There's no single sticker price for AI, because "AI consultancy" covers everything from a two-week strategy review to a fully built, integrated product. The good news: the cost is more predictable than most people expect once you know which model you're buying and what drives the number.
The main pricing models
Most AI work is priced in one of three ways. Which one fits depends on how clearly the work is defined.
| Model | Best for | Indicative range* |
|---|---|---|
| Fixed-price audit / strategy | A clear, scoped piece of work with a defined deliverable | € few k – low tens of k |
| MVP / pilot build | Proving one high-value use case in production | € low – mid tens of k |
| Time & materials / retainer | Ongoing build, support and iteration | Day rate or monthly |
*Indicative ranges you'll see in the European market — not a quote. Real figures depend entirely on scope. We give a fixed quote after a short, free scoping call.
What actually drives the price
- Scope — one use case is far cheaper than "transform the whole company".
- Data readiness — clean, accessible data lowers cost; messy or missing data adds prep work.
- Integration — connecting to your existing systems is often where the real effort sits.
- Compliance — EU AI Act and GDPR requirements add rigour (and value) for higher-risk use cases.
- Build vs. support — a one-off build costs less up front than an ongoing, maintained system.
How to keep the cost down
The single best way to control AI spend is to start small. Instead of committing to a big programme up front, scope one high-value use case, prove it works, and measure the return before you scale. That's exactly why our second conversation comes with a working MVP rather than a 60-page document — you see real value (and real cost) before the big investment.
What this looks like at Crux Digits
We start with a free consultation and a fixed-price AI Audit & Strategy when you need direction, or go straight to a scoped MVP when the use case is already clear. Every roadmap includes the projected ROI, so you can see the return before you commit.