What happened to justinwelsh.me?

Justin Welsh is a sharp operator. So this is not a hit piece. It is a warning about the ground everyone builds on, told through the numbers.

Between 2021 and 2025, justinwelsh.me grew its organic search traffic about 480 times. The engine behind that growth was a programmatic glossary: hundreds of "what is X" definition pages, each one tuned to catch a search. It worked, right up until it did not.

The cliff

Traffic plateaued through 2025 at roughly 8,600 to 11,500 visits a month. Then Google's December 2025 and March 2026 core updates landed, both aimed squarely at scaled content. The fall was steep and steady: from about 8,685 visits in December 2025 to 740 in June 2026. That is close to a 94% drop in six months.

Here is the tell. The keyword count did not collapse. It peaked at 9,179 in January 2026, the same month traffic started falling off a cliff. The pages were still indexed. They had simply been demoted off page one, where the clicks live.

The response

Welsh did the rational thing. He deleted the entire glossary. Every /glossary/ URL returns a 404 today, and the site's navigation was trimmed right back. Then he leaned on the asset Google cannot switch off: a newsletter audience of more than 230,000 readers.

The lesson

This is the whole point, and it is not new, but the data makes it hard to argue with:

Rented traffic gets recalled. Owned audience compounds.

Search demand is borrowed. An algorithm gives it and an algorithm takes it away, on its own schedule, without warning. An email list, a community, a base of people who know your name: that is yours. It survives core updates.

We pulled the full Semrush series, checked the live site and the Wayback Machine, and wrote the honest post-mortem, charts and all.