AI for Expat Tax Firms in the Netherlands
The structured work hidden inside an advisory business
Expat tax advisory looks like judgment work. In practice, a significant share of each engagement is structured, repeatable, and document-heavy: collecting the same set of documents from each assignee, checking completeness, cross-referencing employment contracts against Dutch tax code conditions, assembling a packet the advisor then reviews.
That gap — between the judgment your clients pay for and the structured work that consumes your team's time — is where AI is built to operate.
What AI can and cannot do in this context
Where AI is well-suited:
- Checking whether a submitted document set is complete against a defined checklist
- Extracting key fields from employment contracts (start date, gross salary, expected stay, country of origin) and flagging where they fall outside standard 30%-ruling eligibility criteria
- Drafting a structured packet summary the advisor reviews and approves — rather than assembling it from scratch
- Generating a client-facing status update from intake form data, without the advisor writing it manually
Where human judgment stays essential:
- Interpreting edge cases — the assignee whose situation doesn't fit the standard eligibility pattern
- Advising on the decision to apply (not just the mechanics of the application)
- Signing off on every client-facing output before it leaves your office
- Managing the RVO relationship and any review process that follows
The practical result: the structured work that currently falls to a junior advisor or office manager can be handled by an AI system that runs the checklist, extracts the fields, and prepares a draft for human review. The advisor's time shifts toward the work that requires them.
How Orellis approaches this
Orellis builds and deploys AI systems for Dutch SMEs. In the expat tax context, that means:
- Operations mapping. We walk through your current 30%-ruling intake and packet-prep workflow — who does each step, how long it takes, where errors occur.
- Compliance and risk screen. Any task touching personal data or regulated advice requires documented human-in-the-loop. We identify which steps in your workflow qualify and which can operate autonomously.
- First implementation, shipped during the engagement. Not a recommendation document — a working system. The advisor team can run it solo by the end of the first week.
- Handoff and runbook. A one-page operating guide: how to run it, how to stop it, who owns it, when to review it.
We build on infrastructure designed for EU data residency, with an audit trail of every prompt and output. Every AI-generated output in a client-facing context includes a disclosure. We are not a legal advisor; the compliance screen is a starting point, not a legal filing.
Is this the right fit for your firm?
This approach works best for firms that have:
- A defined, repeating intake process (not every engagement is bespoke from the start)
- An advisor or operations lead who is willing to be the named "operator" for the system
- Enough volume that the structured prep work is a real time cost — typically 20+ active assignee cases at any point
If your firm handles 5 cases a year, the setup cost isn't justified. If you're handling 30 or more, the math changes.
Tell us one thing.
What is the single most time-consuming structured task in your 30%-ruling or expat tax intake process? Send one line to rachel@orellis.ai — we'll tell you whether AI is the right tool for it, at no cost.