AI process audit
€2,500
credited
We record how your processes actually run, rank the candidates, and check honestly whether the data is reachable. You get a document with a plan and prices.
- 2-3 weeks
- 3-5 ranked candidates
- Data access checked
AI at work
We bring AI into your processes.
We automate one process at a time, connect it to the systems you already run, and stay responsible for it afterwards - so nobody in your company has to become an AI expert.
Ready when you are.
- ChatGPT
- Claude
- Gemini
- Mistral
- DeepSeek
- Perplexity
Where things stand
The technology can do more than reaches most companies.
Between an assistant somebody operates and a process that runs on its own lie two further stages. Most companies are on the first one and take it for the end of the road.
- 01
The assistant at the desk
Someone asks, the model answers. Useful for whoever is sitting there - and over the moment they close the window. Almost every company already pays for this through its Microsoft or team subscription.
- 02
The process that runs on its own
An event starts it - an email, an invoice, an appointment. It reads from your systems and writes back into them, logged, within a defined scope. A person signs off anything binding. It runs at night too.
- 03
The system across departments
Several processes interlock, agents work on a task over hours and days, fetch what they need and come back when a decision is due. People lead and supervise instead of typing.
The question is rarely whether a process can be automated. It is how much of it is worth automating, and in what order - which is exactly what the audit answers. We do not deploy finished tools that somebody else built AI into: we build the system your process needs. That is why the limit is almost never the technology.
What actually works in practiceWhy it stays undone
Two gaps, and the second one weighs more.
01
Nobody in the company is there for it
Between a possibility and a process that runs every day sits work: sorting out access, connecting systems, writing test cases, keeping it running. Nobody has time for that alongside the day job, and the role is hard to hire - whoever could fill it usually works somewhere else.
02
And before that: nobody knows what to do
Which process justifies the effort at all, and what would a system have to do for something to genuinely change? That question needs two viewpoints at once - your own process with everything that has grown into it, and where the technology stands. The two rarely meet in one person. So it usually stops at a tool subscription.
We close both. The audit works out what is worth doing and in what order; then we build it and keep it running. No handover in between to somebody who has to read themselves in again.
Why us
We work with these systems ourselves - every day.
What you get at the end is not a slide deck or a training handout, but a process that runs.
- Current models
- We build with what is available today, not with last year. Where these models are strong and where they tip over is something we know from working with them.
- Integration
- The value appears where the system hangs off your ERP, inventory, mailbox and file storage. That is the most demanding part of a project - and our craft.
- Our own systems in production
- We build and run our own software. What a system needs after nine months, we learned from operating one, not from a study.
- Building included
- We advise which process is worth automating - and then build it. Same people, no handover to a third party.
In practice
Where AI already takes work off the table.
Eight processes that occur in almost every company. Each page shows how it really runs today, which systems are involved, what a person still signs off - and where it is not worth doing.
01
Quotations
An enquiry and a drawing become a costed quotation.
Mechanical engineering · Trades · Wholesale02Invoice processing
Read documents, match them, check them, post them cleanly.
Wholesale · Logistics · Tax advisory03Customer enquiries
First qualification and a suggested appointment, before anyone picks up the phone.
Trades · Professional services · Construction & building services04Document search
Answers out of records nobody can find any more.
Mechanical engineering · Industrial suppliers · Law firms05Shared inbox
Sort, match, prepare - before anyone reads.
Wholesale · Logistics · Professional services06Tenders
Evaluate bills of quantities before anyone starts costing.
Construction & building services · Mechanical engineering · Professional services07System integration
Move data between systems without copy and paste.
Wholesale · Retail · Industrial suppliers08Reports and analyses
Recurring compilations nobody enjoys producing.
All sectorsWork out what is worth doing - then build it, then run it.
Three stagesProcess
from €6,900
each further one from €4,900
One purpose, one automation, properly embedded. The systems involved are connected, the scope of action is defined, your people are shown how it works.
- Systems connected
- Autonomy fixed in writing
- Signed off against test cases
System
from €45,000
after an audit only
For when the company is to be rebuilt around AI. Across departments, with purpose-built tools wherever no interface exists.
- Across departments
- Roles and permissions
- Own infrastructure
Every price says "from" - the scope decides. After the audit there is a fixed price; before it, any number would be a guess.
Operations
An AI system without maintenance is broken within nine months.
Models change, interfaces disappear, prompts drift, processes move. We watch, keep up and keep developing - so the system still does in a year what you built it for.
- Carefrom €290/month per process
- Monitoring, model and API updates, fixing what breaks.
- Managed€1,490/month
- A named contact, continuous development, reporting.
How a project runs
- 01
Audit
Record the processes, rank the candidates, check data access.
- 02
Build
Build the system, connect it, define its scope, test it.
- 03
Sign-off
Checkable criteria instead of promised effects.
- 04
Operations
Monitor, keep up, keep developing.
From our insights
The bottleneck is never the building
An agent that drafts quotations is built in two days. The question of whether it can reach the data it needs inside your business costs weeks. Almost every failed AI rollout fails on the second question, having started with the first.
Read the pieceWhich process eats the most time at your company?
Tell us about it. In the audit we record it, work out what is worth automating, and you get a plan with fixed prices - credited against the build if you go ahead.
Request a process audit€2,500, credited