afrexai-ai-readiness（Afrexai 准备就绪）

AI Readiness Assessment

Run a structured AI readiness audit for any organization. Scores 8 dimensions, identifies gaps, produces a prioritized 90-day action plan with budget ranges.

When to Use

• Before investing in AI/automation tools
• Board or leadership requesting AI strategy
• Evaluating build vs buy decisions
• Annual technology planning

How It Works

Score each dimension 1-5 (1=not started, 5=optimized):

1. Data Infrastructure (Weight: 3x)

• [ ] Centralized data warehouse or lakehouse operational
• [ ] Data quality monitoring automated (freshness, completeness, accuracy)
• [ ] API-first architecture for core systems
• [ ] Data governance policy documented and enforced
• [ ] PII/PHI classification and access controls active

Score 1: Spreadsheets and siloed databases Score 3: Warehouse exists, some pipelines automated Score 5: Real-time streaming, quality >99%, full lineage

2. Process Documentation (Weight: 2x)

• [ ] Top 20 revenue-impacting processes mapped end-to-end
• [ ] Decision trees documented for each process
• [ ] Exception handling paths defined
• [ ] Time-per-task benchmarks established
• [ ] Process owners assigned

Score 1: Tribal knowledge, nothing written down Score 3: Major processes documented, some outdated Score 5: Living documentation, updated quarterly, covers 80%+ of operations

3. Technical Talent (Weight: 2x)

• [ ] At least 1 person understands ML/AI concepts at implementation level
• [ ] Engineering team comfortable with APIs and integrations
• [ ] DevOps/infrastructure person can deploy and monitor services
• [ ] Data analyst can query and interpret model outputs
• [ ] Security team understands AI-specific attack surfaces

Score 1: No technical staff beyond basic IT Score 3: Good engineering team, AI knowledge is theoretical Score 5: Dedicated AI/ML engineer, cross-functional AI literacy program

4. Budget & ROI Framework (Weight: 2x)

• [ ] AI budget allocated (not pulled from "innovation" slush fund)
• [ ] ROI measurement criteria defined before project starts
• [ ] Kill criteria established (when to stop a failing project)
• [ ] Total cost of ownership model includes maintenance, retraining, monitoring
• [ ] Benchmarks set against current manual process costs

Budget Reality by Company Size:

5. Change Management (Weight: 1.5x)

• [ ] Executive sponsor identified and actively involved
• [ ] Communication plan for affected teams drafted
• [ ] Training budget allocated
• [ ] Pilot team identified (volunteers, not voluntolds)
• [ ] Success metrics shared openly with organization

Score 1: Leadership says "just do AI" with no plan Score 3: Exec sponsor exists, some team buy-in Score 5: Change management playbook active, regular town halls, feedback loops

6. Security & Compliance (Weight: 2.5x)

• [ ] AI-specific data handling policy written
• [ ] Vendor security assessment process includes AI criteria
• [ ] Model output logging and audit trail planned
• [ ] Regulatory requirements mapped (GDPR, HIPAA, SOX, SOC 2, EU AI Act)
• [ ] Incident response plan covers AI failures

Score 1: No AI-specific security considerations Score 3: General security strong, AI gaps identified Score 5: AI governance framework active, regular audits, compliance automated

7. Integration Readiness (Weight: 1.5x)

• [ ] Core systems have APIs (CRM, ERP, HRIS, etc.)
• [ ] Authentication/authorization supports service accounts
• [ ] Webhook or event-driven architecture available
• [ ] Test/staging environment mirrors production
• [ ] Rollback procedures documented

Score 1: Legacy systems, no APIs, manual data entry Score 3: Major systems have APIs, some manual bridges Score 5: API-first architecture, event-driven, CI/CD for integrations

8. Strategic Alignment (Weight: 1x)

• [ ] AI initiatives map to specific business objectives (not "innovation")
• [ ] 3-year technology roadmap includes AI milestones
• [ ] Competitive landscape analysis includes AI adoption by rivals
• [ ] Board/leadership educated on AI capabilities and limitations
• [ ] Failure tolerance defined (acceptable experiment failure rate)

Score 1: AI is a buzzword, no concrete strategy Score 3: Strategy exists, loosely connected to business goals Score 5: AI embedded in strategic plan, quarterly reviews, competitive moat building

Scoring

Weighted Total = Sum of (Score × Weight) / Max Possible × 100

90-Day Action Plan Template

Days 1-30: Foundation

• Complete this assessment with honest scores
• Document top 5 processes by time spent × error rate
• Audit data infrastructure gaps
• Set budget and kill criteria

Days 31-60: Pilot

• Select highest-scoring use case (high data readiness + clear ROI)
• Deploy single agent or automation
• Measure daily: time saved, error rate, cost
• Weekly review with stakeholders

Days 61-90: Scale or Kill

• If pilot ROI > 2x: plan 2 more deployments
• If pilot ROI < 1x: diagnose root cause, pivot or kill
• Document learnings regardless of outcome
• Update 3-year roadmap based on reality

7 Assessment Mistakes

1. Scoring yourself too high — External validation beats internal optimism
2. Ignoring data quality — AI on bad data = faster wrong answers
3. Skipping change management — Technical success + team rejection = failure
4. No kill criteria — Zombie projects drain budget and credibility
5. Buying before understanding — Tool purchases before process documentation = shelfware
6. Ignoring security until audit — Retrofitting AI security costs 3-5x more than building it in
7. Comparing to tech companies — Your readiness bar is YOUR industry, not Silicon Valley

Industry Benchmarks (2026)

Get your industry-specific context pack ($47) → https://afrexai-cto.github.io/context-packs/

Calculate your AI revenue leak → https://afrexai-cto.github.io/ai-revenue-calculator/

Set up your first AI agent → https://afrexai-cto.github.io/agent-setup/

Bundles: Pick 3 for $97 | All 10 for $197 | Everything Pack $247