AI Decision & Governance Sprint

Phase Zero:
Establishing the Foundation for AI Decision-Making

You define how decisions about AI are made - grounded in how your organization actually works.

This work starts with how your organization operates today - so decisions about AI reflect real workflows, not assumptions.

Together, we surface what is often implicit:

  • how work flows
  • where knowledge lives
  • how decisions happen in practice
  • where inconsistencies or gaps exist
Why This Phase is Necessary

Establishing structure before complexity is introduced.

AI is entering organizations through multiple paths at once:

  • Individual staff usage
  • Vendor-integrated features
  • External expectations from boards, partners, or funders

In most cases, this creates movement without shared structure.

This phase provides a point of alignment before those patterns become embedded.

The Realities We Surface

The five realities we define

Phase Zero focuses on identifying and clarifying five core realities that shape how AI can function within an organization:

1

Business Reality

How work actually gets done across teams, roles, and responsibilities

2

System Reality

Where work is expected to happen across tools, platforms, and formal systems

3

Representation Gap

Where business reality and system reality do not align. This is often where risk and inefficiency emerge.

4

Decision Reality

How decisions are actually made, including ownership, escalation, and ambiguity

5

Data Reliability

Whether the data being used within systems is consistent, accurate, and usable for AI-related processes

definition before action

AI interacts with existing systems—it does not operate independently of them.

Each of the five realities directly affects how AI behaves in practice:

  • Business Reality determines how outputs are used
  • System Reality determines where AI is applied
  • Representation Gaps introduce risk, inconsistency, and failure points
  • Decision Reality determines whether AI use is governed or ad hoc
  • Data Reliability determines whether outputs can be trusted

If these are not clearly understood, AI implementation tends to amplify existing issues rather than resolve them.

How Phase Zero Works

Phase Zero is structured as a 4–6 week engagement designed to fit within existing organizational workflows.

The process is designed to integrate into ongoing work while building an accurate, shared understanding of how the organization operates in practice.

It typically includes:

  • A small number of focused working sessions
  • Targeted input from key team members
  • Review of existing systems, workflows, and structures
  • Synthesis and mapping of how work, decisions, and data actually function
Outputs

Your Decision Foundation

At the end of Phase Zero, your organization has a clear, shared understanding of how AI should function within your existing work—and the structure to support it.

1

A structured view

of how work, systems, and decisions actually function across your organization

2

A shared definition

of the five core realities shaping how AI operates in your environment

3

Clear visibility into risk and misalignment

across workflows, systems, and decision-making

4

A decision structure for AI use

that defines how decisions are made, owned, and escalated

5

A grounded starting point for next steps

so future AI efforts (policy, tools, implementation) are built on clarity, not assumption)

What This Enables

This phase gives your team the clarity and structure needed to move forward with confidence.

Phase Zero creates the conditions for more effective work in later phases.

It enables:

  • Clearer, more consistent decision-making across teams
  • A shared approach to handling AI-related questions
  • Stronger alignment between systems and real workflows
  • Reduced risk from informal or fragmented usage
  • A foundation your team can build on with confidence
next step

Start with clarity

Phase Zero offers a structured way to understand how your organization operates today and how AI should fit into that reality.

If your team is trying to make sense of AI-related questions, emerging risks, or next steps, this is a grounded place to begin.

Let’s talk

Share a bit about your organization and what you’re trying to make sense of - we’ll respond from there.

Email us to start the conversation