You know the moment. You open a website to check opening hours, and before you can read a single line, a box jumps up asking what you agree to. You click the biggest button, because you only wanted the hours.
We did that to our own visitors. One day we asked ourselves a simple question: do we really need this banner?
The question that started it
A banner is only there because something sits behind it. So we went looking for what sat behind ours. A tool to measure visits. Another to watch how pages load. Two advertising pixels, from Meta and Reddit. A service that supplied the banner and our legal pages.
Then we asked of each one: what do we actually learn from this, and is it worth asking every visitor for a favour?
For none of them was it worth it. So we took them all out. Google Analytics 4, Datadog RUM, the Meta and Reddit pixels and iubenda are gone from crewdle.com. Google Analytics is gone from our app and our admin console too.
We also found something we were not proud of. Some of our server code had been sending Google a Crewdle user id and an organization id. We deleted those calls. That one stung a little, and it made the decision easier.
What a visit looks like now
Picture a neighbour looking up your company on their phone. They land on crewdle.com. No banner appears. Nothing asks them to choose. The website sets no cookies at all.
Their visit is still counted, and we will say plainly how. A small script of our own tells a Crewdle server in Canada that a page was viewed. It leaves nothing on their phone.
To avoid counting one person ten times in a day, the server makes a visitor number from a random salt, the site, the IP address and the browser type. The salt changes every day and the old one is deleted. So tomorrow, that same neighbour is a stranger to us. We cannot follow them from day to day, even if we wanted to.
The IP address is used in memory to make that number and to guess an approximate place, such as country, region and city, from a free database called DB-IP Lite. It is never logged or stored. The stored rows have no column for a name, an email, an organization, an IP address or a browser type, and our tests check that.
We keep the raw counts for 13 months. The summaries built from them have no end date.
Testing without following you
We like to try two versions of a page and see which one helps people more. The usual way is a cookie, so you see the same version tomorrow. We pick the version on our server, from that same daily visitor number. Nothing is placed on your side. When someone signs up, their account records where they first came from and, for a test, its name and the version they saw.
What we still keep, and why we tell you
It would be easy to write a post that says we store nothing, anywhere. That would not be true, and you would deserve better.
On the website, there are no cookies. Your browser holds a few small settings for you: your language, how you like to read our blog, your choices on the 3D pages, and a form that failed to send, kept up to three days so it can try again. Where you came from on your first visit is remembered only until you close the tab.
In the Crewdle app, things differ. To sign in, the app needs cookies. They include one that remembers someone signed in from this browser, two that remember your theme and language, one for your session, and two that carry a Microsoft sign-in for about five minutes. Stripe also sets two when you pay. Our privacy page lists these and a few others from outside providers, such as reCAPTCHA on forms.
These are there to make the product work. They are not there to follow you around.
What we think
We think privacy should be the default, not a pop-up.
If a visit can be counted without leaving anything on your device, count it that way. If a test can be run without following a person, run it that way. If something truly needs your attention, we will put it in front of you in plain words, not behind a wall of buttons.
We are not claiming we track nothing, because we count visits. We are not giving a legal opinion, because how a rule applies to you is a question for you and your advisers. We are saying what we collect, how, and for how long, so you can judge for yourself.
Our website and platform are hosted in Canada (data residency). Each customer chooses Canada, the United States or the European Union as the place their data is stored.
For the wider picture, read AI and Quebec’s Law 25 and Private AI mode.
You can read the full details, item by item, on our Privacy page.
Ask about this article 
