This worked report shows how an organization’s answers become an Artebello AI Readiness Index™ score and a practical action plan. The fictional example uses the 18-question base assessment without industry or regional additions. It is not a client case study, a benchmark study, or a claim about outcomes achieved by Artebello.
The example organization
Imagine a professional-services business considering an internal knowledge assistant. Leaders have a defined objective and a budget sponsor, but the team’s documents are spread across systems. Some documents have unclear access permissions, and staff do not yet share an approved process for checking AI-generated answers.
The proposed first use case is to help staff find approved operating guidance. The pilot should answer from a limited, curated document collection; show the source of each answer; and send unsupported questions to a person. The organization should repair data and governance gaps before connecting a broader collection of sensitive documents.
All 18 answers
Each number corresponds to the assessment’s 1–5 scale. These are invented answers for demonstrating the calculation.
| Pillar | First answer | Second answer | Third answer | Rounded pillar score |
|---|---|---|---|---|
| Business | 4 | 4 | 3 | 67 |
| People | 2 | 3 | 3 | 42 |
| Process | 3 | 3 | 4 | 58 |
| Data | 2 | 2 | 3 | 33 |
| Technology | 3 | 4 | 3 | 58 |
| Governance | 2 | 3 | 2 | 33 |
For Business, the average answer is 11 ÷ 3 = 3.6667. The pillar conversion is round(((3.6667 − 1) ÷ 4) × 100) = 67.
The six rounded scores sum to 291. The overall index is round(291 ÷ 6) = 49. A score of 49 falls within Artebello’s Emerging band of 40–59. The public methodology explains the complete scoring model.
What the result tells leadership
Business is the strongest pillar in this example. Data and Governance are the lowest. This suggests that sponsorship is ahead of the controls needed for the proposed assistant. It does not establish that the business is 49% compliant or that its pilot has a 49% chance of success.
The current diagnostic’s illustrative comparisons put Data 11 points below its reference, Governance 8 points below, and People 4 points below. These three become the suggested priorities. The references are demonstration values, not observed industry averages. Any percentile-style display is also illustrative.
Evidence review before approving a pilot
| Gap | Evidence to inspect | Example action | Acceptance criterion |
|---|---|---|---|
| Data ownership | Document inventory and named owners | Curate the first collection and assign owners | Each included document has an owner, current version, and access classification |
| Permissions | Identity groups and document access rules | Enforce source permissions in retrieval | A test user cannot retrieve restricted material |
| Oversight | Approved-use policy and review process | Define when staff must verify an answer | Test users can locate the original source and escalate unsupported answers |
| Skills | Role-based task exercise | Train the pilot group using realistic questions | Each participant demonstrates verification and safe escalation |
These are example acceptance criteria to adapt. Sensitive or regulated uses can require additional specialist review. A readiness score alone does not approve a system for deployment.
A worked 90-day plan
Days 1–30: establish the foundation. The executive sponsor confirms the business problem and the budget boundary. The data owner curates a limited document set. The governance owner records the intended use, exclusions, oversight process, and escalation route. The team measures how long staff currently take to find verified answers.
Days 31–60: run a bounded evaluation. The technical owner connects the approved collection and enforces permissions. The pilot group tests a prepared set of answerable, unanswerable, ambiguous, and restricted questions. Reviewers record accuracy, source quality, permission failures, and the effort needed to correct outputs.
Days 61–90: decide what happens next. Compare task time, answer quality, and review effort with the baseline. Fix recurring failures before expansion. Continue, narrow, redesign, or stop the pilot based on the evidence. Update the risk register and repeat the assessment using the same profile after changes have been implemented.
What a useful follow-up report should contain
A later report should retain the original answers, question set, assumptions, and dates. It should explain which practices changed, which evidence supports those changes, and which gaps remain. Reporting a higher score without improved operating evidence is not enough.
Track both capability and business results. A new policy and a completed training session can improve readiness. They do not, by themselves, demonstrate saved cash, improved service, or safe production behavior.
Use the example
- Download this worked report.
- Review the executive readiness checklist.
- Check production launch gates.
- Take your own AI readiness assessment.
For an evidence review and tailored roadmap, discuss a guided assessment with Artebello.