The Enterprise Experience Architect Evaluation Rubric

View raw

{
  "title": "The Enterprise Experience Architect Evaluation Rubric",
  "slug": "eea-rubric",
  "objectNumber": 6,
  "version": "1.0",
  "status": "published",
  "source": "Series Three · Article 4 (capstone)",
  "sourceArticleUrl": "https://medium.com/enterprise-experience-architecture/you-can-describe-the-job-you-cannot-name-it-74d88bd07aae",
  "summary": "A seven-criterion hiring and leveling instrument for the Enterprise Experience Architect: six scored qualities and abilities, plus one gated, unscored baseline — Technical Fluency — that disqualifies a candidate regardless of the other six scores. Each criterion carries a real evidence question, what to listen for in the answer, and a disqualifier, so it evaluates specific answers rather than a checklist anyone can pass.",
  "lastUpdated": "2026-08-14",
  "changelog": [
    {
      "date": "2026-08-14",
      "version": "1.0",
      "note": "Published. Both gate conditions on this object's status (Standard Project Brief v2.9: Series Three, Article 4 published; the interactive scorer built — item 8) are satisfied, content verification against the live article is closed (item 37), and the object's editorial framing (summary, tagline) has been reviewed and approved by the author. Status/version flip only — no content or editorial changes in this entry."
    },
    {
      "date": "2026-08-14",
      "version": "0.2",
      "note": "Content-verification pass against the live, published article closes Build Ledger item 37. Six of seven criteria and Technical Fluency's original three fields confirmed unchanged; one genuine gap found and patched — Technical Fluency's interviewing-methodology note (`interviewingNote`, new optional schema field), missing from the draft this object was originally built from. `sourceArticleUrl` set to the confirmed live URL. Still draft: pending author review of this object's own editorial framing."
    },
    {
      "date": "2026-08-13",
      "version": "0.1",
      "note": "Initial authoring, staged as draft. The seven criteria (six scored, one gated baseline) are locked, verbatim text sourced from the article's final local draft — see the content-verification note in this file's body. Interactive scorer built the same session (Build Ledger item 8's tool-interaction-states gap, closing the EEA Rubric half). Not yet published: pending author review of this object's own editorial framing and a live-article content-verification pass."
    }
  ],
  "type": "diagnostic",
  "tagline": "Score the function nobody wrote a rubric for — before someone else names the job first.",
  "rubricCriteria": [
    {
      "name": "Pragmatic Visionary",
      "scored": true,
      "evidenceQuestion": "Describe an architecture you designed correctly and then deliberately compromised. What did you give up, what did you protect, and who did you have to convince?",
      "listeningFor": "A specific concession to a legacy system, with a clear line around what stayed non-negotiable.",
      "disqualifier": "A candidate whose examples are all north star and no compromise has never governed a real enterprise. So has one whose examples are all compromise."
    },
    {
      "name": "Diplomatic Consensus-Builder",
      "scored": true,
      "evidenceQuestion": "Describe forcing a distributed group to build against a standard they did not want. Not the standard — the forcing.",
      "listeningFor": "Named counterparts in engineering and product, a real objection, and a resolution that did not require escalation — alignment extracted, not volunteered.",
      "disqualifier": "Every example resolved by executive mandate. That is borrowed authority, and this seat cannot run on it."
    },
    {
      "name": "Relentless Pattern Matcher",
      "scored": true,
      "evidenceQuestion": "When did you find the same thing built twice, and what happened next?",
      "listeningFor": "The merge, not the audit. Finding redundancy is observation; consolidating it is architecture.",
      "disqualifier": "The duplication got documented and stayed duplicated."
    },
    {
      "name": "Ecosystem Cartography",
      "scored": true,
      "evidenceQuestion": "Walk me through how a change to one backend record propagates to the front-line surfaces in your last environment.",
      "listeningFor": "They can actually draw it, including the paths that surprise people.",
      "disqualifier": "The map stops at the API boundary."
    },
    {
      "name": "Governance & Scale Strategy",
      "scored": true,
      "evidenceQuestion": "What did you build that other teams built with, and what happened when they wanted to deviate?",
      "listeningFor": "A contribution model, a versioning discipline, and a real deviation they either absorbed or refused.",
      "disqualifier": "The standard existed and nobody used it. Adoption is the evidence; publication is not."
    },
    {
      "name": "Business Translation",
      "scored": true,
      "evidenceQuestion": "Give me the sentence you used to get funding for governance work, and tell me what the CFO said.",
      "listeningFor": "Hours, tickets, cycle time, inference cost, incident count — a figure someone else could verify.",
      "disqualifier": "The case was made on quality, consistency, or user experience alone. That case has been losing budget reviews for twenty years."
    },
    {
      "name": "Technical Fluency",
      "scored": false,
      "evidenceQuestion": "Two parts, and both are required. First: hand them a real API payload or JSON schema from your own systems and ask them to walk through what it binds and what it would break if a field changed. Second: what is the last thing an engineer told you was infeasible, and what did you do?",
      "listeningFor": "On the first, an accurate read of the schema without an engineer translating it for them. On the second, they understood the constraint well enough to respecify around it.",
      "disqualifier": "They cannot read the schema unassisted — this is the gate itself, not a soft signal, and nothing else on this rubric compensates for failing it. On the second question: they escalated, or they accepted it without understanding it. Both produce the same specification.",
      "interviewingNote": "Do not hand an external candidate a live production payload. Build one sanitized reference schema, structurally real but stripped of anything proprietary, and reuse it across every interviewer — the point is whether the candidate can read a schema unassisted, not whether they can read your schema, and a shared reference is what keeps the gate calibrated across a hiring panel instead of drifting interviewer to interviewer."
    }
  ]
}