Tailwind Labs joins Shopify and stops selling Tailwind

Adam Wathan said on September 9th that Tailwind Labs is joining Shopify. In the same post he closed sign-ups for every commercial product his company sells.

Tailwind CSS stays MIT-licensed and the same team keeps maintaining it. Tailwind Plus and ui.sh keep the customers they have and will take no new ones.

That second part is the story.

On the commercial side, we'll no longer be trying to grow the business around Tailwind. All existing customers will of course maintain their access to products like Tailwind Plus and ui.sh, but we're closing sign ups for new customers to focus on Tailwind CSS at Shopify.

Start with the scale, because it makes the rest strange. Wathan writes that the framework "is installed over 110 million times per week". We pulled the registry numbers ourselves: npm served 110,850,620 downloads of tailwindcss in the week ending September 6th, against 152,836,339 for react. That is roughly three installs of Tailwind for every four of React (downloads are not humans, but the distortion flatters both packages equally). The separate @tailwindcss/vite package added another 38,979,953 on top.

So a dependency running at three-quarters of React's install volume was paid for by selling website templates, and the man who built both has just decided the templates were the wrong shape.

He says why, and it is not the reason you would guess. "We built a great little website template business around Tailwind over the years, but deep down I've always wanted the framework to be developed in service of a real product." The complaint is not that templates failed to pay. It is that a template shop hands you no users whose problems drag the framework anywhere it wouldn't already go.

That is a good argument. But it is also the argument every acquired founder makes, so look instead at what the buyer gets: merchant storefronts, an admin console, checkout, the Shop app, and what Wathan calls Shopify's "explorations into agentic commerce". Tailwind is already, in his words, "a load-bearing very important part of the stack at Shopify".

Our read is that this is a hire with a framework attached, not a purchase of revenue. Neither side named a price (nobody does when the number would disappoint), which is probably what you would expect from a company whose disclosed business is templates. What Shopify gets is the small group of people who understand the compiler, plus first call on where they point it.

Does that hurt you? In one specific way. An MIT licence does not un-license itself, and the git history keeps working no matter who signs the cheques (a fork is a git clone away, and Tailwind's is unusually easy to maintain), so the code is safe. The priority queue is what moves. Tailwind v4 shipped a Rust-backed engine and CSS-first configuration because the maintainers had opinions about build tooling. The version after it will likely be written by people whose day job includes a merchant admin with thousands of screens, and "in service of a real product" is a promise about whose bug report gets fixed on a Tuesday.

The strongest case against us is in the announcement itself, and it is self-interest. Shopify was, Wathan writes, "one of the very first companies operating at scale to see the potential in Tailwind CSS", and a company that has bet its own interface on a framework has every reason to keep that framework healthy. And it cuts one way only. Self-interest probably guarantees maintenance. It does not guarantee neutrality, and a framework used by ChatGPT, X, Cloudflare and Reddit — Wathan's own list — now has a maintainer with a house style.

Watch Tailwind Plus. Existing customers keep their access; nobody promised them anything new. If no new template lands there by the end of March 2027, the commercial side is over and the roadmap belongs to a merchant platform. We would take that bet at even money. A v5 release note about agentic commerce would settle it sooner.