AI readiness assessment for mid-market companies

Is your company ready for AI? The honest answer for most $5M–$25M companies is “partially.”

An AI readiness assessment is how you find out which part.

It scores five dimensions of your business and tells you where the gaps are. The data, the workflows, the team, the tooling, the people at the top.

Most leaders guess at readiness and guess wrong. They overrate the technology and underrate the humans who have to change how they work.

This is the diagnostic that comes before the build.

Key Takeaways

• Five dimensions: A readiness assessment scores data, workflows, team capability, tooling, and leadership alignment together.
• People over tech: Most companies overestimate technology readiness and badly underestimate how ready their people are.
• Hours, not weeks: A readiness assessment is a diagnostic you run in hours; it is not a consulting engagement.
• A prioritized roadmap: The output tells you what to fix first, what to build next, and what to skip.
• Run it first: Running the assessment before hiring a firm saves real time and money on both sides.

What does an AI readiness assessment measure?

An AI readiness assessment measures five dimensions: data readiness, workflow readiness, team readiness, tooling readiness, and leadership readiness. A score in each one shows where you are strong and where the work has to start.

Readiness is not one number. A company can have clean data and a team that refuses to change; an average would hide that. You score each dimension on its own.

• Data readiness: Whether your customer, financial, and operational data is accurate, accessible, and consistent enough for AI to use.
• Workflow readiness: Whether your core processes are documented well enough that a system can follow them.
• Team readiness: Whether your people can adopt new tools and want to, beyond one or two enthusiasts.
• Tooling readiness: Whether your current stack can connect to AI without a rebuild of how you operate.
• Leadership readiness: Whether the people setting priorities will fund, defend, and use the work themselves.
• The weakest link: Your real readiness is the lowest of the five scores, not the average across them.

Each dimension maps to one of the AI maturity levels that determine your starting point. Scoring all five tells you which level you actually occupy, not the one you hope you do.

What makes a company AI-ready?

A company is AI-ready when its core operations are documented, at least one executive uses AI personally and pushes adoption, there is budget for a real engagement, and the team is willing to change how it works.

Readiness is mostly about conditions you can name. Vague enthusiasm for AI is not one of them. The four below are the ones that separate companies that move from companies that stall.

• Documented core operations: Your main workflows are written down clearly enough that someone new could run them.
• An executive who uses AI: At least one leader works with AI daily and pushes the team to adopt it.
• Budget for the work: Money set aside for a 3–6 month engagement, not a one-month experiment that proves nothing.
• Willingness to change: The team accepts that AI means doing familiar work in unfamiliar ways.
• A named owner: Someone whose job includes maintaining the AI work, so it improves instead of quietly decaying.

The executive condition matters most. A leader who personally understands giving AI the right context about your business will set realistic expectations and defend the work when adoption gets hard.

What do most companies get wrong about readiness?

Most companies assume readiness means owning the right software. It does not. The common mistakes are buying tools before fixing process, skipping documentation, and underestimating how hard senior people resist change.

These errors share a root. They treat readiness as a purchase rather than a state of the business. The three below show up in almost every stalled engagement we see.

• Mistake 1, the software trap: Believing the newest AI tool makes you ready when readiness lives in process and people.
• Mistake 2, skipping documentation: Launching AI on undocumented workflows, so the system has nothing reliable to learn from.
• Mistake 3, ignoring resistance: Underestimating how much senior team members quietly resist tools that change their judgment.
• The pilot trap: Treating a one-month tool trial as proof of readiness, when adoption is a longer test.

Documentation is where most companies are weakest and least honest. The fix is structuring a knowledge base that AI can actually use, which turns scattered memory into something a system can read.

When should you run an AI readiness assessment?

Run a readiness assessment at three moments: before you evaluate AI consulting firms, before you buy any AI tools or licenses, and when the CEO feels ready but is unsure the company is.

Timing matters because the assessment changes what you do next. Run it too late and you have already spent money on the wrong thing. These three triggers are the natural moments.

• Before evaluating firms: A clear readiness picture lets you brief any firm accurately and judge whether their proposal fits.
• Before buying tools: Knowing your gaps stops you from paying for licenses your team is not ready to use.
• When the CEO is ready: A leader who feels ready can confirm whether the wider company is, before committing budget.
• After a failed start: When an earlier AI push stalled, an assessment names the gap that quietly caused it.

If you are running the assessment because you fear you are behind, read the signals that your company is falling behind first. It separates real urgency from competitor noise.

How do you run an AI readiness assessment?

You can run a readiness assessment three ways: a 10-question self-assessment, a 3-minute voice assessment, or a full formal audit. Start with the self-assessment; most companies learn enough from it to decide their next step.

The three methods sit on a spectrum of depth. A quick self-score tells you whether to keep going. The deeper assessments confirm the detail and feed directly into a build.

• Self-assessment, 10 questions: The Phos AI Readiness Scorecard scores all five dimensions and returns a prioritized view in minutes.
• Voice assessment, 3 minutes: A spoken walkthrough that surfaces gaps a checkbox form tends to hide from you.
• Formal assessment, full audit: A complete review run inside the AI Foundations phase, mapping every workflow and data source.
• Combine the quick two: Run the Scorecard and Voice Audit together; the checkboxes and the conversation catch different gaps.

For the voice route, the Phos AI Voice Audit asks about your operations out loud and listens for the hesitation that signals a gap. Either tool scores you within the hour.

What comes after the assessment?

If you are ready, you start with AI Foundations. If you are not, you fix the gaps first; usually documentation, team alignment, and data cleanup. Either way, the path from assessment to first deployment runs 60–90 days.

The assessment is only useful if it changes what you do on Monday. A score with no next move is trivia. The path forks cleanly based on what the five dimensions revealed.

• If you are ready: Begin the foundations work; context packs, decision rules, and the first proven workflows.
• If documentation is weak: Write down your core operations before any tool touches them.
• If the team is not aligned: Close the gap between the executive who is ready and everyone else.
• If data is messy: Clean and consolidate the customer and financial data your systems will depend on.
• Either way, the clock: Most companies reach a first deployment within 60–90 days of an honest assessment.

Companies that score well move quickly; this is how Phos builds AI foundations for companies that are ready, turning a strong assessment into a first deployment inside a single quarter.

An AI readiness assessment costs nothing. Getting it wrong costs months.

The five dimensions are scorable today, and the gaps they reveal are fixable in a quarter. Start with the Scorecard or the Voice Audit; learn your real score before committing a budget.

The companies that move first are not the ones with the best tools. They are the ones who knew exactly where they stood before they started.

Find out where your company stands before you spend a dollar

You have seen the five dimensions and where companies usually break. The next move is knowing your own score, not guessing at it; the Scorecard and Voice Audit give you that in under an hour.

Phos AI Labs turns AI strategy into running operations. We design the foundations, train your team inside real workflows, and rebuild the processes that matter most; until AI is not something your business uses occasionally, but how it actually runs.

• Strategy before systems: We establish what to build and what to leave alone before recommending a single tool.
• AI Foundations that hold: We install the operating manuals, context packs, and decision rules your team runs on for years.
• Training inside real work: We build fluency in your actual Slack, HubSpot, and QuickBooks workflows, not staged demos.
• Private AI Workspace: We design a shared company-wide AI environment built around your operations, knowledge, and team.
• Operations redesign: We rebuild the workflows that carry the most weight, from financial close to customer onboarding.
• Honest judgment, always: We tell you what will work and what will not, before you spend a dollar.
• We stay until it compounds: We are done when the business runs differently, not when the setup is complete.

400+ engagements. Clients include Zapier, Coca-Cola, Medtronic, Dataiku, and American Express.

If you want a readiness picture you can act on, see how Phos approaches this.

Frequently asked questions about AI readiness assessments

How long does an AI readiness assessment take?

The self-assessment takes minutes and the voice assessment runs about three. A formal audit inside the AI Foundations phase takes days, since it maps every workflow and data source in detail.

Can a founder’s personal AI use scale to the whole company?

Rarely on its own. One founder running AI well proves it works; spreading that to 85 employees needs documented workflows, training, and a shared workspace, which is exactly what a readiness assessment surfaces.

The owner wants results this quarter. Is that realistic?

Sometimes. A company that scores well on readiness can reach first deployment in 60–90 days. A company with weak documentation or messy data needs the fix work first, which a readiness assessment prices honestly.

How do we get skeptical senior partners on board?

Show them the score, not a sales pitch. A readiness assessment names concrete gaps in data, workflows, and tooling, which gives skeptical partners something specific to evaluate instead of a vague promise about AI.

Do we need to hire a firm to run a readiness assessment?

No. The Phos Scorecard and Voice Audit are free self-assessments that give you a real picture. A firm adds value later, when you want the formal audit and the build that follows it.

What does an AI readiness assessment cost?

The self-assessment and voice assessment cost nothing. The formal audit is part of a paid engagement; running the free version first tells you whether that engagement is worth scoping at all.

Which dimension do most companies score lowest on?

People, not technology. Team readiness and leadership alignment are where most $5M–$25M companies fall short. The tooling is rarely the blocker; the willingness to change how work gets done usually is.

What happens if we skip the assessment and just buy tools?

You pay for licenses your team does not adopt and workflows the tool cannot follow. The assessment is hours of work that prevents months of an expensive deployment going quietly unused.