The short answer
A paid AI readiness assessment typically costs $2,500 to $7,500 and takes two to three weeks. You are paying for a senior person to interview your team, look at your systems and data, and tell you which AI projects are worth doing first, what they would take, and which ones to skip.
If you are not ready to spend anything yet, start with Lumeor’s free online AI readiness assessment. It is a quick self-serve starting point that helps you decide whether a paid assessment is worth it.
What the free assessment covers and what it does not
The free online assessment is a self-serve check you complete on your own. It is useful for a quick, honest look at your starting point, such as how organized your data is, how much repetitive work your team does, and whether someone can own an AI project.
What it cannot do is see your actual processes. It does not know that your invoices arrive as scanned PDFs, that your CRM has three duplicate fields for the same thing, or that your practice management system only exposes its API through a partner program. Those details decide what a project costs and how long it takes, and they only come out in interviews and a hands-on systems check.
Where the money goes
| Activity | Share of effort | What happens |
|---|---|---|
| Kickoff and scoping | 5–10% | Agree goals, departments in scope and who to interview |
| Stakeholder interviews | 30–40% | Walk through daily work, bottlenecks and the tools each team uses |
| Systems and data review | 20–25% | Check API access, data quality and where documents live |
| Use case ranking and estimates | 20–25% | Score opportunities by value, effort and risk, with rough build costs |
| Report and readout | 10–15% | Written findings, roadmap and a live session with leadership |
The interviews and systems review are what separate a useful assessment from a generic slide deck. A list of “AI use cases for your industry” is free on the internet. A list ranked against your data, your systems and your staff is what you are paying for.
What a good assessment produces
- A ranked shortlist. Usually three to eight opportunities, each with the problem it solves, how it would work and a rough cost band. Common winners are document handling such as invoice processing, inbox triage and internal knowledge assistants.
- A data and access check. Which systems have usable APIs, which need a partner program or vendor approval, and which data needs cleanup before any AI can use it. Our guide on connecting an LLM to internal data covers what we look for.
- Clear “do not build” calls. Places where an off-the-shelf tool, a feature you already pay for, or a simple process change beats a custom build.
- Risks and policy notes. Data handling, vendor terms, and for healthcare, a reminder that any vendor touching patient data needs a signed BAA.
- A first project. One recommended starting point small enough to deliver in weeks, so the assessment leads to something real.
How to keep the cost down
- Take the free assessment first. It gives you a vocabulary for the conversation and shortens kickoff.
- Narrow the scope. Start with the one department where the pain is most obvious. You can assess the rest later.
- Line up interviewees in advance. Scheduling delays stretch a two-week engagement into four or more.
- Gather system access early. Read-only access or a screen share with the system owner saves hours of back and forth.
How Lumeor prices assessments
We price each assessment as a fixed fee after a short call to agree scope, departments and deliverables. The written scope lists who we will interview, which systems we will review and what you get at the end. If the free assessment and a conversation are enough to point you at a single obvious first project, we will tell you that instead of selling you a paid assessment. For broader strategy work beyond an assessment, see what AI consulting costs.