Choose the right transformation pattern

Many organizations begin their AI journey with a broad mandate to "do AI", but that framing rarely leads to meaningful progress. Transformation patterns give that ambition shape. Each pattern represents a distinct strategic intent: a specific answer to what kind of value you're creating, for whom, and how agents make that possible. By aligning on a pattern early, teams move from vague AI initiatives to targeted conversations about vision, investment, and outcomes.

This article introduces the six patterns and explains how the maturity requirements differ across them. Use this overview to understand the full landscape before deep-diving into the specific pattern that matches your current priorities.

Tip

Patterns are design choices, not stages. You don't graduate from one to the next. You pick the patterns that match your intent, and you run them in parallel, each with its own ownership, governance, and metrics. Learn more about how the patterns connect to the maturity model and the Center of Excellence in Agentic transformation patterns.

The six patterns at a glance

Each pattern identifies a scale-breaker, the capability driver most likely to limit your ability to scale a pattern. It's the constraint you need to address first if you want sustainable scale.

Tip

Every scale-breaker maps to one of the five pillars of the Agentic AI adoption maturity model. Your current maturity in that pillar tells you how far you are from being able to scale the pattern reliably. Use the scale-breaker for each pattern as your first investment signal.

Pattern Purpose Who or what the agent serves Assist or execute Scale-breaker
Employee AI enablement Make every employee more capable The individual employee Assist Organization and culture
Business expert empowerment Scale an expert's knowledge Anyone who needs the expert's guidance Assist Technology and data (knowledge quality)
Workplace and IT services Run internal services end-to-end Employees using IT, HR, and finance services Execute Business strategy (service design)
Core business process transformation Transform business-critical workflows The business and its customers Execute Business strategy (process redesign)
External engagement Serve customers and partners directly People outside the organization Execute Governance and security
AI-first capabilities Create net-new capabilities A new capability or market Execute Technology and data (orchestration)

How patterns differ in maturity requirements

The five capability drivers remain consistent across all patterns. What changes is the depth of maturity required in each driver. Use the following table to understand where each pattern concentrates its demands.

Pattern AI Strategy and experience Business strategy Governance and security Technology and data Organization and culture
Employee AI enablement 200 200–300 200 200–300 300
Business expert empowerment 200 200 300 300 300
Workplace and IT services 300–400 400 400 300 200–300
Core business process transformation 500 500 400 400 400
External engagement 300–400 400 500 400 200
AI-first capabilities 500 500 500 500 500

Important

Your pattern determines where you invest, not just how much. Investing in the wrong capability driver for your chosen pattern wastes resources and doesn't unblock scale.

Start from intent, not technology

The fastest way to find your pattern is to name what's driving your AI agenda. Three motivations map cleanly to the patterns:

If your goal is to do something you couldn't do before, use AI-first capabilities to drive innovation.

How to choose

Choosing a pattern is an alignment exercise. You're balancing two things: the business outcomes and vision you want to reach, which is your ambition, and your readiness to deliver them, which is your maturity. The right pattern is the one that matches a meaningful outcome and a level of readiness you can realistically reach. Work through the following four steps.

Step 1: Define the outcome and vision you want

Start with the business result, stated as a measurable change rather than a technology. Examples include lower cost to serve, faster cycle time, higher output quality, more expert capacity, a better customer experience, and new revenue.

Then connect that result to your longer-term vision. How far do you intend to take this part of the business with agents over the next one to three years? Your first outcome sets the entry point. Your vision sets the destination.

Map the outcome to a pattern using the six patterns and start from intent, not technology. The same broad goal can land on different patterns depending on ambition. "Improve customer service," for instance, could map to Employee AI enablement to make your service agents faster, Workplace and IT services to run internal service requests, or External engagement to put an agent in front of customers. Be explicit about which one you mean.

Step 2: Assess your readiness

Rate where you are today across the five capability drivers using the adoption maturity model:

  • AI strategy and experience: Where AI fits in your priorities.
  • Business strategy: How AI connects to business outcomes.
  • Governance and security: What agents are allowed to do.
  • Technology and data: Whether your foundation supports scale.
  • Organization and culture: Whether people are ready to change.

Your readiness profile is an honest snapshot of what you own, govern, and run today—not what you aspire to. Note your weakest driver because it limits how much you can do, no matter how strong the others are.

Step 3: Align ambition with readiness

Each pattern has a target maturity profile. Lay your readiness profile over the target for the pattern your outcome points to, and identify the gap:

Your readiness versus the pattern's target What it means What to do
Meets or exceeds the target Strong fit now Proceed. Deliver, then scale.
One or two drivers fall short You have a clear scale-breaker Close that gap first, or run a bounded pilot while you close it.
Several drivers fall well short Ambition is outrunning readiness Choose a more accessible pattern as a stepping stone, one that builds the very capabilities the ambitious pattern needs.

The biggest readiness jump is the assist-to-execute line. Patterns where agents only assist (Employee AI enablement, Business expert empowerment) keep governance light. Patterns where agents execute (the other four) demand much higher governance, security, and business-strategy maturity, because the agent acts in systems, runs a process, or faces customers. If your vision is an execute pattern but your readiness is assistive, start on the assist side to build the necessary capabilities.

Step 4: Sequence and revisit

Most organizations run two or three patterns at once. Start with the patterns that have high outcome value and the smallest readiness gap. Early wins build the culture, platform, and governance maturity that harder patterns need. They also fund bigger bets.

Reassess quarterly. As your maturity rises, patterns that were out of reach move from "far" to "close," and your destination vision becomes a realistic next step.

Tip

Ambition sets the destination. Readiness sets the starting line. Choose deliberately to close the distance between them. Don't start with a pattern that's beyond your current maturity. Agent initiatives often stall when teams take on more than they can own, govern, and run today. Match each initiative to your current capabilities, and invest in the one gap that holds you back.

Guardrails

Keep these constraints in mind when finalizing your pattern selection:

  • Cap at two or three patterns. Your weakest capability is your ceiling. Don't pick patterns that exceed your current or near-term maturity.
  • Leadership decides, informed bottom-up. Pattern selection must be a leadership commitment, not a bottom-up wish list.
  • Every pattern needs a named outcome and owner. No owner means it's a wish, not a choice.

Next steps

Explore the patterns that match your ambition in more detail:

  • Last updated on