AI maturity assessment: is your company ready?

An AI maturity assessment shows whether the company has what artificial intelligence needs to deliver results: a written process, organized data, prepared people, a baseline number and usage rules. It doesn't measure how much technology the company has. It measures whether AI will find a solid foundation or just speed up the disorganization.

The practical result is an answer to two questions: can the company implement AI yet? And, if so, which process should it start with? This article provides a framework for running this assessment with your own team, focused on the areas where revenue happens: sales, customer service and marketing. The full implementation path is in the guide to implementing AI in your business.

What does the assessment evaluate, and what doesn't it?

It's common to confuse AI maturity with technical sophistication. Subscribing to several tools, or having someone on the team who understands language models, doesn't make the company ready. Often the opposite: scattered tools create more places where information gets lost.

The assessment looks at what AI will need day to day:

a clear process to carry out;
reliable data to consult;
people who know how to work alongside it;
a metric to prove it worked;
rules about what it can and can't do.

If one of these pieces is missing, the AI may work in the demo. But it stalls when it meets the real operation.

The five dimensions of AI maturity

Each dimension has a central question, a sign that the company is ready and a warning sign:

Dimension	Central question	Sign it's ready	Warning sign
Process	Is the work designed?	Funnel stages and qualification criteria written down	"Every rep does it their own way"
Data	Is the information in one place?	Customer history in the CRM, with stages kept up to date	Conversations on each person's personal phone
People	Will the team use it?	Manager involved and someone with time to own it	Project pushed from the top, with no owner
Measurement	Do we know today's number?	Response time and conversion are known	"I think we convert well"
Rules	Do we know what AI can do?	Usage policy and limits defined	Everyone pastes whatever they want into any tool

The weakest dimension defines the starting point. Great data doesn't help if nobody knows which process the AI will carry out.

The four maturity levels

After looking at the dimensions, a company usually fits into one of these levels:

Level 1: ad hoc use

People use AI tools on their own, with no rules and no measurement. The sales process depends on whoever is handling the customer. What to do: write down the process and the usage policy before any project.

Level 2: organized foundation

The funnel is designed, the CRM is used and the company knows its main numbers. There's no AI running processes yet. What to do: choose the first process and roll it out with a measured pilot.

Level 3: AI in one process

One process already runs with AI, with written rules, handoff to people and a tracked metric. What to do: consolidate the operation and choose the second process based on the results of the first.

Level 4: AI integrated into operations

Several processes use AI and share the same CRM and the same data. What AI learns in customer service reaches the sales rep and flows back to marketing. What to do: regularly review the rules, the knowledge base and the metrics, and adjust the policy as the operation changes.

Skipping levels is the most common way for a project to stall. A Level 1 company that hires an AI agent for WhatsApp will spend most of the project discovering it has no qualification criteria.

How to run the assessment in practice

1. Get the right people together

The owner or director, the sales manager and whoever is responsible for customer service and marketing. Each one sees part of the funnel. A short conversation is enough for the first round.

2. Answer the questionnaire honestly

For each question below, the answer is "yes" or "no." "Sort of" counts as no.

Process

Does the sales funnel have defined stages, with entry criteria for each one?
Is there a written criterion for what a qualified lead is?
Is it clear when a conversation moves from customer service to the sales rep?

Data

Is the history of conversations and deals kept in the CRM, and not on each person's phone?
Are stages and loss reasons filled in regularly?
Is the information the AI would use (prices, policies, FAQs) written down and up to date?

People

Is there one person with the time and authority to own the project?
Does the sales manager support the change?
Has the team received any guidance on using AI?

Measurement

Does the company know the average first response time to leads?
Is the conversion rate per stage known?
Are leads linked to their source campaign or channel?

Rules

Is there a policy on which data can go into AI tools?
Is it defined how customers will be told they're talking to a virtual assistant?
Is it clear who is accountable for what the AI does in each process?

3. Map where the funnel loses the most

With the questionnaire done, sketch the funnel on a sheet of paper and mark where leads wait, where reps repeat work and where information gets lost. Those points are the candidates for the first AI process.

4. Choose the first process

Cross-reference the candidates with the dimensions. The first process should have volume, a clear rule, low risk and a metric the company can already measure.

Take Nexo's free assessment and find out which stage of your funnel AI pays off in first.
Take the free assessment

How to read the results

Count how many "yes" answers each dimension got. Reading them is simple:

Almost all dimensions with "yes": the company can start a pilot in the chosen process.
One weak dimension: fix that dimension in parallel with planning the pilot. If it's measurement, start measuring now.
Two or more weak dimensions: the first project isn't AI. It's getting sales organized.

There's no universal cutoff score. The goal isn't to reach a number, but to know which piece is missing before spending on tools.

What to do when the company isn't ready yet

Not being ready is a useful result. It prevents a project that was going to stall. The most common fixes are these:

Process: write down the funnel and the qualification criteria together with the sales team.
Data: put the CRM to real use, with stages and loss reasons filled in. When the current CRM doesn't support the process, CRM consulting helps decide between fixing it and replacing it.
People: name an owner and prepare the team. The plan is in how to train your team to use AI.
Measurement: start recording response time and stage conversion, even in a simple way.
Rules: write a short usage policy. A template is in AI usage policy for your company.

Repeat the assessment after these fixes. The result usually changes faster than the owner imagines, because much of the work is deciding and writing, not buying.

Common assessment mistakes

Answering based on how the company should be, not how it is.
Letting only the owner answer, without the people in the operation.
Confusing having tools with being ready.
Evaluating the entire company at once instead of looking at the sales funnel.
Running the assessment and not choosing a process at the end.

Frequently asked questions

What is AI maturity?

It's how well the company has the conditions for AI to deliver results: a written process, organized data, prepared people, known metrics and usage rules. It has nothing to do with how many tools the company subscribes to.

Does a small company need an assessment before using AI?

Yes, but it can be simple. A conversation with the owner and whoever runs sales and customer service, using a short questionnaire, already shows whether a piece is missing before the first project.

What should I do if the company is at the lowest level?

Write down the sales process and the AI usage policy before any project. With the funnel designed and the CRM in use, the company moves up a level and can start a measured pilot.

Which dimension is usually the weakest?

It varies from company to company. Process and measurement come up often, because many sales operations depend on whoever is handling the customer and don't track conversion by stage.

Conclusion

An AI maturity assessment isn't meant to grade the company. It's meant to show which piece is missing and which funnel process deserves the first AI. Look at process, data, people, measurement and rules, start with the weakest dimension and choose a sales or customer service process with a known metric.

Want to know what's missing for AI to deliver results in your sales team? Take Nexo's free assessment.
Take the free assessment

Once the foundation is ready, see how Nexo runs AI implementation for businesses, from assessment to measuring results in sales.

AI maturity assessment: how to tell whether your company is ready