Is Your EA Team Ready for AI Augmented Architecture?
AI Augmented Architecture is a practice shift, not a tool installation. A team is AI-ready when five factors are in place: governed repository, sufficient MDG quality, architect capability to work with AI tools, stakeholder relationships that can absorb self-service, and platform configuration for EA GraphLink. Most teams that attempt AI integration before all five factors are ready achieve either nothing useful or actively damaging results: AI-generated answers from a poorly governed repository are worse than no AI integration because they create false confidence in unreliable data.
Key Takeaways
- AI Augmented Architecture requires five factors in place: all five, not just one or two
- The most common missing factor is MDG quality: not platform configuration or licensing
- Teams that score below 12 on the combined assessment are not ready for EA GraphLink deployment
- The 5-factor score routes teams to the appropriate Sparx Services engagement
- Starting with Amplify (architect capability) before repository governance is in place is a common sequencing mistake
What Is AI Augmented Architecture?
AI Augmented Architecture means the EA practice uses AI tools: EA GraphLink, Kernaro AI Hub, Kernaro Assist, Microsoft Copilot: to expand the reach and speed of architecture intelligence delivery. It does not mean AI replaces architects. It means architects spend less time extracting and formatting data from the repository, and more time on judgment-intensive architecture work. Stakeholders get self-service access to architecture intelligence without needing an architect to produce a report.
The shift is significant: from a periodic report-production model to an always-available intelligence model. But that shift only works if the underlying architecture data is trustworthy, structured, and consistently governed.
The 5-Factor Assessment
Score each factor from 0 to 4:
- 0 = Not in place
- 1 = Partially in place, significant gaps
- 2 = Mostly in place, some gaps
- 3 = In place, minor gaps
- 4 = Fully in place and maintained
Factor 1: Repository Governance
What it means: The Sparx EA repository has active MDG profiles, enforced element type restrictions, mandatory tagged values, naming conventions, and a functioning review process. Packages are organized by domain and ownership is assigned.
How to assess it:
- Can you run a query against the repository and get complete, consistent results for Application Components with lifecycle status and owner?
- Does every top-level package have an assigned owner?
- Is there an active architecture review process for changes to baseline content?
- Have you scored 14+ on the Repository Governance Checklist?
Score 0-1: Repository has no structured governance: significant remediation needed before any AI integration Score 2: Repository has partial governance: Deploy engagement will close the gaps Score 3-4: Repository governance is sufficient for EA GraphLink deployment
Factor 2: MDG Quality
What it means: The MDG profiles in use are correctly configured, consistently applied, and produce repository content that EA GraphLink can index and query reliably. Element types match MDG standard types, tagged values are populated consistently, and relationship types are correct.
How to assess it:
- What percentage of Application Components have all three core tagged values (lifecycle status, owner, criticality) populated?
- Are relationship types consistently used (Serving vs Association vs Composition)?
- Have you ever run an MDG validation report and resolved the findings?
- Is there a process to prevent new elements being created with incorrect types?
Score 0-1: MDG is installed but not governance-grade: EA GraphLink will produce incomplete or incorrect output Score 2: MDG quality is partial: Discover assessment will identify specific gaps Score 3-4: MDG quality is sufficient for reliable AI integration
Factor 3: Architect Capability
What it means: The architecture team can work effectively with AI-augmented tools: they understand how EA GraphLink surfaces data, can interpret Kernaro Assist outputs, can identify when AI answers are incorrect (and trace the issue to the underlying data), and can use AI tools to increase throughput without compromising architecture quality.
How to assess it:
- Has the team received formal ArchiMate training in the last 18 months?
- Can team members identify and correct MDG quality issues independently?
- Is there a structured process for using Kernaro Assist in architecture work?
- Does the team have a shared understanding of where AI-generated architecture outputs require human judgment before use?
Score 0-1: Capability gaps will limit the value of any AI integration Score 2: Capability is functional: targeted Amplify development will close the gaps Score 3-4: Team is ready to work effectively with AI-augmented tools
Factor 4: Stakeholder Relationships
What it means: Business and IT stakeholders are familiar with the EA function, have received architecture outputs, and have showed willingness to use self-service architecture intelligence. Stakeholders who have never engaged with architecture outputs will not independently adopt a new AI interface for architecture data, regardless of its quality.
How to assess it:
- Do you have a regular (quarterly or more frequent) architecture briefing for senior stakeholders?
- Have you delivered architecture-driven insights to a business stakeholder in the last 90 days?
- Is the EA function represented in project gate review processes?
- Have any stakeholders expressed interest in direct access to architecture data?
Score 0-1: Stakeholder relationships are insufficient to support self-service rollout Score 2: Relationships exist but are not mature: stakeholder engagement program needed before rollout Score 3-4: Stakeholder relationships are ready to support Kernaro AI Hub or Copilot integration
Factor 5: Platform Configuration
What it means: EA GraphLink is deployed and configured, or the deployment environment is provisioned and ready. The Sparx EA repository uses a supported database backend (SQL Server, Oracle, MySQL). HTTPS endpoint for MCP Server is available. M365 Copilot licensing is in place if Copilot integration is the target.
How to assess it:
- Is the Sparx EA repository on a SQL Server, Oracle, or MySQL backend?
- Is there a server provisioned for EA GraphLink deployment?
- Is an HTTPS endpoint available for the MCP Server?
- Has the M365 Copilot license level been confirmed for MCP connector support?
Score 0-1: Platform prerequisites are not met: Connect engagement cannot proceed Score 2: Platform partially prepared: specific gaps can be addressed in Connect scoping Score 3-4: Platform is ready for EA GraphLink deployment
Scoring Guide and Routing Table
Total score = Sum of 5 factors (0-20)
| Score | Readiness State | Recommended Next Step |
|---|---|---|
| 0-7 | Not ready | Discover: assess all five factors, produce remediation roadmap |
| 8-11 | Foundation work needed | Deploy (governance) + Amplify (capability), then reassess |
| 12-15 | Conditionally ready | Discover to confirm gaps, targeted Deploy or Amplify based on findings |
| 16-18 | Ready for integration | Connect: EA GraphLink deployment and Copilot/Kernaro integration |
| 19-20 | Fully ready | Connect (immediate) + Kernaro AI Hub planning for GA 2026 |
The Most Common Sequencing Mistake
Teams frequently attempt to sequence AI integration before Factor 1 (Repository Governance) and Factor 2 (MDG Quality) are addressed. The rationale is usually: “let’s see what the AI can do and then improve the data based on what it returns.”
This is backwards. AI tools do not reveal data quality gaps in ways that build trust. They produce answers that look authoritative. Stakeholders receive those answers and act on them. When the answers turn out to be based on stale lifecycle statuses or incorrect owner tags, the damage is to the credibility of both the architecture function and the AI integration.
The correct sequence: governance first, then AI integration. A Discover engagement that scores all five factors and produces a sequenced remediation plan prevents this mistake.
FAQ
What does “AI Augmented Architecture” mean? AI Augmented Architecture is an EA practice model where AI tools: including EA GraphLink, Kernaro AI Hub, Kernaro Assist, and Microsoft Copilot: extend the reach and speed of architecture intelligence delivery. Architects work faster and with more information; stakeholders access architecture data without waiting for reports. It requires a well-governed, MDG-quality repository as its foundation.
What is the minimum MDG quality needed for EA GraphLink? As a practical minimum, EA GraphLink needs: ArchiMate MDG active and consistently used, 80%+ of Application Components with lifecycle status populated, 80%+ with owner tagged, and primary capability-to-application relationships in place. Falling below these thresholds means EA GraphLink returns results with significant gaps: and Copilot answers reflect those gaps.
How long does it take to become AI-ready? For teams scoring 8-11 (foundation work needed), the remediation path is typically 4-8 months: governance and MDG work via Deploy (2-4 months), architect capability development via Amplify (concurrent, 3-6 months), then Connect for EA GraphLink deployment. Teams scoring 12-15 can move to Connect in 2-4 months after Discover confirms the specific gaps.
What factor do most teams fail on? Factor 2: MDG Quality: is the most common gap. Teams frequently have Sparx EA well-configured (Factor 5) and reasonable stakeholder relationships (Factor 4) but have not enforced MDG to the level required for reliable AI integration. Tagged values are defined but not mandatory; element types are mixed; relationships are informal. This is the gap that Discover most often surfaces.
Can I start with Amplify before Connect? Amplify (architect capability development) can run concurrently with Deploy (governance infrastructure) and should precede Connect. Amplify before governance is in place produces architects who are capable but working in a governance-deficient repository: which limits what they can do. The effective sequence is Deploy and Amplify in parallel, then Connect. Amplify alone without governance does not move the AI readiness score.
What does Discover assess for AI readiness? Discover is Sparx Services’ assessment engagement ($25K-$75K). For AI readiness, it assesses all five factors: repository governance audit (MDG configuration, access control, package structure), MDG quality scoring (element type consistency, tagged value completeness, relationship validity), architect capability assessment, stakeholder relationship mapping, and platform readiness check. Output is a scored AI readiness assessment and a sequenced remediation roadmap.
Know Your Score Before You Invest
The Sparx Services Discover engagement provides the authoritative AI readiness assessment for your team: five factors scored, gaps identified, and a concrete roadmap for moving to Connect.
