10000 Feature request: Improve cookieless tracking · Issue #25117 · PostHog/posthog · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Feature request: Improve cookieless tracking #25117

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
robbie-c opened this issue Sep 20, 2024 · 5 comments
Open

Feature request: Improve cookieless tracking #25117

robbie-c opened this issue Sep 20, 2024 · 5 comments
Labels
enhancement New feature or request team/web-analytics

Comments

@robbie-c
Copy link
Member
robbie-c commented Sep 20, 2024 •
edited
Loading

Feature request

Is your feature request related to a problem?

Our previous iteration of cookieless tracking is linked here: #1571

It works by storing the data that would go in the cookie, in memory instead, which means it's lost on a page reload.

Describe the solution you'd like

The state-of-the-art approach to cookieless tracking is creating a privacy-preserving hash server-side, using some properties, e.g., user agent, IP address, hostname, and a daily salt. No PII is stored, as the daily salt is thrown away, so it'd be impossible to reverse-engineer a user's identity from a hash.

Describe alternatives you've considered

We could keep our current approach of doing things in memory.

Additional context

There's a lot of details to get right:

  • How does this work at the API level? Do clients in this mode leave the distinct ID blank, which would be replaced server-side by the hash?
  • How do session IDs work? As this is stateless, we wouldn't have a way of rolling a new session only after 30 minutes of activity, without adding additional state.
  • How often does the hash need to rotate? Daily is common, could users customize this setting?
  • What other docs do we need? Presumably, people would use this mode because they don't have cookie consent, do we need to add docs saying not to send other PII in this mode?
  • How does the upgrade path work when cookie consent is given?
@robbie-c robbie-c added enhancement New feature or request team/web-analytics labels Sep 20, 2024
@KlotzJesse
Copy link
KlotzJesse commented Sep 23, 2024 •
edited
Loading

For me this is crucial to finally switch to posthog, a good reference point could be the data journey fathom analytics is doing:
https://usefathom.com/data

Posthog should also document like that, communicate it clearly so without giving law advices everyone can easily verify it's gdpr compliant and should be the new default tracking method, as its privacy first.

@ghoti143
Copy link
Contributor
ghoti143 commented Oct 2, 2024 •
edited
Loading

Should this solution be Schrems II compliant? That is, should IP addresses for EU originating requests only be processed by servers owned by EU entities and hosted in the EU?

Speaking of the EU, in this article, the EFF attempts to make the case that a browser fingerprint falls under the category of personal data and may not be legal under the GDPR's "legitimate interest" clause; thus requiring consent. This is a 6 year old article, however. Have the conclusions ever been tested?

@robbie-c robbie-c mentioned this issue Nov 6, 2024
Closed
@robbie-c robbie-c mentioned this issue Nov 28, 2024
Open
@robbie-c robbie-c mentioned this issue Dec 10, 2024
Merged
3 tasks
@ericdoerheit
Copy link

We use plausible.io which uses server-side hashes as far as I know. Would be great to see something similar integrated into posthog!

@motz0815
Copy link

@robbie-c I saw that #27290 has been merged a few days ago. Any news on an open / closed beta of this? I'm really stoked to try out cookieless as soon as there is something to use and would be happy to help with any feedback.

@motz0815 motz0815 mentioned this issue Jan 29, 2025
Merged
@JPustkuchen
Copy link

Thanks for this proposal! Just came here looking for a solution to map a session as fallback while using "memory" if someone denied cookies. That's a wonderful idea to not have a totally fresh "session" with each request, if cookies were denied. And still ensuring privacy, not knowing who it was, but just that it might have been the same person.

@robbie-c robbie-c mentioned this issue Mar 17, 2025
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement New feature or request team/web-analytics
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
6 participants
@ghoti143 @JPustkuchen @robbie-c @ericdoerheit @KlotzJesse @motz0815
0