ctaio.dev Ask AI Subscribe free

AI Readiness Audit

Know Where You Stand Before You Invest

30-day AI readiness assessment

A structured assessment of your organization's AI readiness across 6 dimensions, benchmarked against Gartner data. The output is a scored profile, gap analysis, and a roadmap your board can act on.

2026 UPDATE

Vendor capture risk: six new audit questions

The audit added a vendor-capture module in 2026. The trigger was the OpenAI acquisition of TBPN on 2026-04-02, a deal that made one fact uncomfortably explicit: AI model providers are not neutral utilities. They acquire media properties, sign exclusive distribution arrangements, and shape the public narrative around AI in ways that change what "best practice" looks like for the rest of us. The right response is not to panic. It is to add concentration, portability, and narrative-exposure questions to your readiness work and then act on the answers.

01

How much of your AI surface depends on a single model provider?

Pick the percentage by spend, by user-facing surface area, and by integrations. If any single number exceeds 60%, you are running concentration risk. The OpenAI–TBPN acquisition on 2026-04-02 is the most recent reminder that model providers are not neutral infrastructure. They acquire media, sign exclusive distribution, and shape narratives in ways that affect your roadmap whether you noticed or not.

02

Can you switch your primary model provider in a single quarter without a public outage?

Run the thought experiment. If your largest provider doubled prices, was acquired by a competitor of yours, or shipped a policy change incompatible with your product, how long until you are stable on an alternative? Under one quarter is portable. One to two quarters is workable. Beyond two quarters is captured.

03

Is your data sovereignty story honest at the regional level?

Where do your prompts, completions, and embeddings actually sit during inference? Which jurisdiction governs them? Which provider sub-processors are in the chain? If the answer is "we trust the vendor SOC2," that is not data sovereignty. It is vendor trust dressed in compliance language.

04

Do you have dual-vendor SLAs in production, or only on paper?

A second provider in evaluation is not a hedge. A second provider in production, behind a routing layer, handling at least 10% of live traffic, is a hedge. Most companies stop at the first. The audit specifically tests whether the secondary path has been exercised in the last 60 days.

05

What is your exposure to vendor narrative capture?

The OpenAI–TBPN deal is the cleanest 2026 example. A media property your competitors and customers watched daily is now owned by an AI provider you may or may not be using. Narrative capture matters because public framing shapes board expectations, regulatory attention, and talent flow. Map which media, conferences, and analyst houses are owned or sponsored by your AI providers. Do not be surprised by stories framed in their favour.

06

Where would model portability break first?

Fine-tuned weights, vendor-specific tools, evaluator setups, and retrieval pipelines are the usual lock-in surfaces. Identify the top three you would have to rebuild on a switch. Estimate the engineer-weeks. If you cannot estimate them, you have not done the work to be portable.

THE PROBLEM

Most organizations don't know what's holding them back

Gartner's June 2025 survey of 195 software engineering leaders found that only 16% believe their delivery processes are ready for AI. Only 14% think their workforce is prepared. Only 12% rate their architecture as AI-ready. These aren't companies ignoring AI — they're companies that have invested but can't pinpoint why results aren't materializing.

84% feel their processes aren't AI-ready
86% feel their workforce isn't prepared
88% feel their architecture isn't ready

Source: Gartner AI-Driven Disruptions in Software Engineering Survey, June 2025

WHAT YOU GET

Four deliverables in 30 days

AI Readiness Scorecard

Per-dimension scores across all 6 pillars, benchmarked against Gartner's 2025 data. Radar chart visualization showing your profile against industry averages.

Gap Analysis Report

Your weakest dimension identified with root cause analysis. Downstream impact mapping showing how one bottleneck affects your entire AI capability.

6-12 Month AI Roadmap

Prioritized action plan sequenced by impact and dependency. Specific initiatives, owners, timelines, and success metrics for each dimension.

Board-Ready Executive Summary

A 5-page document your CEO can hand to the board. Current state, risks, investment case, and recommended next steps in business language.

HOW IT WORKS

The 30-day process

Week 1

Discovery & Stakeholder Interviews

  • Executive alignment interviews (CEO, CTO, business leaders)
  • Current AI initiatives inventory
  • Data landscape mapping — sources, pipelines, quality
  • Regulatory and compliance exposure scan
Week 2

Architecture & Process Assessment

  • Software architecture review (API maturity, cloud readiness)
  • CI/CD and delivery process audit
  • Model serving infrastructure evaluation
  • Technical debt assessment against AI workload requirements
Week 3

Workforce & Governance Audit

  • Engineering skills assessment (AI literacy, prompt engineering)
  • Organizational structure analysis for AI integration
  • Governance framework review (policies, ethics, compliance)
  • Vendor relationship and tool landscape audit
Week 4

Scoring, Roadmap & Delivery

  • 6-dimension readiness scoring with Gartner benchmarks
  • Gap analysis with root cause identification
  • Prioritized 6-12 month AI roadmap
  • Board-ready executive summary and presentation

FREE VS PROFESSIONAL

Self-assessment vs. full audit

Not sure if you need the full audit? Start with the free AI readiness self-assessment on prommer.net. It gives you a directional score in under 3 minutes. If three or more dimensions score below 50, the full audit will show you exactly why and what to do about it.

Free Self-Assessment 30-Day Audit
Time to complete3 minutes30 days
Scoring methodSelf-reported answersStakeholder interviews + technical review
Dimensions assessed6 (directional)6 (deep, with sub-scores)
BenchmarkingGartner averagesGartner + industry-specific peers
Architecture reviewNoHands-on code and infra review
Data quality testingNoPipeline and quality assessment
Custom roadmapGeneral recommendationsPrioritized 6-12 month plan with owners
Board-ready reportNo5-page executive summary
CostFree$25,000-$50,000

Frequently Asked Questions

What is an AI readiness audit?
An AI readiness audit is a structured 30-day assessment of your organization's preparedness for AI adoption and scaling. It evaluates six dimensions: delivery processes, workforce skills, architecture, data quality, governance, and leadership alignment. The output is a scored assessment, gap analysis, and a prioritized 6-12 month roadmap.
How much does an AI readiness audit cost?
A standard 30-day AI readiness audit runs $25,000-$50,000 depending on organization size and complexity. This includes stakeholder interviews, architecture review, data quality assessment, and a board-ready deliverable. For companies spending $500K+ on AI initiatives that aren't delivering results, the audit typically pays for itself by redirecting that spend toward the actual bottleneck.
How is this different from the free self-assessment?
The free AI readiness assessment at prommer.net gives you a directional score in 3 minutes based on self-reported answers. The audit goes deeper: stakeholder interviews across your organization, hands-on architecture and code review, data quality testing, governance gap analysis, and a custom roadmap. Think of the self-assessment as a blood pressure check and the audit as a full physical.
What does the audit deliverable look like?
You receive a comprehensive report with: per-dimension scores benchmarked against Gartner data, a gap analysis identifying your weakest dimension and its downstream impact, a prioritized 6-12 month roadmap with specific actions, a board-ready executive summary, and optionally a 30-minute presentation to your leadership team.
Who should be involved from our side?
The audit works best with access to 8-12 stakeholders: the CEO or COO (for strategic context), CTO or VP Engineering (architecture and process), data team leads (data quality and governance), 2-3 engineering managers (workforce skills), and whoever currently owns AI initiatives. Total time commitment from any individual is 60-90 minutes.
Can the audit be done remotely?
Yes. Most audits are conducted entirely remotely via video interviews, screen-sharing for architecture review, and shared access to code repositories and data documentation. On-site visits are available for organizations that prefer in-person discovery, typically during Week 1.

Start with a conversation

A 30-minute call to understand your situation and whether the audit is the right next step. No pitch, no pressure.