Arbitrary values are not being emitted
Dynamic class construction building the class name at runtime, a content path that misses the template file, or an older Tailwind version that does not support the arbitrary syntax.
Classes present and applied
Running Tailwind CSS?
Tailwind is the fastest way to style a UI until the content configuration cuts a class you only use at runtime, the v4 engine refuses to recognize a plugin that worked on v3, or three different teams have each invented their own `btn-primary` and nobody wants to touch the audit. We go through the config, the build output and the components, and we leave you with a system that is actually consistent.
What it is
Tailwind CSS is a utility-first CSS framework created by Adam Wathan and first released in 2017. Rather than writing component classes, you compose designs from small single-purpose utilities (`flex`, `pt-4`, `text-sm`, `bg-gray-800`) directly in markup, and an unused-class purge keeps the final stylesheet small.
Version 4, released in 2025, replaces the PostCSS plugin with a native engine, drops `tailwind.config.js` in favor of CSS-first configuration via `@theme`, and adds container queries, the `3xl` breakpoint and a faster incremental build. The upgrade path is straightforward for simple projects and significantly more involved for codebases with plugins, deep theme extensions or server-rendered dynamic class names.
The honest version
No technology is all upside. This is where Tailwind CSS earns its place and where it tends to cause trouble.
Sound familiar?
The Tailwind CSS problems that bring people to us, and what you have once each one is fixed.
Dynamic class construction building the class name at runtime, a content path that misses the template file, or an older Tailwind version that does not support the arbitrary syntax.
Classes present and applied
A content configuration that no longer matches the project structure after a folder move, or a path glob that excludes the files actually using Tailwind.
Content paths accurate, styles emitted
A codebase mid-migration from v3 to v4, with CSS-first config half done, plugins not ported, and the design visibly different depending on which page you look at.
Project fully on v4 with parity
Classes constructed from string templates or stored in a CMS, invisible to the Tailwind scanner, that work in dev because every class is present and fail once the production purge runs.
Safelists or refactor so purge is honest
Multiple component libraries, three competing button styles, color values hard-coded in some files and tokenized in others, and nobody wants to be the one to rewrite it.
One tokenized system, enforced in code
`@apply` used in CSS files that are not processed through Tailwind, or applied to classes that are purged before the directive resolves.
@apply in processed files, scoped correctly
A late-loaded stylesheet, a build that split critical CSS incorrectly, or a Tailwind preflight race with framework-level style injection.
Styles present at first paint
What we do
Call us for one page that looks wrong, or the entire design system.
How it works
Tell us what's going wrong. We look at it, then tell you plainly what's wrong and what it will cost.
We fix the cause, not the symptom. If something is down, getting it back up comes first.
Backups, monitoring, updates and access, so the same problem can't quietly come back.
You get a plain-language write-up. Or we stay on for hosting, upkeep and new work.
Often found together
We work on the whole stack, so the problem can be wherever it actually is.
For most projects, yes, but on a schedule that respects the scope. The new engine is faster, container queries are first class, and v3 will eventually stop getting fixes. For codebases with heavy plugin use, custom theme extensions, or server-rendered dynamic class names the migration is a real project. We plan it, migrate in branches, run visual regression against staging, and ship it without surprises.
Almost always because the class name is constructed dynamically or stored somewhere Tailwind's scanner cannot see. The dev build emits more classes, so the issue is hidden. The fix is to either stop constructing class names at runtime, add a safelist for the classes that must survive purge, or widen the content paths to include the files where the classes actually appear.
Yes. That work usually starts with inventorying what is already in the codebase, agreeing on tokens for color, spacing and type, extracting the patterns that repeat into components, and documenting the result so new work does not immediately drift back. We can also layer Headless UI or shadcn/ui on top if a headless library is a better fit than growing one in-house.
Not inherently, but it is easy to overuse. The pattern that works is small utility combinations extracted into components (a button, a card) where the semantics are clear. The pattern that goes wrong is wrapping every utility into a bespoke class, which rebuilds the old CSS problem Tailwind was meant to avoid.
Yes. We review the config, the content paths, the plugin set, the component patterns and the token usage, and give you a short report on what will need tidying up before new features ship cleanly. Then we either do the tidy or hand the plan to your team.
Free estimate first. Send us the repo or describe what is rendering wrong, and we come back with a price you can sign off before any work starts.
Free estimate
Tell us what you're running and what's going wrong. A person reads it and comes back with an estimate.
Your message is with a real person, and we'll get back to you soon.
Rather talk now? Call us: