Download websites that use Newspack by Automattic

Detecting a content management system usually tells you nothing about the publisher. The same WordPress code runs a bakery and a bank. This platform is the exception worth paying attention to, because nobody installs it by accident and nobody installs it for a bakery.

The product is a publishing stack Automattic built on top of WordPress for news organizations, released in 2019 under the GPL v2 license and written in PHP, like the CMS beneath it. What it adds to the base system is the machinery a newsroom needs and a general site does not: subscription and donation flows for reader revenue, newsletter capture, ad placement and article layouts arranged around a masthead. As of October 2026 we detect it on 358 sites.

A count in that range is not a disappointment, it is the shape of the market. The product targets small and mid-sized local and regional outlets, an audience that is numerous but individually tiny and that has been contracting for years. Nearly every domain on the list is a working news publication. Few technology fingerprints classify a site's industry that cleanly, and the classification is precisely what a general WordPress list cannot give you.

Filtering the whole WordPress population for news sites means guessing from URLs, keywords or manual review across an enormous set of domains, and accepting the error rate that comes with it. Filtering for this stack hands you the subset directly. That is why the list is useful to media researchers tracking the health of local journalism, to advertising and syndication networks looking for inventory, and to anyone studying how reader revenue models spread through small newsrooms rather than through national titles.

Two honest caveats travel with it. The signature we match is a theme and plugin fingerprint layered on top of WordPress, so a publisher who has customized heavily, renamed assets or moved to a bespoke front end can drop out of the count while still running the platform underneath. And the managed service, the open source project and their variants share most of that fingerprint, so a detected domain is evidence of the publishing stack and not proof of any particular commercial relationship with the vendor. The list is a reliable answer to which sites publish on this stack. Read it as that, and verify the relationship separately if the relationship is what you need.