An explanation
Methodology.
Webtone is a directory of websites and the third-party code they load. We load each site once with nothing blocked and publish what we find: counts, categories, shapes, and a sound. The aim is not to shame a site for loading ads. It is to let you feel the weight of what normally happens silently.
What counts as a tracker.
A request is counted as a tracker when its destination appears on the Disconnect strict list. This is a conservative definition: a CDN that serves both content and analytics is counted, a subdomain of the site itself is not, and first-party scripts are never counted no matter what they do.
The numbers here describe network behaviour, not intent. A high count does not mean a site is malicious. A low count does not mean a site is safe.
One pass, per site.
Each site is loaded once in a headless browser with nothing blocked. We record every third-party request that fires during a fixed wait window after the page settles, then count the ones whose destination appears on the strict list. The detail page reports exactly what was observed — no simulation of what a blocker would or wouldn't catch.
The quietest trackers are the ones you should worry about the most.
Seven categories.
The radar shape has seven axes — one per category of tracker. A taller point means more trackers in that category on this site, scaled against every other site in the index so relative differences stay readable as the directory grows.
- Advertising
- Ads, ad-tech, and related infrastructure.
- Analytics
- Audience measurement, behaviour logging, A/B testing.
- Consent
- Cookie banners and privacy-policy enforcement layers.
- Content
- Recommendation engines, embedded media, CDN-delivered content.
- Email list identity matching and lightweight outreach.
- Email+
- Aggressive lead-capture and identity enrichment.
- Fingerprint
- Device and browser fingerprinting for silent identification.
Tracking is invisible and silent. This site asks what it would feel like if it weren't.
About the sound.
Each site plays a crowd of voices — one for every tracker, capped so the browser doesn't catch fire. The voices speak in ten languages about privacy and data ownership, mixed through a shared reverb so they feel like people in the same room rather than isolated recordings.
The sound is a metaphor, not a literal signal. A site with three trackers is audibly a conversation. A site with three hundred is audibly a crowd you cannot follow. That gap is the point.
Everything is generated client-side in the browser. No audio is uploaded or streamed; the voices are eleven short MP3s that get layered locally with randomized offsets, pitch, and stereo placement each time.
If you prefer the site quiet, toggle Sound: Off in the header. The setting persists.
← Back to the index