Archive of Christian Ullrich

Organizational Design for the GenAI Age Reflex Area

2026-09-29

Table of Contents

Establishing the Case and Requirements for Organizational Redesign

Recognizing When Individual GenAI Gains Have Become a Structural Issue

Distinguishing Organizational Redesign from Local Workflow Improvement

Testing Whether Existing Structures Still Match How Work Gets Done

Separating Temporary GenAI Effects from Durable Design Requirements

Translating Strategic Priorities into Organizational Design Requirements

Identifying Constraints the New Organization Must Continue to Respect

Deciding Which Organizational Level Actually Needs Redesign

Establishing What Must Be Preserved Through Structural Change

Setting an Evidence Threshold Before Redesigning the Organization

Defining Design Criteria Before Choosing a New Structure

Reconstructing the Architecture of Work

Mapping How Cognitive Work Flows from Need to Outcome

Identifying Work Built Around Scarce Information Processing

Separating Production, Evaluation, Integration, Judgment, and Accountability

Locating Tasks That GenAI Has Materially Changed

Identifying Valuable Work That Becomes Newly Feasible

Tracing Human Dependencies Across the Existing Workflow

Identifying Work That Still Requires Direct Human Contribution

Separating Activities That No Longer Need to Stay Bundled

Recombining Activities That No Longer Need Separate Specialists

Comparing the Existing Work Architecture with a GenAI-Enabled Alternative

Redesigning Around New Bottlenecks and Organizational Constraints

Responding When Production Capacity Is No Longer the Main Constraint

Responding When Verification Becomes the New Bottleneck

Responding When Decision Capacity Cannot Keep Up with Analysis

Responding When Expert Attention Becomes Scarcer Than Output

Responding When Integration Becomes Harder as Local Output Expands

Responding When Exception Handling Dominates Remaining Human Work

Responding When Prioritization Becomes Harder as More Work Becomes Feasible

Responding When Data, Context, or Access Limits GenAI-Enabled Work

Responding When Managerial Attention Becomes the Limiting Resource

Determining Whether a Constraint Has Disappeared or Merely Moved Elsewhere

Redesigning Role and Job Architecture

Redesigning a Role After Routine Production Work Disappears

Broadening a Role into Adjacent Work Made Accessible by GenAI

Combining Roles Whose Historical Separation No Longer Adds Value

Splitting Production, Review, and Accountability into Different Responsibilities

Preserving a Role Whose Tasks Change but Core Responsibility Does Not

Updating a Role Whose Formal Description No Longer Matches Actual Work

Correcting Roles That Gain Responsibility Without Matching Authority

Shifting a Role from Production Toward Orchestration and Judgment

Preserving Hidden Responsibilities When Visible Tasks Are Automated

Retiring or Reconstructing a Role Whose Original Purpose Has Disappeared

Repositioning Specialist Expertise

Moving Routine Specialist Cases into AI-Enabled Generalist Work

Shifting Specialists Toward Exceptions and Difficult Cases

Preserving Specialist Independence for Challenge and Assurance

Expanding Specialist Capacity When GenAI Creates More Work to Review

Choosing Between Centralized and Embedded Specialist Expertise

Distinguishing Access to Specialist Knowledge from Genuine Expertise

Separating Specialist Ownership from Routine Specialist Participation

Creating Shared Specialist Pools for Selective Intervention

Redefining Specialist Functions Around Standards, Architecture, and Judgment

Deciding Which Specialist Capabilities Need a Dedicated Organizational Home

Designing Human-AI and Human-Agent Work Configurations

Using GenAI as an Assistant Within a Human-Owned Task

Delegating a Bounded Task to AI While Humans Integrate the Result

Delegating a Multi-Step Workflow to an Agent with Defined Checkpoints

Supervising Multiple Agents from a Single Human Role

Dividing Routine Cases and Exceptions Between AI and Humans

Running Human and AI Work in Parallel Before Integrating the Result

Allowing Agents to Execute Actions Within Predefined Boundaries

Extending the Amount of Work Delegated Before Human Intervention

Coordinating Several Agents Within One Human-Owned Workflow

Determining Where Direct Human Participation Must Remain Continuous

Redesigning Team Composition and Support

Reducing Team Size After GenAI Expands Individual Capability

Broadening Team Roles Without Losing Necessary Depth

Moving Specialists from Permanent Team Membership to On-Demand Support

Reducing Routine Support Capacity Around High-Value Professionals

Repurposing Support Roles Toward Coordination, Review, and Exceptions

Choosing Between an AI-Enabled Individual and a Multi-Person Team

Adding Independent Review to a Highly AI-Enabled Team

Adding Technical AI Capability to an Existing Domain Team

Varying Team Composition According to Case Complexity

Preserving Human Collaboration Where AI Does Not Replace Its Value

Redesigning Managerial Work

Automating Routine Reporting and Administrative Management Work

Redirecting Managerial Capacity Toward Coaching and Development

Managing Increased Output from AI-Enabled Teams

Supervising Work Performed by Both People and Agents

Concentrating Managerial Attention on Exceptions and Difficult Decisions

Redesigning a Manager Role That Primarily Relayed Information

Responding When Managerial Judgment Becomes a Bottleneck

Setting Quality and Delegation Boundaries for AI-Enabled Work

Making Managers Responsible for Future Capability Development

Redesigning Management When Automation Increases Rather Than Reduces Workload

Reconsidering Management Layers and Spans of Control

Reassessing a Layer That Primarily Aggregates and Transmits Information

Widening Spans After Routine Managerial Work Declines

Preserving Narrower Spans Where Coaching and Judgment Remain Intensive

Removing a Layer and Redistributing Its Remaining Responsibilities

Combining Management Layers Whose Distinct Purposes Have Weakened

Preventing Flatter Structures from Producing More Senior Micromanagement

Repurposing a Layer Instead of Eliminating It

Accounting for Agent Supervision When Setting Management Spans

Correcting Delayering That Creates New Coordination Gaps

Piloting Wider Spans Before Making Permanent Structural Changes

Redistributing Decision Rights and Authority

Moving Decisions Downward When Frontline Capability Increases

Retaining Central Authority Where Local Decisions Create Enterprise-Wide Consequences

Reconsidering Decisions That Remain Central Only Because Expertise Was Scarce

Correcting Decentralized Authority Where Local Capability Is Still Insufficient

Removing Escalations Whose Original Information Constraint Has Disappeared

Preventing Better Central Visibility from Producing Unnecessary Central Control

Moving Decisions Closer to Customers or Operations

Centralizing Decisions That Require Coordination Across Interdependent Units

Separating Improved Analytical Capability from Formal Decision Authority

Rebalancing Decision Rights After Workflows and Roles Have Changed

Designing Human-AI Decision and Accountability Boundaries

Separating AI-Generated Options from the Authority to Choose Among Them

Preventing AI Recommendations from Becoming Unexamined Decisions

Allowing Agents to Execute Actions Without Giving Them Unbounded Authority

Designing Human Approval for Work the Approver Did Not Personally Produce

Assigning Accountability When Several AI Systems Contribute to One Outcome

Separating Recommendation, Approval, Execution, and Accountability Across Roles

Clarifying Ownership After an Automated Workflow Replaces Human Handoffs

Delegating Reversible Low-Consequence Decisions to Automated Systems

Requiring Explicit Human Approval for Difficult-to-Reverse Actions

Resolving Accountability After an Automated Decision Chain Fails

Designing Independent Challenge and Escalation

Setting Escalation Thresholds Based on Consequence, Uncertainty, and AI Autonomy

Introducing Independent Review for High-Consequence Decisions

Routing Difficult Exceptions to Qualified Specialists

Correcting Oversight Roles That Lack Real Authority to Intervene

Clarifying Override and Stop Authority for Autonomous Work

Protecting Reviewers from Routine Acceptance of AI Recommendations

Redesigning Escalation When Review Queues Become Bottlenecks

Preserving Separation of Duties After Work Becomes More Automated

Preventing Central Assurance Functions from Becoming Universal Approval Queues

Moving from Universal Review to Risk-Based and Exception-Based Review

Reducing Handoffs and Coordination Burden

Distinguishing Faster Handoffs from Truly Removed Handoffs

Allowing One Role to Absorb an Adjacent Workflow Step

Removing Routine Approval Transfers That No Longer Add Judgment

Reducing Translation Work Between Specialist Functions

Replacing Repetitive Coordination with Shared AI-Enabled Context

Shortening Cross-Functional Waiting Time Without Losing Ownership

Preventing Removed Handoffs from Reappearing as Review Work

Identifying Coordination Hidden Inside Agentic Workflows

Reducing Repeated Transfers of Ownership Across One Outcome

Rebundling Sequential Work Around End-to-End Responsibility

Redrawing Functional and Organizational Boundaries

Reconsidering a Functional Boundary Built Around Scarce Expertise

Preserving a Boundary That Provides Independent Challenge

Reconsidering Geographic Boundaries After Translation and Information Costs Fall

Redrawing Boundaries That Create Repeated Cross-Functional Dependencies

Reconsidering Boundaries Between Frontline and Back-Office Work

Preserving Specialist Units for Standards, Depth, and Exceptional Cases

Clarifying Boundaries That Become Blurred by Shared AI Systems

Reconsidering Units Whose Main Purpose Is Information Transformation

Preserving Separate Units Where Local Context Still Matters Materially

Testing Whether Organizational Units Should Merge, Split, or Change Purpose

Choosing Structural Forms for GenAI-Enabled Work

Reconsidering a Functional Structure After Expertise Becomes More Accessible

Reconsidering a Matrix Structure When Coordination Costs Change

Moving Toward Product or Outcome-Based Structures

Combining Outcome Teams with Shared Enterprise Platforms

Reassessing Geographic Structures as Distance Matters Less for Cognitive Work

Reassessing Customer-Based Structures as Personalization Becomes Cheaper

Using Networked Structures for More Autonomous Teams

Designing Hybrid Structures for Uneven GenAI Maturity and Risk

Reconsidering Divisional Structures When Shared GenAI Capabilities Span Business Units

Selecting a Structural Form After Work and Boundaries Have Been Redesigned

Creating End-to-End Outcome Ownership

Assigning Ownership Where One Outcome Crosses Several Functions

Consolidating Ownership After AI Removes Intermediate Handoffs

Giving Product Teams Responsibility for End-to-End Results

Assigning Process Ownership Across Humans, Agents, and Systems

Establishing Ownership Across an Entire Customer Journey

Preserving Local Outcome Ownership on Shared Enterprise Platforms

Resolving Conflicting Functional Metrics Around One Outcome

Correcting Outcome Ownership That Lacks Matching Decision Rights

Assigning Ownership When External Providers or AI Perform Much of the Work

Defining Where One End-to-End Outcome Begins and Ends

Balancing Centralization, Federation, and Local Autonomy

Centralizing Enterprise-Wide Dependencies Without Centralizing Local Delivery

Granting Greater Autonomy as Domain Capability Matures

Managing Rapid Local Experimentation Across Business Units

Combining Central Standards with Local Delivery Ownership

Varying Local Autonomy According to Risk and Consequence

Reducing Duplicated GenAI Capability Across Autonomous Units

Correcting a Central Model That Has Become a Delivery Bottleneck

Constraining Local Autonomy When Enterprise-Wide Risks Accumulate

Clarifying Responsibilities in a Federated Operating Model

Moving from Centralized Delivery Toward Federation Over Time

Organizing Shared GenAI Capabilities and Platforms

Deciding Which GenAI Capabilities Should Be Reusable Across the Enterprise

Providing Common Model Access Across Multiple Business Units

Organizing Shared Retrieval and Enterprise Knowledge Access

Providing Common Infrastructure for Agentic Work

Centralizing Reusable Evaluation and Monitoring Capability

Providing Shared Identity, Authorization, and Security Foundations

Building Domain Applications on Common Enterprise Platforms

Consolidating Duplicated Local GenAI Infrastructure

Funding Shared Capabilities Whose Benefits Span Multiple Units

Turning a Successful Local GenAI Solution into an Enterprise Capability

Redesigning Shared Services, Staff Functions, and Centers of Excellence

Automating Routine Production Within a Shared Service

Reconsidering a Staff Function Built Mainly Around Information Intermediation

Using a Center of Excellence to Concentrate Scarce Early Expertise

Correcting a Center of Excellence That Has Become an Approval Bottleneck

Shifting a Center of Excellence from Delivery Toward Enablement

Moving Shared Services Toward Exception-Based Human Work

Refocusing Staff Functions on Judgment, Standards, and Integration

Reducing Central Service Demand as Embedded Capability Grows

Sunsetting Temporary AI Transformation Offices

Preserving Central Functions That Still Provide Scale or Independent Challenge

Rebalancing Workforce Composition

Shifting Workforce Capacity Between Functions as GenAI Changes Demand Unevenly

Rebalancing the Mix of Junior, Mid-Level, and Senior Professionals

Shifting from Routine Support Roles Toward Expert Review and Exception Handling

Rebalancing Generalist and Specialist Capacity

Reassessing the Ratio of Managers to Individual Contributors

Determining How Much Dedicated Technical AI Expertise the Organization Needs

Redeploying People Whose Existing Work Shrinks

Rebalancing Human Capacity Across Production, Review, Integration, and Judgment

Correcting a Workforce That Becomes Too Senior-Heavy

Planning Human Capacity Alongside Growing Agent Capacity

Redesigning Entry-Level Work and Career Architecture

Redesigning Entry-Level Roles After Routine Learning Work Is Automated

Replacing Apprenticeship Paths That No Longer Arise Naturally from Production

Giving Junior Employees Supervised Responsibility in AI-Enabled Work

Shifting Progression Criteria Toward Demonstrated Judgment

Rebuilding Career Paths After Intermediate Rungs Disappear

Using Rotations to Replace Narrow Repetition as a Learning Mechanism

Redesigning Promotion Criteria When Output Volume Becomes Less Meaningful

Avoiding Entry-Level Roles That Demand Senior Judgment Without Development

Preserving the Future Management Pipeline as Early-Career Work Changes

Designing Parallel Generalist, Specialist, and Leadership Career Paths

Preserving Expertise, Learning, and Succession

Preserving Expertise When Experts Perform Less Routine Work

Capturing Tacit Knowledge Before Experienced Employees Leave

Creating Deliberate Practice When Real Work Provides Fewer Repetitions

Building Successors for Highly Leveraged AI-Enabled Experts

Maintaining Selected Skills Through Periodic Unaided Practice

Using Simulation to Develop Judgment for Rare or Difficult Cases

Preserving Professional Communities After Expertise Becomes More Distributed

Transferring Expertise When Critical Work Moves Between Roles or Functions

Preventing AI Assistance from Replacing Human Mentorship

Maintaining Independent Human Competence as AI Takes Over Routine Execution

Allocating Productivity Gains and Released Capacity

Distinguishing Time Savings from Usable Organizational Capacity

Using Released Capacity to Increase Output

Using Released Capacity to Improve Quality

Using Released Capacity to Shorten Cycle Times

Using Released Capacity to Reduce Backlogs and Unmet Demand

Preserving Deliberate Slack for Resilience and Unexpected Work

Converting Durable Excess Capacity into Staffing Reduction

Using GenAI to Perform Valuable Work That Was Previously Uneconomic

Responding When Lower Costs Create Additional Demand

Reallocating Capacity After the Bottleneck Moves Elsewhere

Reconsidering Organizational Scale, Sourcing, and Firm Boundaries

Bringing a Capability In-House After GenAI Reduces Its Minimum Viable Scale

Reconsidering Outsourcing Built Primarily on Labor Cost Advantages

Buying Agentic Services Instead of Building Internal Delivery Capacity

Insourcing Strategically Critical Expertise Despite External Availability

Reassessing Contractor and Contingent Workforce Models as Routine Cognitive Work Shrinks

Reassessing the Role of Consultants in AI-Enabled Knowledge Work

Choosing Whether to Build, Buy, or Partner for a GenAI-Enabled Capability

Using External Specialists for Expertise Needed Intermittently Rather Than Continuously

Reassessing the Minimum Viable Size of a Function

Redrawing the Boundary Between the Organization and Its External Ecosystem

Redesigning Performance Measures and Incentives

Replacing Output Volume as the Primary Measure of Knowledge Work

Avoiding AI Usage as a Proxy for Performance

Rewarding People for Sharing Reusable AI-Enabled Practices

Preventing Productivity Gains from Automatically Becoming Higher Workload Targets

Making Verification and Quality Work Visible in Performance Systems

Balancing Individual Contribution with Team and System Outcomes

Rewarding Disciplined Experimentation Without Rewarding Careless Failure

Removing Incentives to Preserve Manual Work for Budget or Headcount Protection

Preventing Metrics from Rewarding Excessive or Unsafe Automation

Shifting Performance Measures Toward End-to-End Outcomes and System Health

Designing Organizational Resilience Around Critical AI Dependencies

Managing Dependence on a Single Model or AI Provider

Preparing for Failure of a Shared Enterprise AI Platform

Reducing Dependence on a Small Number of Critical Specialists

Restoring Human or Non-AI Fallbacks for Critical Work

Preserving Capability When Employees Can No Longer Work Without AI

Preventing Central Review Functions from Becoming Single Points of Failure

Maintaining Viable Alternative Providers or Technical Paths

Designing Reduced-Function Operating Modes for AI Disruption

Preparing for Common-Mode Failures Across Many Agents or Workflows

Balancing Efficiency Gains Against Concentration and Resilience Risks

Testing, Transitioning, and Reversing Structural Change

Piloting a Structural Change Before Enterprise-Wide Adoption

Running Old and New Organizational Arrangements in Parallel

Using Temporary Roles During an Organizational Transition

Defining Reversal Criteria Before a Structural Experiment Begins

Sequencing Permanent Changes After Capabilities and Workflows Stabilize

Preserving Institutional Knowledge Before Removing a Role or Unit

Tracking Hidden Work After Responsibilities Are Removed or Moved

Revising or Reversing a Redesign That Does Not Produce the Expected Results

Sunsetting Temporary Structures After Their Transition Purpose Ends

Maintaining Clear Accountability While Old and New Structures Coexist