Original research · Published 2026-07-30

YouTube to MP3 Safe? 70-Site Transparency Study (2026)

People choosing an online converter mostly want to know three things: will it make me sign up, will it bury me in ads, and where do my files actually go? We measured what 70 converters do on their public homepages, and how often that matches what they claim.

Key findings

89.1%

enforce HTTPS

57 of 64 analysed

14.1%

publish neither a privacy policy nor terms of service

9 of 64 analysed

20.3%

publish a DMCA or copyright page

13 of 64 analysed

4.5

third-party hosts referenced by the median homepage

32 sites reference 5 or more

Claims versus what the page does

Marketing copy is easy to check against the page it sits on. Three claims appear constantly in this category, so we measured all three.

sites claim to be ad-free but load advertising code

of the 23 that make the claim

sites claim "no signup" but still show a signup or login entry

of the 34 that make the claim

46.3%

load advertising code

25 of 54 that could be assessed

11/32

sites promising local-only processing show a visible WebAssembly trace

absence of a trace is not proof the claim is false

“Your files never leave your device”

Half the sample tells visitors their files never leave the device. Browser-side conversion needs WebAssembly, which usually leaves a trace in the page. We found that trace on 11 of the 32 sites making the claim. The remainder may still process locally and load those scripts on demand, so this is a transparency observation, not evidence that any claim is false.

Some of these homepages cannot be audited at all

10 of the 64 homepages we analysed are JavaScript shells: the HTML the server sends contains almost no content, and everything the visitor sees is assembled in the browser afterwards. For those sites we report undetermined rather than ad-free. Reporting them as ad-free would turn “we could not see it” into “it is not there”, which is a different claim. That is also why the advertising figures on this page use a denominator of 54 rather than 64.

Brand reuse: 1 of 64 analysed sites put a well-known converter brand in the page title while the registrable domain has no relationship to that brand.

Claimed bitrate: 23 sites advertise 320 kbps; 1 site advertises 192 kbps. We recorded the claim only. This study did not convert any file, so it does not verify the output bitrate.

How the sample was built

Distinct registrable domains appearing in top search results for eight converter-related queries, covering both YouTube-specific converters and general online audio, video and file converters. This is not a census.

youtube to mp3 converter free online
best youtube to mp3 converter 320kbps no signup
youtube to mp3 320kbps
free mp3 converter online no registration

Subdomains were folded into their registrable domain so each operator counts once. Of 70 domains, 64 returned a homepage we could analyse and 6 did not respond or refused the request; the unreachable ones are excluded from every denominator rather than counted as failures.

What we measured

HTTPS enforcement: Whether a request to the http:// address is redirected to https://.
Advertising code: Whether the homepage HTML references a known ad network or contains a known ad loader such as an adsbygoogle or pop-under initialiser.
Third-party hosts: How many distinct external hosts the homepage loads scripts or frames from, excluding the site's own registrable domain.
Privacy policy: Whether the homepage links to a privacy policy page.
Terms of service: Whether the homepage links to a terms of service or terms of use page.
DMCA or copyright page: Whether the homepage links to a DMCA notice or copyright policy page.
Signup entry: Whether the homepage shows a signup, register or login entry point.
Ad-free claim: Whether the homepage copy states the service carries no advertising.
No-signup claim: Whether the homepage copy states no account or registration is needed.
Claimed maximum bitrate: The highest bitrate advertised in the homepage copy, in kbps. Recorded as a claim only; no file was converted to verify it.

Limitations

Sample of 70 registrable domains drawn from search results, not a census. Percentages describe this sample only.
All measurements come from the static HTML of each public homepage. Content injected after page load by JavaScript is not visible to this method.
Advertising figures use a smaller denominator (54) because 10 homepages are JavaScript shells that static analysis cannot assess.
Where a site claims local processing but no WebAssembly trace was visible, we report the absence of visible evidence. We do not treat that as proof the claim is false, because such scripts are often loaded on demand.
Presence of a policy link was detected from homepage markup. A link being present was not taken as proof the policy page is complete or accurate.
No conversion job was submitted and no media file was downloaded on any site in the sample.

How the data was collected

Raw HTML snapshots of every analysed homepage are retained so any figure on this page can be re-checked against the page as it appeared on the collection date.

Only public homepages and public policy pages were requested. No conversion job was submitted and no audio or video file was downloaded.
No access control, CAPTCHA or bot protection was bypassed. Sites that blocked the request were recorded as unreachable and excluded, not retried through a workaround.
Requests were spaced out and limited to the few pages each measurement needed.
Only publicly observable facts were recorded. This report does not rank the sites or judge which is better.
Raw HTML snapshots are retained so every figure can be re-checked against the page as it appeared on the collection date.

Last updated 2026-07-30. If you are writing about this data and need a figure clarified, the underlying per-site records and homepage snapshots can be shared on request.