Originally published: August 27, 2025
Jan Dietz's Design & Engineering Methodology for Organizations (DEMO) starts from a stronger claim than most AI strategy discussions do: an enterprise should be understood first in terms of its essential transactions, commitments, and value-creating acts, not in terms of its current workflows, software stack, or reporting structure. If that claim is right, then AI reinvention begins not with automating existing processes, but with understanding what the enterprise actually is.
This article applies that DEMO insight to AI strategy. The central idea is simple: identify the human acts that create value, then design AI to support and strengthen them rather than treating automation as a substitute for enterprise understanding. For a related discussion of conceptual schema and ontology, see What Does an Ontology Actually Give You?. For the governance implications of human intent and delegated execution, see Intent-Driven AI Delegates.
DEMO provides a framework for understanding an organization's core by separating its stable essence from its variable implementation. It focuses on two essential types of human actions:
DEMO reveals the "business DNA" by mapping these essential acts. The point is to distinguish what is structurally central from what is merely one current implementation. That distinction helps keep technological change aligned with the enterprise's actual value creation.
Most organizations use AI to automate existing processes. Think of chatbots replacing call centers or machine learning optimizing advertising campaigns. This approach treats AI as a mere productivity tool, leading to three key risks:
This operational focus also appears in broader enterprise AI failures. A recent MIT Project NANDA study, "The GenAI Divide: State of AI in Business 2025," reported a 95% failure rate for generative AI pilots in enterprises to deliver measurable returns. Whatever the exact number across industries, the larger point is the same: technology-first adoption, without a clear model of business value and human accountability, often produces pilots that do not translate into durable operational results.
The Design & Engineering Methodology for Organizations (DEMO), developed by Jan Dietz, analyzes organizations at the level of their essence. It abstracts away from implementation details so that the transactions that actually create value become easier to see. At its core, DEMO identifies three types of acts performed by human actors:
The key DEMO insight is that only Original Acts (O-Acts), carried by human actors, create business value and commitments. I-Acts and D-Acts are still necessary, but they are support activities for O-Acts. For AI strategy, this distinction matters because it helps separate what may be safely automated from what still requires human judgment, authorization, and accountability. This aligns with the broader distinction between information processing and meaning discussed in What I Mean by Knowledge, Information, and Semantics.
Even if software agents perform some coordination acts, they must do so on behalf of human stakeholders and within human authorization structures. DEMO is useful here not because it solves every ethical question, but because it keeps intent, commitment, and accountability attached to the humans who stand behind the transaction.
Let's illustrate this with a simplified example for ProMatch Inc., a fictitious professional services marketplace. Imagine a core transaction: "Matching a Client with a Service Provider."
A DEMO transaction typically involves a Requester and an Executor engaging in a coordinated dialogue, leading to a Production Act.
Transaction: Matching a Client with a Service Provider
| Step | Act Type | Actor | Essential Act |
|---|---|---|---|
| 1 | C-Act (rq) | Client | Requests a service provider match |
| 2 | C-Act (pm) | ProMatch | Promises to find a suitable match |
| 3 | P-Act (ex) | ProMatch | Executes the matching process (O-Act) |
| 4 | C-Act (st) | ProMatch | States a match has been found |
| 5 | C-Act (ac) | Client | Accepts the proposed match |
In this sequence, the P-Act (Execute matching process) is the core Original Act where ProMatch creates value by identifying a suitable connection. The C-Acts are the communication steps that make this P-Act possible and formalize the commitment between the Client and ProMatch. Looking at the transaction this way helps separate core value creation from information processing and documentation.
A DEMO-guided strategy focuses less on automation in the abstract and more on strengthening essential value-creating activities. Consider the fictitious example ProMatch Inc.
Current AI Usage: ProMatch uses AI for operational tasks like automated matching and content generation. This is an "efficiency-focused" approach.
The Problem: Their AI investment has not created a competitive advantage. Competitors now have similar algorithms, leading to commoditization and reduced client loyalty.
DEMO Analysis and Solution: A DEMO analysis shows that ProMatch's core value comes from human-centered acts. These are the Original Acts that matter most, such as:
The current AI usage supports informational acts (processing data), but not these Original Acts. The DEMO-guided response is to design a new strategy that enhances these human acts with AI.
For example, an AI could analyze business data to surface patterns a human consultant might miss. The human role remains to interpret emotional and cultural signals, make judgment calls, and take accountability for the final outcome. Looking at ProMatch in terms of essential transactions rather than current processes creates room to redesign the business more deliberately, including with AI, software agents, or logic-based reasoning components where they genuinely improve execution.
As AI becomes more sophisticated, a critical question emerges: Who is responsible when things go wrong? To preserve the integrity of your business, a clear distinction must be made between authorized AI and autonomous AI.
For any AI-assisted decision, the accountability test is straightforward: if a human remains responsible for the outcome, the act can be legitimate; if responsibility is displaced onto the machine, it is not. This keeps authorization chains intact and positions AI as an aid to judgment and execution rather than as a substitute for accountable commitment. The same governance theme is developed more fully in Intent-Driven AI Delegates.
In the coming years, information processing, document generation, and routine coordination are likely to become increasingly automated. What remains, and what may become a competitive advantage, is the essence of human contribution:
The organizations that thrive will be those that use AI to strengthen these human capabilities while preserving accountability and trust.