Article
What does an AI consultant do?
An AI consultant works out where AI helps in your business, and where it does not. An implementation partner builds it. Here is how it works.
In short
An AI consultant looks at how your business works and advises where AI helps, and where it does not. An implementation partner builds and integrates the solution. Some consultants do both. When you choose, pay attention to ownership, privacy and cost.
What does an AI consultant do?
An AI consultant works out where AI helps in your business, and where it does not. They look at how the work is done now, which systems you use and where a lot of time goes. Then they advise what you can tackle first, and what you are better off leaving alone.
That sounds simple, but it takes knowledge of two sides. On one side, you need to understand what AI can do and where it goes wrong. On the other side, you need to understand how a business works: who does what, where the data is kept and which rules apply.
The title AI consultant does not say much on its own. One mainly gives advice. Another also builds. So ask exactly what someone does.
What do you get from an AI consultant?
You get a clear picture of where AI fits in your business, with a plan to get started. Usually that consists of:
- An overview of your main processes: from enquiry to invoice, from question to answer.
- A list of opportunities: which work is suitable for AI or automation, and in what order.
- What is not suitable, and why.
- The rules that come with it, such as the GDPR and the EU AI Act.
- A proposal with costs, so you can decide.
Good advice can also be: you do not need AI here. Sometimes a better spreadsheet, a standard integration or a clear working agreement is enough.
What is an AI implementation partner?
An AI implementation partner builds and puts in place what has been designed. They connect systems, set up AI tools, test with real data and explain how it works. After delivery they often also take care of maintenance.
Where the consultant answers the question of what you should do, the implementation partner answers the question of how it will work. An implementation partner builds, for example, an API integration, an AI assistant on your own documents or an AI agent that carries out a task.
What is the difference between an AI consultant and an implementation partner?
The difference is advice versus execution. Side by side:
| What | AI consultant | Implementation partner |
|---|---|---|
| Role | Investigates and advises | Builds, integrates and tests |
| You get | An analysis and a plan | Working software or settings |
| Main question | What do we do, and why? | How do we get it working? |
| Useful if | You do not know where to start | You know what you want and are looking for someone to make it |
If you work with two parties, the advice has to be handed over to the builder. That can go well, but knowledge can also get lost. Not everything the consultant heard in conversations ends up in the report.
Why do I do both?
I am an AI consultant and I build the solution myself. First I do the analysis and the proposal. Then you decide whether I also build it for you. That way you talk to the same person from analysis to delivery. That is also why I call myself an AI agency (in Dutch).
The advantage is that when I build, I know what was said in the analysis. I do not have to hand anything over to another team. My advice is about what I can make and maintain myself, in the systems you already have. If something does not fit there, I say so.
The proposal is yours. Even if you then do it yourself, or have someone else build it. Read how that works in my way of working.
When is someone else a better fit?
Someone else is a better fit if your project is bigger or more specialised than one person can carry. A few examples:
- A whole team needs to work on it at the same time. For example when introducing a new ERP system across many locations. I am a one-person business.
- You have your own IT department that builds. Then you may only need advice, or only a training.
- It involves AI that assesses people. For example in recruitment or in granting credit. The GDPR and the EU AI Act set extra requirements for that. In that case, bring in a lawyer as well.
- You only want a subscription to an AI tool. Without an integration or custom work, you can often go to the vendor directly.
When do you need an AI consultant?
Hiring an AI consultant is useful if you want to start with AI, but do not know where. Do you recognise one of these situations?
- Your team already uses AI tools, but there are no agreements about them. Read Your team uses ChatGPT without agreements.
- You read a lot about AI, but do not know what it means for your business. Read You do not know where to start with AI.
- There is a lot of recurring work with text: quotes, emails, reports or reading documents. See for example making quotes with AI.
- Vendors are offering you AI features, and you want an independent view.
How does a project with an AI consultant work?
A project starts with looking and listening, and only then with technology. With me it goes like this:
- Introduction. A first conversation about your business and your question. Afterwards you know whether it is a fit.
- Analysis. I visit, talk to the people who do the work and look at your systems and data. Where is the repetitive work? Where does it go wrong?
- Proposal. You get a short written proposal: what can be improved, what it costs and where you start.
- Implementation. If you choose to go ahead, I build one improvement and integrate it with your systems. A person keeps checking what the AI does.
- Explanation and maintenance. I explain to someone in your team how it works. After that I maintain it, if you want me to.
You decide after each step whether to continue. Start small, with one process. If that works well, you choose the next step.
What does an AI consultant cost?
That differs per consultant and per assignment. With me the first conversation is free. After that I do a paid business analysis. In it I visit, talk to your people and look at your systems and data. Then you get a proposal with a fixed price for the next stage.
Do you then want it built? Then I build one improvement, integrate it with your systems and explain it to a colleague. What that costs is stated in the proposal in advance, so you know where you stand. Would you like rules and training for your team first? Then you start with the AI policy & training session, in half a day. Also read how your price is put together.
Other consultants charge differently. Ask every party in advance what the analysis costs, what you get afterwards and whether the next stage has a fixed price.
Do you become dependent on an AI consultant?
You do not have to, if you make good agreements about it in advance. It comes down to three things:
- Ownership. Who owns the code, the settings and the data? With me, they are yours. If you stop the maintenance, you take the code and your data with you.
- Explanation. Someone in your team needs to know how it works. I explain it to a colleague after delivery.
- Accounts. Put subscriptions to AI tools your team uses in the name of your own company as much as possible, with a business account.
Does a consultant work with personal data of your customers or employees? Then you sign a data processing agreement together. The GDPR requires this when another party processes personal data on your behalf.
Does an AI consultant also help with the rules?
Yes, rules are part of good AI advice. Two laws often play a role: the GDPR for personal data, and the European AI Act.
Article 4 of the EU AI Act is about AI literacy. It has applied since 2 February 2025 to businesses that use AI. Since an amendment that came into force on 27 July 2026, it says: you take measures to support the AI literacy of your people. You do not have to guarantee a particular level per person. Think of explanation, practice and clear agreements, set down in an AI policy.
Want to arrange this with your team? Read more about AI training and rules.
This is an explanation in plain language, not legal advice.
Which questions do you ask in the first meeting?
With these questions you can quickly see whether an AI consultant suits you:
- Do you only give advice, or do you build it too? And if so, yourself or through someone else?
- Do you receive commission from software vendors?
- Who owns the code, the settings and the data if we stop?
- How do you handle personal data? Will we sign a data processing agreement?
- What does the analysis cost, and will I get a fixed price afterwards?
- Can you explain what AI is not good at in our case?
Bring something yourself too: the three tasks that take the most time, a list of your systems and who in your team already uses AI. The first conversation with me is free.
How do I help you as an AI consultant?
I look at where AI helps in your business, and where it does not. Then I build it myself, in the systems you already use. I work from Ede, for businesses across the Netherlands with 10 to 250 employees. Read more about AI consultancy, or book a conversation.
Frequently asked questions
What is an AI implementation consultant?
Someone who advises on AI and also helps with putting it in place. The term sits between consultant and implementation partner. Ask what someone does themselves: advise, build, or both.
Can I get just the advice?
Yes. After the business analysis you get a proposal. It is yours, even if you carry it out yourself or have someone else build it.
How long does a project take?
With me, the recommended length of the analysis is one week. For implementation I allow four to six weeks. Exactly how long it takes is stated in the proposal.
Do you need to use AI already before you bring in an AI consultant?
No. You can also start without any experience. Then the analysis starts with how your business works now, not with the technology.
Does an AI consultant ever advise against AI?
A good one does. Sometimes an ordinary integration, a better template or a clear working agreement is enough. That is also part of honest advice.
Sources: Regulation (EU) 2026/1744 (Digital omnibus on AI) (in Dutch), which amends Article 4 of the AI Act (EU) 2024/1689, and Dutch Data Protection Authority (Autoriteit Persoonsgegevens): Data processing agreement (in Dutch). Consulted on 7 October 2026.
Written by Bart van Engelenhoven Founder of Uniekaiconsultancy in Ede, the Netherlands. Bart analyses how SMEs work and builds software, integrations and AI applications himself. Last updated: 23 October 2026
Read more
Article · 15 min read
How do you map your processes before you automate?
Mapping processes before you automate: which process to choose, what to note per step and how to see what software, AI or a person does.
Read moreSound familiar?
You don't know where to start with AI
Everyone talks about AI, but what's in it for you? Start with your own work, not the tool. How to choose a small first step that suits your business.
How to tackle itTerm
AI policy
An AI policy is your company's set of agreements on using AI: which tools, which data and who checks. What goes in it, in brief.
Read moreNext step
Curious what could work for you?
- The first conversation is free and without obligation
- By phone or at your premises
- You speak directly with me
Bart van Engelenhoven You tell me what takes up your time, I tell you honestly whether I can help.