In March we signed two projects in the same week. For a restaurant we built on WordPress. For a SaaS company we rebuilt on Nuxt with a headless CMS. Both were correct, and if we had swapped them, both clients would have been unhappy within a year. The technology question has no universal answer, but it does have a repeatable way of getting to the right one. This is how we work through it.
When WordPress is the right choice
WordPress wins when the site is mostly pages and posts, the budget is modest, and the people editing it are not developers. A restaurant that changes its menu weekly, a law firm publishing two articles a month, a tradesperson with eight service pages: all of these get more value from a mature editor and a plugin for every common need than from a bespoke codebase. The hosting requirements are modest (current PHP and database versions are all it asks for), and almost any Czech agency or freelancer can take the site over later. A quality WordPress website from us starts at 15,000 CZK for a small site and 35,000 CZK for a corporate site with custom design.
When WordPress becomes the problem
The SaaS company we mentioned had 47 plugins and a six-second load time. Nobody had chosen that; it accumulated. A form plugin, a caching plugin to fix the slowness the form plugin caused, a security plugin because the caching plugin had a vulnerability, a page builder because the theme could not do the pricing table. Each decision was reasonable on its own. Together they produced a site nobody dared update.
The warning signs we look for: more than fifteen active plugins, a page builder that owns the layout, custom data (products, locations, members, bookings) squeezed into posts and custom fields, developers who say a change is risky because of what else it might break, and mobile performance that no caching plugin can rescue. At that point WordPress is not wrong because it is WordPress; it is wrong because the project has outgrown what a content-first CMS is for. We go into the mechanics in our headless CMS and Core Web Vitals articles.
The custom alternative
Our default custom stack is a Nuxt frontend with a headless CMS for content and whatever backend the data needs. Pages are generated ahead of time or rendered on the server, so they pass Core Web Vitals without a caching plugin, and because the HTML is complete before any JavaScript runs, search engines index it without the pitfalls described in Google's JavaScript SEO guidance. There are no plugins to update and no third-party code deciding how your pricing table looks. The trade-offs are honest ones: a higher upfront cost (our custom builds start around 70,000 CZK), a smaller pool of people who can maintain it, and an editing experience that is good but less familiar than the WordPress admin.
Decision matrix: 5 factors
| Factor | WordPress wins | Custom wins |
|---|---|---|
| Budget | Under roughly 60,000 CZK | Over roughly 80,000 CZK |
| Content volume | Blog, simple pages | Complex data, API integrations |
| Update frequency | Client edits daily without help | Updates need specific layouts or come from other systems |
| Performance | Under 3 s is enough | Under 1.5 s, high traffic, conversion-critical |
| Future scaling | Stays roughly as it is for 2-3 years | New features, integrations and a second language planned |
If three or more rows land in the same column, that is usually the answer. If it is two against three, the deciding factor is who maintains the site: a client with no developer on call should lean WordPress even if the data model argues for custom.
Total cost of ownership over 3 years
Indicative Czech market ranges, not our price list. The point is not the exact numbers but the shape: WordPress is cheaper to build and more expensive to keep, custom is the reverse, and over three years the two converge more than most clients expect.
| Item | WordPress | Custom (Nuxt + headless) |
|---|---|---|
| Build | 35,000 - 120,000 CZK | 70,000 - 300,000 CZK |
| Hosting (3 years) | 10,000 - 50,000 CZK | 0 - 20,000 CZK |
| Plugins and licences (3 years) | 5,000 - 25,000 CZK | 0 - 10,000 CZK |
| Maintenance (3 years) | 36,000 - 216,000 CZK | 36,000 - 108,000 CZK |
| Total | 86,000 - 411,000 CZK | 106,000 - 438,000 CZK |
Details in the website cost breakdown and maintenance cost guide; our own figures are on the pricing page. See also WordPress vs. custom and speed optimization.
Three scenarios and what we recommended
- A dental clinic with ten pages, online booking through a third-party widget and a blog. WordPress. The booking lives elsewhere, the content is simple, and the receptionist updates opening hours herself.
- A B2B manufacturer with 600 products, technical datasheets, three languages and a price list that comes from the ERP. Custom. The product data has a shape WordPress fights against, and the ERP integration would be a fragile plugin chain.
- An established WordPress site that is slow but structurally sound. Neither. Remove the dead plugins, replace the page builder on the three templates that matter, fix images and hosting. A rebuild would have cost five times more for the same result.
Pitfalls on both sides
- Choosing custom because it sounds premium. A five-page brochure site in a custom framework is more expensive to change and harder to hand over, with no benefit the visitor can see.
- Choosing WordPress because it is free. The software is free; the hosting, licences, updates and the eventual cleanup are not. See the table above.
- Letting the page builder decide. If the layout lives in a builder, you are locked to that builder, its performance and its licence. We treat heavy builders as a reason to reconsider the whole stack.
- Skipping the maintenance plan. Both options need one. WordPress needs it for security updates; custom needs it for dependency updates. A site nobody is responsible for is a liability either way.
- Migrating without redirect mapping. If you move from one to the other, every old URL needs a 301 to its new equivalent, or the rankings go with the old site. Our migration playbook covers the full process.
A short checklist
- List what the site must do in year three, not just at launch.
- Count the kinds of data: pages and posts only, or products, locations, members, bookings?
- Name the person who will edit the site and the person who will maintain it.
- Write down every external system the site must talk to (CRM, ERP, accounting, booking, payments).
- Set a performance target in numbers, not adjectives.
- Put a three-year budget next to the build budget before comparing quotes.
Frequently asked questions
Is WordPress free?
The software is. Hosting, the domain, premium plugins and themes, and the maintenance hours are not, and over three years they usually exceed the original build price. More on WordPress vs. custom.
Can I migrate from WordPress to custom later?
Yes, typically in 6-12 weeks. Content exports cleanly through the REST API or WP-CLI; the work is in the data cleanup and the redirect map. Details in the headless CMS article.
How to decide
There is no universal answer. Sometimes WordPress. Sometimes custom. Sometimes the right move is to fix what you already have. Run the five-factor matrix, be honest about who maintains the site, and the choice is usually obvious. If it is not, schedule a free consultation and we will tell you straight, including when the cheaper option is the better one.