What we keep about you, and why

LSVH works from Harderwijk and decides what happens to your data. There is no office to visit, so mail is the quickest way to reach us. There is a button for it at the bottom of this page.

Only what you fill in yourself: your name, your company, your email address, your phone number if you give one, your description of your situation and your answers to the questions.

When you send the request off, we record it with a reference against it: your answers, your description and the range it came to. That sits in a Cloudflare database (D1) rather than only in our mailbox, so we can find your request again when you write to us about it later. The reference is in your confirmation.

If you accept a quote, we record when that happened, which version it was and the IP address it came from. That is the evidence of the agreement — as much yours as ours. At that point we ask for your billing address, because an invoice without the customer's name and address is not a valid one; we use those details for nothing else. If you ask something through your own page, we keep that question with the request.

We buy nothing in, we top nothing up from other sources and we build no profile of you.

To answer your request and put a price against it. That is preparation for an agreement, and it is also the basis for keeping it: without those details we cannot send you anything.

Cloudflare also counts requests per IP address to stop abuse. That is about numbers rather than about who you are, and it is how the chatbot stays standing without a tracking script next to it.

A request that leads to nothing is kept for eighteen months. After that an hourly job strips your name, your company, your address and your description out of it, and the links stop working. What is left is that a request existed and what became of it, without who it was.

If it becomes work, it comes with records we are required to keep for seven years. That is the law rather than a choice of ours.

Four parties, each for one thing. Cloudflare puts the site online, keeps it there and holds your request in D1; its server logs include your IP address. Cloudflare Workers AI runs the chatbot: what you type goes there to produce the next question, is not used to train models and is not retained. Resend delivers the mail carrying your request to our mailbox. And if you accept a quote, the work becomes a task in our own private GitHub repository: what has to be built, your description and your email address, so whoever works on it can reach you.

Nobody else. No advertising networks, no analytics package, and we never sell anything on.

The site sets none at all. So there is nothing to ask permission for, which is why no bar stands between you and the page.

The chatbot does remember what you filled in, but inside your own browser and only while the tab is open. Close it and it is gone. We cannot reach it.

For each question we send a short summary of what you have said so far and your last two messages to the language model — not everything you ever typed. That limits what travels and what could leak.

The model never names an amount. The price comes out of a calculation on our own rates; the chatbot only fills in the fields that calculation needs.

You can ask what we hold about you, have it corrected, have it deleted, or object to what we do with it. One message is enough; no form and no reason are needed. We answer within a month.

If you think we are getting it wrong you can complain to the Dutch data protection authority. We would rather hear it first, because then we can put it right.

If something on the site changes such that different data passes through it, this page changes with it. The date below comes from this file's own history, so it cannot fall behind what it describes.