Guides

Practical instructions for reducing unnecessary storage of visitor IP addresses and other identifiers. Not every configuration is available on every hosting plan, so provider-level logging must always be checked separately.

Search guides Category

Fundamentals & checks

IP processing and IP storage: the crucial difference
A website cannot be delivered without IP processing. However, this does not mean that the address has to be saved permanently.
Anonymize IP or not save it at all?
Anonymization can be useful; if the address is not needed at all for the purpose, not saving is the more data-efficient architecture.
Cookieless is not automatically tracking-free
Even without cookies, local storage, session storage, IP-based hashes or server-side identifiers can combine visits.
Unique visitors without tracking: why exact counting requires recognition
Exactly unique visitors require that multiple page views can at least temporarily be assigned to the same instance.
What does my hosting provider store about website visitors?
Website operators must consider the application, web server and hosting/security layer separately.
Checklist: does my website store unnecessary visitor data?
A practical check from browser storage to network requests to server, provider and security logs.
Linux: find your own test IP in log files
With your own test IP you can check which accessible server logs a single page view ends up in.
MySQL/MariaDB: find possible IP and visitor columns
A pure schema query finds suspicious column names without reading stored visitor data.
PHP: Check REMOTE_ADDR and passed client IP
Source code search shows places that use visitor IP or proxy headers; a hit does not prove storage.
Check Set-Cookie header with curl
curl makes server-side cookies visible, even if JavaScript cannot read them because of HttpOnly.
Check cookies, localStorage and sessionStorage yourself
Browser developer tools show stored identifiers that a pure HTML scanner cannot fully detect.
Check third-party requests with browser DevTools
The network view also shows requests that JavaScript only creates after loading and that are not visible in HTML.
WordPress: Data protection self-test with WP-CLI
WP-CLI lists active plugins and relevant options without changing the installation.
Reverse Proxy: Check forwarding of the visitor IP at the origin
Configurations and headers show whether a proxy passes the original client IP to the origin.

Web servers, reverse proxies & CDN

Apache access logs without visitor IP addresses
Current Apache 2.4 configuration: Avoid access logs or use a log format without a client IP.
nginx access logs without visitor IP addresses
Turn off access logging or use your own nginx log format without remote_addr and forwarded IP.
Caddy: Reduce IP addresses in access logs
Minimize Caddy logs with current filter functions and take remote_ip, client_ip and forwarded headers into account.
Traefik: Configure access logs to save data
Remove client IP, headers and query parameters from Traefik access logs.
HAProxy: Logs without persistent client IP
Use HAProxy as an upstream security layer without unnecessarily perpetuating the visitor IP in normal request logs.
Cloudflare: Don't share visitor IP to origin
Cloudflare processes the IP at the edge; With Remove visitor IP headers you can limit their forwarding to the origin.
Caddy: Reconstruct original IP behind reverse proxy only if necessary
Caddy only evaluates client IP headers if trusted_proxies are configured; This allows you to consciously decide whether the Origin needs the visitor IP.

WordPress

WordPress: Statify as a simple, data-efficient page view counter
Statify counts page views instead of visitors and, according to the project documentation, does not store IP addresses or visitor profiles.
WordPress: Set Koko Analytics to be as data-efficient as possible
Koko offers cookie, cookie-free and pageview-only modes. For maximum data minimization, pageview-only is the clearest option.
WordPress: Configure Burst Statistics to save data
Burst is self-hosted but offers different detection modes. For “We don’t save!” The privacy setting must be chosen consciously.
WordPress: Classify independent analytics in a data protection-conscious manner
Independent Analytics does not store raw IP, but creates a reusable visitor ID from IP, user agent and salt.
WordPress: use Antispam Bee with data minimisation
Filter comment spam locally, avoid third-party providers and be aware of IP-dependent options.
WordPress: Don't permanently save comment IP
Customize WordPress comments so that the normal comment IP is not permanently retained as metadata.
WordPress: Avoid Gravatar third-party connections
Avatars can create a connection to Gravatar when viewing a page; Local avatars or disabled display avoid this request.
WordPress: Contact Form 7 without unnecessary additional services
Keep Contact Form 7 itself slim and check optional reCAPTCHA, Flamingo and other integrations separately for storage.
WordPress: Configure WP Statistics to save as much data as possible
WP Statistics works locally and without cookies; Data protection options prevent additional personal storage and data sharing.
WordPress: MailPoet without opening and click tracking
MailPoet can reduce or disable engagement tracking for newsletters; Also check diagnostic logging and optional data transfer.
WordPress: Classify Wordfence and IP-based security functions
Wordfence uses IP-related security data and external infrastructure; Clearly separate security purposes and normal visitor analysis.
WordPress: Minimize Solid Security Logs and IP Detection
Solid Security logs security events and uses IP addresses for lockouts; Storage location and retention are configurable.
WordPress: Check WP Activity Log and saved IP addresses
WP Activity Log stores, among other things, the source IP for events; for strict data economy, critically examine the scope and purpose.
WordPress: Consciously configure login blocks and IP storage
Login limiters typically work IP-based; Attack defense does not automatically justify unlimited storage.
WordPress: Gravity Forms without saved submitter IP
Gravity Forms can switch off the storage of the IP during form submissions and automatically delete entries after a period of time.
WordPress: Turn off Fluent Forms IP logging and country detection
Fluent Forms provides filters to disable IP logging and country detection derived from request headers.
WordPress: Explicitly activate SlimStat IP anonymization
In current versions, SlimStat stores full IP addresses again by default; Anonymization or hashing must be activated consciously.
WordPress: Jetpack Stats processes visitor IP despite aggregated display
Jetpack shows aggregated statistics to website operators, but Automattic temporarily maintains stats logs with visitor IP.

WooCommerce

WooCommerce: turn off voluntary usage tracking
WooCommerce offers voluntary usage data reporting; it can be deactivated in the advanced settings.
WooCommerce: Only use IP geolocation when needed
WooCommerce can geolocate customer IP locally using MaxMind data; cache mode also generates an IP-based URL hash.

Security & error diagnostics

CrowdSec: Attack IP as security data instead of visitor statistics
CrowdSec processes IP addresses of detected attackers specifically for security decisions and has defined retention and degradation rules for this purpose.
Bugsnag: Do not use IP as automatic browser user ID
By default, Bugsnag uses the IP address as the user ID in the JavaScript SDK; collectUserIp can prevent this collection.
Sentry: remove personal error data before sending
Sentry offers SDK scrubbing, server-side rules and relay; If you never want data to leave your system, it must be removed before sending.
Rollbar: IP-based error history and diagnostic data
Rollable can group events by IP and display location information; This function must be examined critically for strict data economy.

Other CMS

Drupal: anonymous page views with statistics
The current Statistics module for Drupal 10.3+/11 focuses on anonymous content views instead of visitor profiles.
Joomla: Use data protection functions and check extensions separately
Joomla's privacy functions help with information and consent, but do not replace checking logs, extensions and third-party resources.
Joomla: User Actions Log without IP address
Joomla can log user actions without additionally including the IP address in the actions log.
TYPO3: Configure IP anonymization and logs correctly
TYPO3 has a current ipAnonymization setting - but it does not work when creating new log entries and does not replace a log check.

Statistics & analytics

Matomo: configure analytics for data minimisation
Look at IP anonymization, tracker proxy and web server logs together – not just the Matomo database.
Matomo: Remove original IP before the tracker
The Matomo Tracker Proxy can remove the visitor IP before Matomo processes it - stricter than just subsequent IP masking.
Umami: cookieless does not mean without visitor identification
Umami does not store IP as a metric, but forms sessions from IP, user agent and website ID. Distinct IDs can additionally connect profiles.
Plausible: Understanding IP processing and daily visitor recognition
Plausible does not store the raw IP, but processes the IP and user agent for location and a daily changing visitor ID.
GoatCounter: Correctly classify sessions, IP processing and aggregates
GoatCounter stores aggregates by default; IP and user agent are processed in RAM for up to eight hours for session detection.
GoAccess: Anonymize IP addresses when evaluating logs
GoAccess can mask client IP addresses when reading in; The original web server log must still be configured separately to save data.
PostHog: Correctly classify cookieless mode
At PostHog, cookieless means no browser ID, but server-side daily recognition from IP and user agent.
Countly: Understanding IP processing for geolocation
According to the documentation, Countly does not store raw IP, but processes it first to determine location.
Simple Analytics: Classify analytics without visitor recognition
Simple Analytics does not document cookies, local storage identifiers, or storing or hashing IP addresses.
Fathom Analytics: cookieless, but with short-term IP processing
Fathom does not set analytics cookies, but processes IP and user agent for daily rotating visitor signatures.
Microsoft Clarity: Consent Mode prevents cookies, not every measurement
Clarity can work without cookies before consent, but still carries out requests and basic measurements.
Pirsch: cookieless with daily visitor ID from IP and user agent
Pirsch does not store the IP, but processes it with user agent, date and salt to create a visitor ID that can be used for up to 24 hours.
Piwik PRO: measure without cookies and without session hash
Piwik PRO can switch off cookies and session hash; then each event is treated as a new session and visitors are not recognized.
Cloudflare Web Analytics: classify cookieless RUM connection
Cloudflare Web Analytics uses an external beacon and describes the measurement as without personal end-user profiles; The network request still remains.
Plausible proxy: Visitor IP is still passed on to Analytics
A first-party proxy visually hides the analytics request but does not eliminate Plausible's IP processing.

Forms, CAPTCHA & spam protection

Contact form without tracking and unnecessary IP storage
Process contact forms locally, only collect necessary fields and build spam protection without unnecessary third-party connections.
CAPTCHA and spam protection: first local, then external
Use honeypot, time check and server-side rules before external CAPTCHA services and make third-party connections visible.
Cloudflare Turnstile: CAPTCHA replacement is still a third party
Turnstile avoids classic image CAPTCHAs, but sets up Cloudflare communication and should therefore be evaluated separately from local spam protection.
Google reCAPTCHA: Consciously evaluate third-party connections
reCAPTCHA integrates Google into the bot check; Local spam defense is the first choice for data-saving forms.
hCaptcha: integrate data efficiently and recognize third-party connections
hCaptcha is also an external challenge service; only load where protection is required and configure server checking separately.
Getform: Form service collects additional visitor data
In addition to form fields, Getform also documents IP address, location, operating system, browser and device type.
Brevo Forms: CAPTCHA means additional third-party processing
Brevo offers Google reCAPTCHA or Cloudflare turn styles for forms; Both options involve an additional external service.

Newsletters & email tracking

Newsletter without opening and click tracking
Send newsletters without installing invisible tracking pixels and personalized tracking links for each recipient.
Brevo: Anonymize opening and click tracking
By default, Brevo assigns opens and clicks to recipients; Anonymous tracking separates these events from specific contacts.
Mailchimp: Turn off open and click tracking
Mailchimp enables open and click tracking by default; both functions can be deactivated per email.
MailerLite: Deactivate Open Tracking specifically
MailerLite allows opening measurement to be switched off; other measurements such as click or e-commerce tracking must be controlled separately.
Buttondown: Open Tracking is opt-in
Buttondown does not necessarily activate opening tracking; When activated, a receiver-specific tracking pixel is inserted.

Hosting & providers

Hetzner Webhosting: Configure access logs to save data
Hetzner web hosting does not require permanent visitor IP logging; the documented standard retention of logs is seven days.
IONOS web hosting: check anonymized logs and storage
IONOS documents anonymized web hosting logs; The hosting comparison shows a retention period of eight weeks.
STRATO web hosting: separate anonymized customer logs and security logs
STRATO provides anonymized customer logs; Security-related processing of complete IP addresses must be considered separately.
manitu web hosting: Activate IP anonymization yourself
At manitu, the anonymization of web server logs is a conscious setting; The replacement by 127.0.0.1 is documented in the hosting comparison.
Mittwald Hosting: Check anonymized access logs and error logs separately
Mittwald anonymizes access log IP addresses; Error logs may contain IP addresses and have a shorter retention period.

Website builders

IONOS Website Builder: Classify visitor logs and web analytics
IONOS directly documents anonymized IP addresses and 21 days of storage of certain visit data for website builders.
STRATO homepage builder: Check statistics, cookies and external content
STRATO documents its own visitor statistics, cookie control and external content as separate functions.
Squarespace: Reduce analytics cookies and unique visitors
Squarespace Analytics uses, among other things, cookies for up to two years for unique visitors; Analytics cookies can be deactivated.
Webflow Analyze: cookieless means local storage here
Webflow Analyze does not use cookies for visitor recognition, but uses local storage and anonymous identifiers.
Jimdo Creator: Consciously activate or deactivate statistics
Jimdo Creator Statistics is cookieless and anonymized; the function can be completely deactivated.
one.com Website Builder: Check consent banners and tracking integrations
one.com offers a consent banner; additional analytics and advertising integrations must be evaluated separately.
Wix: Minimize cookies, analytics and session recordings
Wix distinguishes between essential and consent-requiring analytics cookies; Session recordings are an additional, separately activated function.
Hostinger Website Builder: Only activate tracking integrations specifically
Google Analytics, Hotjar, Meta Pixel and other tracking services are optional integrations of the builder.

PHP & website resources

PHP sessions: only use when state is really needed
PHP sessions typically generate a recognizable session ID; Minimize lifespan, cookie properties and server-side data.
External fonts and CDNs: avoid unnecessary third-party connections
Deliver fonts, JavaScript and CSS locally if there is no technical reason for external retrieval.
Only load YouTube, maps and other embeds as needed
Do not automatically connect external media when the page is accessed; Display preview image or placeholder locally and only load embed after action.
Disqus: Evaluate third-party comments and tracking separately
Disqus processes technical data and tracking signals for embedded comments; Publishers can disable data sharing.
YouTube nocookie: Data protection mode does not replace a two-click solution
youtube-nocookie reduces certain storage, but still establishes a third-party connection when loading the player.
Cloudflare Zaraz: server-side tagging is still third-party processing
Zaraz can load marketing and analytics tools differently, but does not automatically eliminate their data processing.