Artificial intelligence is becoming easier for Kenyan businesses to access. A small company can now use AI to draft customer replies, analyse spreadsheets, summarise documents, prepare marketing material, automate repetitive work and help staff search internal information.

The harder question is no longer whether AI is available. It is whether your business is ready to use it well.

A company can pay for ChatGPT, Gemini, Microsoft Copilot or another AI service tomorrow and still get little value from it. The problem may be poor data, no clear business goal, employees who do not know how to review AI output, weak privacy controls or no way to measure whether the tool is actually saving money.

That is why we built a free AI Readiness Calculator for Kenyan businesses.

Take the free AI Readiness Calculator →

It takes about three minutes, has 18 questions and scores your business across six areas: Strategy, People, Data, Governance, Technology and ROI. This guide explains what those six areas mean and what to do with your result.

Why AI readiness matters in Kenya

Kenya is already treating AI as an economic and business issue rather than a distant technology trend.

The Kenya AI Strategy 2025–2030 sets out a national direction for AI, while the government’s implementation roadmap identifies MSMEs among the priority sectors for AI use. The roadmap also emphasises digital infrastructure, data, talent, governance, investment and responsible AI adoption.

A company can have excellent AI tools and still fail because its records are disorganised, employees do not trust the system, nobody owns the project or customer information is being handled carelessly.

For a Kenyan SME, the practical lesson is simple:

Do not start with the AI tool. Start with the business problem.

What does AI readiness actually mean?

AI readiness is the ability of a business to adopt AI in a way that is useful, measurable and manageable. It does not mean you need a data-science department, expensive servers or your own AI model.

A five-person business can be ready for a simple AI pilot while a much larger company can be unready because its data and processes are a mess.

Our assessment looks at six areas.

Readiness area What it asks
Strategy Do you know what business problem AI is supposed to solve?
People Can employees use AI, check its work and adapt their workflows?
Data Is the information AI needs accurate, accessible and safe to use?
Governance Are privacy, security and human-review rules in place?
Technology Can your current systems support a low-risk AI pilot?
ROI Can you measure whether AI saves time, cuts costs or increases revenue?

A weakness in one area does not always stop a project. But some weaknesses matter more than others.

For example, a business with excellent internet, modern computers and enthusiastic employees may still not be ready to put customer records through an AI workflow if it has no rules for handling personal information.

That is why the calculator does not rely on a simple average alone.

1. Strategy: What exactly should AI improve?

“Use AI in the business” is not a strategy. A useful starting point is a specific process with a measurable problem.

For example:

  • Customer-service staff spend three hours a day answering repetitive questions.
  • A salesperson spends every Monday preparing the same weekly report.
  • Staff manually classify hundreds of invoices.
  • A business takes two days to turn meeting notes into client proposals.
  • A retailer has years of sales data but does not use it to forecast demand.

These are possible AI use cases because the current problem is clear.

Before buying anything, write down:

The task, the current cost or delay, what you want AI to improve, and how you will measure the result.

If you cannot explain the problem without mentioning the name of an AI product, the project is probably still too vague.

2. People: Can your staff actually use the tool?

AI adoption often looks like a technology project, but employees determine whether it sticks.

Someone still has to know what to ask the tool, provide useful context, recognise a bad answer and decide when human judgement is required. Staff do not all need to become AI experts. They need skills relevant to their jobs.

For a customer-service team, that might mean learning how to draft replies from approved information and verify them before sending.

For an accountant, it might mean knowing which financial records can be analysed with an AI tool and how to check calculations.

For a marketing team, it might mean using AI for research and drafts without blindly publishing invented statistics or claims.

The goal is not simply to make staff “good at prompts.” It is to make them good at using AI inside a real workflow.

3. Data: Is your business information usable?

AI becomes much more useful when a business has reliable information to work with.

That can be a problem for smaller companies where records may be spread across:

  • WhatsApp conversations;
  • employee laptops;
  • Google Sheets;
  • paper files;
  • email inboxes;
  • accounting software;
  • point-of-sale systems; and
  • several versions of the same document.

Before attempting a complex AI project, ask whether the information is complete, current and accessible.

If your stock records are wrong, an AI demand forecast will not magically fix them.

If your customer records contain duplicates, missing fields and outdated phone numbers, automating the process may simply automate the errors.

Sometimes the best first step toward AI adoption is not buying an AI tool. It is cleaning up the information you already have.

4. Governance: What information are employees allowed to give AI?

This is one of the most important readiness areas for Kenyan businesses.

Kenya’s Office of the Data Protection Commissioner has specifically highlighted privacy risks linked to AI, including increased collection of personal data, lack of informed consent and unauthorised use of data. The ODPC has also stressed the need for appropriate safeguards and Data Protection Impact Assessments where necessary.

That does not mean a business should be scared of using AI, it means employees need rules.

At minimum, a business should decide:

  • which AI tools are approved for work;
  • what customer or employee information may not be entered into public tools;
  • how confidential company information should be handled;
  • when AI output must be reviewed by a person;
  • who is responsible when an AI-assisted decision affects a customer; and
  • what happens when an employee discovers an AI error or data incident.

A simple one-page internal AI policy is better than having ten employees invent their own rules.

If your business processes personal data, read the ODPC guidance on data protection and AI and consider whether you need specialist advice for your specific use case.

5. Technology: Can you run a small pilot without breaking anything?

Being AI-ready does not require rebuilding your entire technology stack. In many cases, the best first pilot is deliberately boring.

You might test AI on:

  • drafting internal reports;
  • summarising non-sensitive documents;
  • turning meeting notes into action lists;
  • categorising customer enquiries;
  • preparing first drafts of product descriptions;
  • extracting information from approved documents; or
  • helping staff search a controlled internal knowledge base.

A useful pilot should be small enough to stop if it fails. That is very different from connecting an untested AI system directly to payments, customer accounts, payroll or other critical systems.

6. ROI: Is AI actually making the business better?

AI demos can be impressive, but business results are more important. Before a pilot starts, measure the current process.

Suppose an employee spends eight hours a month preparing a recurring report. After introducing AI, the same report takes three hours.

You have saved five hours per month.

Now compare the value of those hours with:

  • the AI subscription;
  • staff training time;
  • integration costs;
  • review time; and
  • any additional software required.

ROI does not always have to mean direct cash savings.

A useful pilot could also improve:

  • response time;
  • sales conversion;
  • error rates;
  • customer retention;
  • output per employee; or
  • turnaround time.

But there should be a number somewhere. If six months pass and nobody can explain what the AI system improved, the business should question why it is still paying for it.

What your AI readiness score means

The AI Readiness Calculator gives a score from 0 to 100, but the number is only part of the result.

It also checks whether important foundations are strong enough for the next stage.

The five stages are:

Build the foundations

Your business should first improve its basic digital processes, data, skills or governance. Buying more AI software is unlikely to solve the underlying gaps.

Foundation stage

You have some useful building blocks, but important weaknesses could make an AI project difficult to manage or measure.

Pilot preparation

You are close. The next step is to fix the specific readiness blockers identified in your result and define one measurable pilot.

Pilot ready

Your business has enough of the right foundations to test AI in a focused, low-risk use case. The goal is not to deploy AI everywhere. It is to run a controlled pilot, measure the result and learn.

Scale ready

Your business has stronger foundations for expanding AI use cases that have already shown value. Even at this stage, scaling should follow proven results rather than enthusiasm.

A high score does not automatically produce a high readiness level. Weak Strategy, Data or Governance scores, and critical safety gaps, can prevent a business from being marked Pilot Ready or Scale Ready.

A simple 90-day AI adoption plan

If your assessment says you are ready to move forward, you do not need a five-year AI strategy.

Start with 90 days.

Days 1–30: Pick the problem

Choose one process where the current time, cost or error rate can be measured.

Assign one person to own the pilot.

Write down what information the AI tool may use and what it may not use.

Record your baseline before changing the workflow.

Days 31–60: Run a controlled pilot

Use the tool with a small number of employees or on a limited set of tasks.

Require human review.

Track failures as seriously as successes.

Ask employees what is actually happening in the workflow rather than relying only on the vendor’s dashboard.

Days 61–90: Measure the result

Compare the pilot with your baseline.

Did it reduce hours?

Did output improve?

Did errors increase?

Did customers notice an improvement?

Did staff actually use it?

Then make one of three decisions:

Scale it, improve it or stop it.

Stopping a weak AI project is not failure. Continuing to pay for one that creates no measurable value is.

AI readiness is not the same for every business

A business does not become “AI ready” once and stay that way forever. You may be ready to use AI for drafting marketing copy but not ready to let it analyse sensitive employee records.

You may be ready to automate internal reports but not customer-facing decisions. Readiness depends on the use case.

That is why our calculator is best treated as a practical starting point, not a compliance certificate or an independently validated industry benchmark.

It is designed to answer a more useful question:

What should this business fix or test next?

If you have not taken the assessment yet, start there.

Check your business’s AI readiness now →

If you are exploring AI as an individual rather than implementing it inside an existing company, you may also want to read our guide on how to make money with AI in Kenya.

FAQs

What is an AI readiness assessment?

An AI readiness assessment checks whether a business has the strategy, people, data, governance, technology and measurement systems needed to adopt AI effectively. It helps identify gaps before the business spends heavily on AI tools or automation.

How do I know if my business is ready for AI?

A business is more likely to be ready for an AI pilot when it has a clear use case, usable data, staff who can work with the tool, basic privacy and security controls, suitable technology and a way to measure whether the pilot creates value.

Do Kenyan businesses need an AI policy?

There is no single AI policy template that every Kenyan business must use, but businesses should set clear internal rules for AI use, especially where employees handle customer information, employee data, confidential records or other personal data.

Can employees put customer information into ChatGPT or other AI tools?

Businesses should not assume that customer or employee information can safely be entered into a public AI tool. Before personal or confidential data is used, the business should consider its data-protection obligations, the tool's terms, access controls and whether the information is necessary for the task.

What score do I need to be AI ready?

There is no universal AI readiness score. The Kenya MMF Calculator assessment uses a 0–100 diagnostic with readiness gates. A high overall score is not enough on its own if important data, governance or strategy gaps remain.

Is the AI Readiness Calculator free?

Yes. The Kenya MMF Calculator AI Readiness Calculator is free to use, requires no login and processes the assessment in your browser. Your answers are not sent to or stored by the calculator.