Introduction: The AS-IS Challenge in Clients Without Documentation

In Enterprise Architecture (EA) projects, the very first and most critical step is establishing an accurate AS-IS baseline. However, in many companies, formal documentation simply does not exist. Knowledge about applications, integrations, and infrastructure often lives only in the minds of employees.

In environments like this, trying to capture requirements directly into tools such as SAP LeanIX—or relying on long email threads—becomes slow, inefficient, and error‑prone.

To address this challenge and accelerate the Assessment phase, we developed and successfully applied the Technical AS-IS EA Canvas methodology.
You can access the methodology assets here:
👉 link

Our approach innovates by separating human interaction from metadata capture, using AI to produce high-fidelity insights from the workshop.

1. The Tactile and Visual Workshop (The Canvas)

The heart of the methodology is the Technical AS-IS EA Canvas, a large-format (A0/A1) printed board designed for collaborative mapping.

Key characteristics:

• Structure: The Canvas visually separates the technology landscape into On-Premise, Cloud, and External, aligned with the TOGAF ADM phases.
• Interaction: Participants use colored post-its (following ArchiMate patterns for Application, Technology, and Business) to map systems, while strings represent integration flows.
• Capabilities: Tools like SAP Signavio Process Explorer are used during the session to anchor discussions and ensure the mapping reflects true business capabilities.

2. High-Fidelity Data Capture (Audio + AI)

The major differentiator of this methodology is how the technical metadata is captured:

• Recording & Cognition: Instead of taking notes, the architect focuses on facilitating the session using a guided Cheat-Sheet. The entire session is recorded.
• AI Synthesis: The audio is then processed by an AI engine that extracts the essential metadata:
  • protocols
  • integration frequency
  • triggers
  • pain points
  • non-functional requirements
  • security considerations
  • and other technical details often lost in traditional workshops

This ensures no insight is missed.

3. The Result: Structured Data Ready for SAP LeanIX

The combination of photo + audio + AI extraction produces a reliable and complete dataset.

• Structured Output: AI produces clean, structured metadata from the discussion.
• Direct Import: This dataset is formatted and imported into SAP LeanIX, feeding the EA repository with a high-confidence AS-IS.
• Focus on To-Be: With the AS-IS complete, architects can move directly to Gap Analysis and To-Be Architecture, drastically accelerating the project lifecycle.

The Technical AS-IS EA Canvas methodology demonstrates how tactile, visual engagement combined with AI capabilities solves one of the biggest challenges in Enterprise Architecture: capturing knowledge in undocumented environments.

We were able to extract all relevant requirements, applications, and integration flows—building a robust EA baseline in SAP LeanIX with confidence and speed.

This approach not only reduces risk but transforms the AS-IS assessment into an engaging, high-value experience for all participants.

What about you?

I would love to hear your experiences with AS-IS capture in undocumented environments.
How do you deal with missing documentation in EA projects?

Share your thoughts and let’s exchange best practices!

Transparency Note: This post was drafted with the assistance of an Artificial Intelligence model to ensure clarity and structure.