tier 2 — Cross-cutting controls, capacity and resilience15–19, 25–36Human and cultural dynamics: 15, 25–28; assurance and risk: 16–19, 29; technical and human protection: 30–36What capacities, incentives and safeguards affect real-world execution?
Tier 3 — Accountability, redress and preventive redesign20–24, 37–40Escalation and oversight: 20–22, 37–38; impact and closure: 23–24, 39–40Are failures detected, fairly remedied and prevented from recurring?

Distinguishing failure modes

Execution failure: an applicable, valid and accessible process is not followed because of an operational breakdown, inadequate resources, communication failure or defective control. Design failure: the prescribed process is followed, yet the policy, decision criteria, technology or accessibility arrangement is itself defective or fails an applicable obligation. A case may involve both, neither, or insufficient evidence to distinguish them.

Method: reconstruct the system before judging the event

  1. Define the service, population, decision and period under review.
  2. Map organisational hierarchy, delegated authority and accountable oversight.
  3. Reconstruct the prescribed end-to-end process, including exceptions and adjustments.
  4. Identify policies, contractual terms, human-rights and equality duties, regulations and other legal obligations that actually apply.
  5. Build a dated evidence timeline: records, communications, decision reasons, system events, omitted evidence and uncertainty.
  6. Compare the observed pathway with the applicable rules and a defensible baseline; identify lawful discretion separately from departures or failures.
  7. Assess immediate effects, cumulative administrative burden and cross-system consequences.
  8. Trace potential root causes and test preventive changes, including unintended effects and privacy risks.
  9. Monitor measurable outcomes, accessible remedies and independent assurance over time.

40 areas of examination

The initial 24 areas and 16 additional dimensions below are prompts for evidence collection and system design, not an automatic legal compliance checklist. They can be adapted to higher education, employment, public administration, housing, financial services and other organisations.

A. Structure and mandate

No.Examination areaWhat to establishPreventive systems opportunity
1Organisational structureDepartments, divisions, reporting lines and interdependent service unitsOrganisation map and responsibility ownership
2Governance hierarchyBoards, executives, management, supervisory functions and public accountabilityOversight and escalation map
3Delegated authorityWho can decide, authorise, review, override or grant exceptionsDecision rights and documented approvals
4Operating modelService purpose, delivery model, responsibilities and dependenciesService blueprint and accountability matrix

B. Process and requirements

No.Examination areaWhat to establishPreventive systems opportunity
5Process architectureEnd-to-end stages, dependencies, handoffs and decision pointsProcess flow with measurable handoffs
6Policies and proceduresCurrent and historical operating rules, guidance and exemptionsVersion-controlled rules register
7Legislative and regulatory dutiesApplicable laws, regulatory requirements and jurisdictional boundariesDuty-to-process traceability matrix
8Human rights and equalityApplicable rights, proportionality, reasonable adjustments and accessibilityAccessible-by-design safeguards and legal review
9Eligibility and entitlementsService criteria, entitlements, exclusions and applicable exceptionsClear decision criteria and accessible explanation
10Evidence requirementsSource quality, completeness, provenance and access to evidenceEvidence ledger and verification controls
11Decision criteria and discretionThresholds, discretion, departures from policy and reasonsDecision record with reviewable rationale

C. Operations, information and safeguards

No.Examination areaWhat to establishPreventive systems opportunity
12Communication systemsNotices, deadlines, accessible delivery, receipt and responseDelivery confirmation and alternative channels
13Information architectureRecord ownership, retention, correction, data provenance and permissionsTraceable records and correction history
14Technology infrastructurePortals, integrations, automation, outages and fallback pathsMonitoring, accessible fallback and recovery
15Workforce competenceTraining, supervision, staffing capacity and specialist understandingCapability checks and escalation support
16Risk managementForeseeable hazards, severity, probability and cascading harmRisk register and early-warning thresholds
17Internal controlsSegregation of duties, authorisations, checks and reconciliationsControls testing and exception alerts
18Quality assuranceWhether implementation matches approved design and requirementsTesting, sampling and corrective actions
19Performance managementBacklogs, delays, standards and uneven service outcomesBottleneck analytics with fairness indicators

D. Remedy and rebuilding

No.Examination areaWhat to establishPreventive systems opportunity
20Complaints and appealsAccessible internal review, escalation, independence and remedyClear routes and protected reconsideration
21External oversightRegulators, ombudsmen, independent review and auditOversight responsibility map
22Cross-institutional dependenciesEffects of one decision on other systems or organisationsCross-system impact and correction notifications
23Consequence assessmentFinancial, educational, professional, social and other impactsProportionate consequence and mitigation review
24Corrective and preventive designRoot causes, solution testing and learning from repeated errorsPrevention controls, evaluated redesign and monitoring

Additional dimensions: areas 25–40

These expand the original 24 areas without replacing them. They arise from comparing the framework with governance, assurance, risk, continuity and service-design practice. The areas may overlap operationally, but are kept distinct to avoid leaving important failure mechanisms unexamined.

No.Examination areaWhat to establishPreventive systems opportunity
25Stakeholder and service-user participationWhose experiences inform system design, with consent and safeguards for vulnerable participantsCo-design, accessible feedback and representative user testing
26Organisational culture and incentivesPerformance incentives, informal norms, ethical climate and pressures to conceal errorsIncentive review and protected feedback
27Conflicts of interest and independencePersonal, financial and institutional interests affecting decisions or reviewsConflict declarations and independent allocation
28Budget, capacity and resource allocationStaffing, funding, workload and infrastructure constraints driving delays or unequal accessDemand-capacity modelling and transparent prioritisation
29Supplier, contractor and outsourcing governanceResponsibility for third-party decisions, technology, records and service outcomesContract controls, audits and end-to-end accountability
30User journeys and accessibility testingPractical experience across channels, cognitive requirements, language needs and handoffsEnd-to-end assisted-service tests and inclusive design
31Distributional and cumulative impactWho bears repeated burdens across protected or disadvantaged groups and over timeDisaggregated outcomes and cumulative impact review
32Privacy, consent and data minimisationWhether collection, sharing and retention are lawful, proportionate and necessaryPurpose limitation and privacy-preserving design
33Cybersecurity and information integrityUnauthorised access, tampering, identity errors, integrity loss and incident disclosureSecurity assurance, tamper evidence and recovery controls
34Business continuity and operational resilienceWhether critical services remain accessible during disruption, outages or staff absenceFallback access, recovery tests and continuity exercises
35Change management and policy transitionHow new rules, systems and reorganisations affect ongoing cases, exceptions and recordsChange impact assessment and transition safeguards
36Incident reporting and near-miss learningHow failures, near misses and emerging risks are detected without punitive silencingIncident taxonomies and feedback loops
37Speaking-up and anti-retaliation safeguardsHow employees and service users raise concerns safely and independentlyConfidential escalation and retaliation monitoring
38Emergency powers and exceptional proceduresWhen urgent departures from ordinary pathways are permitted and controlledExpiry, authorisation, review and proportionality checks
39Independent evaluation and benchmarkingHow redesigned processes compare with credible baselines and alternative solutionsControlled pilots, benchmarks and external review
40Remedy implementation and closure verificationWhether agreed corrections reach affected records, services and people in practiceVerified remedy completion and downstream confirmation

Four findings that should not be conflated

FindingAnalytical questionAppropriate next evidence
Procedural departureWas an applicable valid step omitted, misapplied or performed by an unauthorised decision-maker?Rules in force, event trail, authority record and reasons
Defective procedureCould a formally followed procedure nevertheless be inaccessible, unlawful, inconsistent or unreasonable?Applicable duties, equality impact, design testing and comparative outcomes
System dependency failureDid a delay, record error or decision propagate harm across connected organisations?Linked records, causal sequence and correction history
Unresolved evidenceIs there insufficient material to conclude whether a failure occurred?Disclosure, retention history, independent review and explicitly recorded uncertainty

Sector-specific regulatory and operational overlays

Apply the common 40-area core first, then check the exact legal duties, regulators, service models and versions of rules that applied to the relevant institution and period. The examples below are starting references, not a substitute for jurisdiction-specific legal analysis.

SectorProcess focusRelevant starting standardsIllustrative preventive checks
Higher educationProgramme regulations, academic appeals, disability support, assessment rules, student recordsOIA Good Practice Framework; Equality Act duties; provider regulations — official referenceSupport actually delivered; accessible deadlines; independent appeal review; submission audit trail
EmploymentRecruitment, occupational standards, workplace adjustments, performance, grievances and dismissalEquality Act 2010; Acas reasonable-adjustments guidance; applicable employment law — official referenceAdjustment implementation; protected reporting; independence of review; decision explanations
Financial servicesProduct suitability, consumer support, customer vulnerability, records, disputes and redressFCA Consumer Duty and relevant FCA rules; Financial Ombudsman remit where applicable — official referenceForeseeable-harm prevention; vulnerable-customer support; transparent records; effective complaint resolution
HousingEligibility, allocation or tenancy, repairs, accessible contact, resident complaints and redressHousing Ombudsman Complaint Handling Code for relevant landlords; housing and equality law — official referenceAccessible channels; adjustment records; timely decisions and remedies; fair escalation
Public administrationStatutory powers, benefits, tax, local services, access to evidence, reconsideration and appealsService-specific legislation and tribunal rules; PHSO good administration principles where applicable — official referenceClear legal authority; defensible discretion; accessible reconsideration; evidence availability and correction

Sector prioritisation across the 40 examination areas

These priority areas support efficient review. They are illustrative rather than an exhaustive legal checklist, and no sector should omit another applicable dimension simply because it is not highlighted.

SectorPriority area numbersParticular review focus
Higher education8, 11, 20, 31, 38Reasonable adjustments, discretionary review, appeals, cumulative impacts and exceptional pathways
Employment3, 15, 26, 27, 37Delegated decisions, management capability, cultural pressures, conflicts and safe reporting
Financial services12, 14, 29, 30, 40Clear communications, automated systems, supplier accountability, accessible support and verified corrections
Housing13, 19, 22, 23, 36Tenancy/repair records, backlogs, inter-agency dependencies, impacts and recurring incident learning
Public administration6, 7, 9, 10, 39Policies, statutory powers, eligibility, evidence standards and independent assessment

Evidence-based diagnostic rubric

The following indicators support triage and investigation, not a legal ruling, validated score or ranking of institutions. They must be assigned separately for structural, procedural and preventive integrity. A missing record is not proof of failure.

IndicatorEvidence thresholdReviewer action
Green — documented controlApplicable rule and authority identified, evidence supports execution, and sampled outcome/adjustment or remedy is verifiedRecord evidence and any remaining risks
Amber — uncertainty or partial performanceEvidence incomplete, controls inconsistent, delivery delayed, or safeguards not independently verifiedRequest missing records, examine alternatives and assess interim protections
Red — substantiated material concernReliable evidence indicates departure from applicable requirements, significant control failure or foreseeable harm requiring urgent examinationEscalate for qualified review, preservation of evidence and proportionate safeguards
Grey — not assessed or not applicableInsufficient evidence or a dimension does not apply to the caseExplain scope, missing information and applicability

Diagnostic record for each integrity level

Record the process stage; current and historically applicable rule; decision authority; source and timestamp; reported experience; institutional explanation; competing interpretation; risk or consequence; proposed safeguard; reviewer; and assessment date. Keep personally identifiable records in protected evidence storage, not on a public catalogue.

Safeguards against false conclusions

Do not average colours into a seemingly scientific overall score. Distinguish legal duties from voluntary best practice, the existence of a policy from its implementation, and harm from established causation. Allow non-applicable and unresolved findings. Where a relevant issue is disputed or potentially consequential, obtain independent subject-matter or legal review.

Five descriptive evidence classifications

The classifications below are not a severity scale: execution failures, design defects and systemic recurrence can coexist. The numbers are identifiers, not a ranking, risk score or determination of legal breach. Record evidence confidence and consequence severity separately. An unverified event should not be assigned to a failure category without sufficient evidence.

CodeClassificationEvidence requiredAction
0Insufficient evidence / not yet verifiedRelevant rules, records or chronology cannot yet support a conclusionRequest and reconstruct records; preserve uncertainty
1Evidentiary alignmentApplicable standards and reliable records support the reviewed steps, adjustments and verified remediesDocument baseline and monitor for drift
2Documented execution gapEvidence indicates a departure from a valid applicable requirement or prescribed processReview root cause, capacity, decision authority and controls
3Potential policy or design deficiencyEvidence indicates that faithfully following the process may still produce foreseeable exclusion, inaccessible service or deficient remediesAssess actual legal duties and test policy or workflow redesign
4Pattern of recurring or unremedied failuresMultiple corroborated events, warnings or independently observed unresolved remedies indicate possible systemic recurrenceConsider independent review, cross-case comparison and prevention controls

Application rule: A case may be marked with more than one code (for example, 2 and 4) when each is independently evidenced. Retain the separate green/amber/red/grey triage flags for urgency and evidence sufficiency; do not collapse the two systems into a single numerical score. Preserve the distinction between legal requirements, formal guidance and voluntary good practice.

Evaluating proposed safeguards

Proposed controls should be evaluated for legal validity, accessibility, completion and correction rates, time to resolution, repeat-error frequency, disproportionate impacts, administrative burden, privacy, human oversight and cost. A process map or algorithm cannot independently decide that a rights violation occurred; human and legal review remain necessary for consequential conclusions.

Research status and boundaries

This page provides a reusable reference for future investigation and collaboration. It is exploratory research, not a commissioned audit, active engineering project, published academic finding or allegation against any individual or organisation. Site updates are dated, and future refinements should preserve earlier versions and distinguish proposals from validated outcomes.

Public standards and reference material