Two quotes for the same company website are lying on the table. One is CZK 12,000 and promises delivery in a week. The other is CZK 60,000 and talks about six weeks. Both mention responsive design, search engine optimisation and a form wired to your inbox. To someone who does not build websites, the only visible difference is a fivefold price.
The difference is not in the price. The cheap quote is not a cheaper version of the expensive one. It is a different product carrying the same name.
This article is not a defence of high prices in general. A high price guarantees nothing on its own and expensive bad websites are everywhere. It is about what can realistically be built for a given amount, where the difference shows up in money, and when paying little is the right answer.
The short answer
- For CZK 12,000 you buy four to twenty four hours of someone's work, depending on their rate. That is not enough to design a structure or write copy. It is enough to fill in a ready made template.
- A template carries decisions someone made for the market average. Your company is not the market average.
- Speed, accessibility and measurement are not add ons. They are decisions taken at the start and the only way to add them later is a rebuild.
- A website nobody measures cannot be improved. It stays a business card that occasionally receives an email.
- A higher price is defensible only when you can see which items you are paying for. A quote with a single number cannot be defended or compared.
For twelve thousand you buy hours, not a website
A website is not a product off a shelf. It is the time of someone who knows how to build it, multiplied by their rate. Czech market rates run from CZK 500 an hour for a starting freelancer to CZK 2,800 for an agency. The same amount therefore buys wildly different amounts of work.
How many hours of work CZK 12,000 buys
Values in hours
| Label | Value |
|---|---|
| Starting freelancer, CZK 500/h | 24 h |
| Experienced freelancer, CZK 1,200/h | 10 h |
| Agency, CZK 2,800/h | 4.3 h |
Ten hours is one working day and a bit. Into that has to fit the first call, the site structure, working through your materials, the visual design, the build itself, deployment, the certificate, the domain, testing on a phone, and handover with an explanation of how to run it.
It cannot be done and nobody does it that way. It gets done differently: buy a template, install the demo content, swap the photos and colours, paste in whatever text the client sent, fill in the contact details. That fits into ten hours and at first glance the result looks like a website worth sixty thousand.
At first glance. The difference shows up at the second.
A template carries decisions nobody made for you
A template is a finished website for nobody in particular. To sell a thousand times it has to suit a cafe, an estate agency, a gym and a joinery. Which means it suits none of them.
In practice it looks like this. The template assumes you have three services, because it has three tiles. It assumes you have a team with photos, because it has a section with portraits. It assumes a three tier price list, because that is what the design contains. A manufacturer, however, has fifteen products, three certificates, one plant and a buyer looking for a datasheet and a lead time.
Template: a finished design is bought, demo content is installed and your information gets dropped into sections someone designed for the market average. What is not in the template will not be on your site. What is in it and you do not need stays there, because removing a section means editing someone else's code.
Custom design: it starts with who the site has to convince and what that person needs to see before they write or call. The page structure follows from that and only then does the visual work start. A section nobody needs never gets drawn. What actually decides deals in your business gets its own place.
A template is not a design. It is a set of someone else's decisions with your logo added
This is not an argument about taste. It is an argument about whether the site leads a visitor to an action or merely informs them that you exist. Structure, the order of sections and the wording are work that somebody has to do. When it is skipped, nothing looks wrong. The only symptom is that nobody gets in touch.
Speed is the first place the difference gets measured
Almost half the websites in the world run on WordPress. It is not a bad system and plenty of fast sites are built on it. The problem is not the system but the stack a cheap job usually ends up with: a template, a visual page builder on top of it, and ten to twenty plugins, each adding its own scripts and styles to every page.
Share of sites passing Core Web Vitals by platform
Values in per cent
| Label | Value |
|---|---|
| Duda | 84.9% |
| Shopify | 75.2% |
| Wix | 74.9% |
| Squarespace | 70.4% |
| Drupal | 63.3% |
| Joomla | 56.9% |
| WordPress | 46.3% |
| Magento | 41% |
More than half of WordPress sites therefore fail Google on speed. Hosted builders such as Wix or Duda do better, because they will not let the site owner load anything they like into a page. The price for that is a closed system, which has its own article. Speed can be bought in two ways: either someone trades your freedom for guardrails, or whoever builds the site keeps watch over it.
What it comes down to is page weight. The browser has to download it, parse it and paint it, and a mobile connection on a building site is not office wifi.
Slowness is not an aesthetic problem. It is a business problem that can be stated in money, because somebody measured it on real orders.
There will be fewer visits, not more
The argument that a website is enough because people will find you held while search engines handed out visits for free. That has changed over the past two years and it keeps changing.
It does not mean search stopped bringing work. It means the visit that does arrive is rarer and more expensive than it used to be. If the site fails to turn it into an enquiry, there is no second chance: that person goes back to the results page and clicks a competitor.
Speed is only one signal in that. Google has confirmed Core Web Vitals feed into how pages are assessed, but it is one item among many and content still comes first. Speed decides when two sites are otherwise comparable. Above all it decides whether the visitor waits at all.
Trust is settled before anyone reads a sentence
This is the least comfortable part of the argument, because it sounds shallow. The data is twenty years old and has only been confirmed since.
The first impression is faster still. Canadian research from 2006 showed that ratings of a page's visual appeal after fifty milliseconds are practically the same as after half a second. Fifty milliseconds is a tenth of a blink. At that point the visitor has not read a single word.
What specifically undermines trust on a cheap site:
- Stock photos your buyer recognises from a competitor's website
- Copy about quality, tradition and an individual approach, which is copy about nothing
- Demo content left in the template: a Lorem ipsum in the footer or a link to a social network you do not have
- A form that says nothing after it is sent, so the visitor has no idea whether the email went anywhere
- No company number, no address of the actual premises and no name of the person they will be dealing with
- A site that scrolls sideways on a phone, because nobody opened it on a phone
None of these is a programming error. All of them follow from a budget that had no room for anyone to look at the site through a customer's eyes.
Accessibility is no longer a bonus, it is law
Since 28 June 2025 Czech act no. 424/2023 Coll. on accessibility requirements for certain products and services has been in force, the national form of the European directive. It covers e-commerce, banking, transport and electronic communications among others. Sites and apps in scope have to meet WCAG at level AA. For online shops the Czech Trade Inspection Authority supervises compliance and the maximum fine is CZK 10 million.
Two qualifications so this does not read as scaremongering: microenterprises providing services, meaning under ten employees and under two million euro turnover, are exempt, and services provided before the act took effect have until 28 June 2030. An ordinary company website that sells nothing is usually out of scope entirely.
That does not make accessibility irrelevant. It means half the argument is legal and the other half commercial: an inaccessible site is unusable for someone with poor sight, someone navigating by keyboard, or anyone trying to read grey on white in direct sunlight.
Most common accessibility failures on home pages
Values in per cent
| Label | Value |
|---|---|
| Low contrast text | 83.9% |
| Image without a description | 53.1% |
| Form field without a label | 51% |
| Empty link | 46.3% |
| Empty button | 30.6% |
| Missing page language | 13.5% |
Security and running costs: the bill arrives later
A cheap website is sold as a one off item. The cost only starts there, because every plugin is a piece of someone else's code that has to be kept updated.
A site with a template and twenty plugins therefore needs someone who updates it every month, checks nothing broke afterwards and keeps an eye on backups. Without that person the site keeps running and looks the same, right up until someone gets in. Then comes a bill that was not in the original quote: restoring from backup, downtime, and if Google has flagged the pages as harmful in the meantime, a drop in the results that takes weeks to recover.
A site built as custom code without third party plugins has an attack surface an order of magnitude smaller. That is not magic, it follows from not running twenty extensions written by strangers.
What the higher price contains
The best way to defend a price is to break it down. Here is the same thing set out side by side.
| What a site has to do | Template for a few thousand | Custom built site |
|---|---|---|
| Page structure | Given by the template, content gets dropped in | Follows from who the site has to convince |
| Copy | Yours, otherwise the demo text stays | Wording and the order of arguments are part of the design |
| Speed | Whatever is left after the template and plugins | A page weight budget held from day one |
| Accessibility | Luck, most often fails on contrast | WCAG 2.2 AA as a condition of handover, contrast calculated |
| Traffic measurement | Google Analytics and a cookie banner with it | Own measurement without cookies, so no banner |
| Admin interface | Universal, does everything except your thing | Only the fields that make sense in your business |
| Third party scripts | Fonts, icons, maps and chat from other people's servers | No third party scripts, everything from your own server |
| A change a year later | Editing a template nobody knows | Own documented code, the change is an hour of work |
| If you and the supplier part ways | Template and plugin licences stay with them | The source code, the data and the domain are yours |
Why the higher price is defensible
Because what it buys is not a prettier look but decisions someone made in your place, and the work of carrying them out. It looks like this.
- 1Work out who is supposed to come to the site and what they need to see before they write. For a manufacturer that is a buyer after a datasheet, for a service business someone who wants a price and a date. This decision shapes everything else.
- 2Build the structure: how many pages, what goes on each, what the home page opens with and where every button leads. Only then does drawing make sense.
- 3Write or rewrite the copy. This is most often the largest item and most often the one missing from a cheap quote. Without it the site fills up with sentences about quality and tradition.
- 4Design the look around your content rather than fitting content into someone else's look. Photos from your own premises instead of stock, because a buyer can tell a real plant from a mood board.
- 5Build it with a page weight budget, with contrast that is calculated rather than guessed, and with keyboard operation. Retrofitting all of that costs more than doing it once.
- 6Set up measurement that shows how many people came, how far they got and where they left. Without it a site cannot be improved, only redone on hunches.
- 7Hand it over with an explanation of how to change the content, and with the source code. A site you can maintain stays alive. A site you are afraid to touch is a year out of date within a year.
When a cheap website is the right answer
I am not claiming little money always means a bad site. It means a smaller scope, and a smaller scope is sometimes exactly what is needed.
One honestly built page for fifteen thousand makes sense when you have one service, one town and one goal: getting a phone call. Such a page can be fast, accessible and convincing, because the scope matches the price. A bathroom fitter working one district will get more out of it than out of a site for eighty thousand.
The bad case is different: a company website with six services, a price list, references, a blog and two languages promised for twelve thousand. There the price is not held by working faster, it is held by leaving out everything except filling in a template. Scope and price do not match and the content, the speed and the result all pay for it.
And then there is the case where no amount of money helps: when a company has nothing to say, will not publish prices, has no photos, no references and nobody with time to supply anything. A website will not fix that. I say so on the first call when it becomes clear.
How to tell you are being sold a template
None of this is aimed at suppliers. It is a way to find out what you are buying before you sign.
- 1Ask whether the quote is broken down into items with prices. Anyone who refuses has told you enough.
- 2Ask what in the quote covers the site structure and who writes the copy. If the answer is that you supply the text, add your own time or a copywriter's fee to the total.
- 3Ask what it will run on and how many plugins will be involved. A template and a few plugins is not wrong in itself, but then ask who updates them year after year and for how much.
- 4Ask for two finished sites and measure them yourself in PageSpeed Insights on mobile. It takes a minute and it is the one number nobody can polish for you.
- 5Open those sites on your phone and try sending an enquiry. You will see exactly what your customer sees.
- 6Ask what happens if you part ways: do you get the source code and the data, or does the site run on licences held in the supplier's name?
When the price difference pays for itself
The gap between the cheap option and the considered one is usually thirty to fifty thousand crowns. That figure looks different next to a single deal.
With a margin of CZK 15,000 on an ordinary deal, a difference of CZK 45,000 is repaid by three extra deals over the whole life of the site, meaning roughly five years. Three deals in five years is one deal every twenty months. For a company doing bathrooms, roofs or custom manufacturing that is a small number.
Put your own margin in. If it turns out the site would need to bring twenty deals a year to pay for itself, the answer is a smaller scope, not a more expensive site.
Frequently asked questions
Why is it worth paying more for a website?
Because the price of a website is mostly time. CZK 12,000 buys four to twenty four hours of work depending on the rate, which covers filling in a ready made template but not structure, copy, speed, accessibility and measurement. The difference shows up in how many visits end in an enquiry, and at a margin of CZK 15,000 per deal the higher price is repaid by three extra deals over five years.
Is a website for ten thousand always bad?
No. One honestly built page at that price makes sense when you have one service and one goal. The bad case is different: a company website with six services, a price list, a blog and two languages promised for the same money. There the price is held by leaving out everything except filling in a template.
What is wrong with templates?
Nothing in itself. The problem is that a template carries decisions made for the market average: three services, a team with photos, a three tier price list. What is not in it will not be on your site, and what is in it and you do not need stays there. On top come plugins covering what the template cannot do, each adding its own scripts to every page.
Does a slow site really rank worse?
Speed is one of the signals Google uses to assess pages, but not the main one. It matters most when two pages are comparable on content. The far bigger effect is on the visitor: making a mobile site one tenth of a second faster raised retail conversion by 8.4 % in the Deloitte and Google study.
Does my site have to be accessible by law?
Czech act no. 424/2023 Coll. has applied since 28 June 2025 and covers mainly e-commerce, banking, transport and electronic communications. Microenterprises under ten employees are exempt for services, and services provided before the act took effect have until 28 June 2030. An ordinary company website that sells nothing is usually out of scope, which does not mean an inaccessible site is pleasant to use.
How do I know I am being sold a template at custom prices?
From the quote. A custom quote is broken into items and contains work on structure and content. A template quote is one number and one sentence about modern responsive design. Then ask for two finished sites and measure them in PageSpeed Insights on mobile.
Summary
- The price of a website is time. CZK 12,000 buys four to twenty four hours depending on the rate, which is enough to fill in a template, not to design anything.
- A template carries someone else's decisions. What is not in it will not be on your site, and what is in it needlessly stays there.
- Speed, accessibility and measurement happen at the start. Adding them later means a rebuild, which costs more than doing it once.
- Customers judge credibility by the look before they read a sentence. Appearance is the single most frequently cited criterion.
- Running costs are part of the price even when a cheap quote omits them. Plugins get updated every month, or the site eventually belongs to someone else.
- A cheap website is not a bad choice when it matches the scope. The bad choice is a big website promised for small money.
Want to know which of this applies to your site
Send me the address of your current website. I will measure it, go through the structure and write back with what I would fix first, what second and what is not worth touching. No commitment and no need to order anything.
If it turns out your site is fine and only needs better copy, I will say so. And if it turns out no website will help you, I will say that too.
See the price list item by item or send a no obligation enquiry.
Sources
- Shean: what web and e-shop development costs in 2025, freelance and agency rates compared
- W3Techs: WordPress share of all websites and of content management systems
- HTTP Archive: Core Web Vitals Technology Report
- HTTP Archive: Web Almanac 2025, Page Weight chapter
- Deloitte and Google: Milliseconds Make Millions
- SparkToro: in 2026 less than one third of Google searches still send a click
- Fogg et al.: How do users evaluate the credibility of Web sites?
- Lindgaard et al.: Attention web designers, you have 50 milliseconds to make a good first impression
- WebAIM Million 2026: accessibility of one million home pages
- Czech Trade Inspection Authority: accessibility of products and services for businesses
- Patchstack: State of WordPress Security in 2026