What cookieless tracking is
Classic analytics drops a cookie with a random ID on the first visit and reads it on every later one. That is how it knows a visitor came back. Cookieless tracking skips the cookie. A common approach is to derive a visitor ID on the server from data in the request, mixed with a secret that changes every day, and to keep nothing in the browser.
Two forces push sites toward it. Privacy rules in many countries require consent before storing identifiers on a device, and many visitors decline. Browsers also cut cookie lifetimes: Safari's tracking prevention deletes cookies created in JavaScript, and other script-writable storage, after 7 days without user interaction with the site. A returning buyer can look brand new.
What you give up
- Returning visitors across days. A daily ID can't tell that Monday's visitor came back on Thursday, so unique visitors over a month run higher than reality.
- Long journeys. A first touch weeks before the purchase can't be joined to the payment by the ID alone.
- Nothing within a day. Visits, pageviews, sources and goals on the same day are counted as usual.
The journey gap is what matters for revenue. The fix is to carry the visitor ID into the moments that last: pass it to checkout so the payment records it, or link a user ID when someone signs in (cross-device tracking).
How to use cookieless tracking well
- Run cookieless by default and switch to a stored first-party ID only when the visitor consents.
- Pass the visitor ID to your checkout so payments link back to the visit that brought the customer.
- Identify signed-in users with your own stable user ID.
- Read returning-visitor and multi-day figures with the daily reset in mind.
Whether you need a consent banner depends on the law where your visitors live and everything else on your site. Ask your own counsel; a cookieless analytics script is one input, not the whole answer.
Common mistakes
- Assuming cookieless means anonymous. Request data can still be personal data; check how the tool derives and stores IDs.
- Comparing visitor counts with a cookie-based tool. Different ways of counting give different totals for the same traffic.
- Losing attribution at checkout. Without the visitor ID on the payment, revenue can't be credited to its source.
How MetricRaven tracks without cookies
MetricRaven works without consent by default. Before consent, the tracker uses a server-derived visitor ID that changes daily and stores nothing that identifies the visitor in the browser. When your cookie banner calls window.analytics.setConsent(true), it keeps a random first-party ID for up to a year, so visits on different days join up; withdrawing removes it.
Payments link to a visitor when checkout passes the visitor ID (for example Stripe Checkout) or the user signs in, so revenue reaches the visit that first brought the customer either way. See Consent & visitor IDs and the cookieless revenue analytics overview.
Questions
- How does cookieless tracking count visitors?
- Usually with an ID the analytics server derives from request data and a daily secret, so the same browser gets the same ID for one day and nothing is stored on the device.
- Is cookieless tracking less accurate?
- Within a day it counts visits and sources the same way. Across days it can't recognize returning visitors without a stored ID, so multi-day visitor totals run higher.
- Can I attribute revenue without cookies?
- Yes, if the visitor ID reaches the payment: pass it to your checkout, or link a user ID when the customer signs in. The payment is then credited to the visit that brought them.
- Do I still need a cookie banner?
- It depends on the laws that apply to you and on the other scripts on your site. Cookieless analytics removes one reason for a banner, not necessarily all of them.