How NOT to commission a website

I sometimes browse freelance platforms and groups where people post projects. Sometimes I’m actually looking for work, sometimes it’s just professional curiosity. And I have to admit, it’s a bit like reading car ads that say “needs a few minor fixes” while the rear wheel in the photo is sitting at a 45-degree angle.

And I regularly come across briefs that sound roughly like this:

“We’re looking for a complete website redesign. We want a modern, dynamic look, higher conversion, more B2B enquiries, better UX/UI, SEO, expanded functionality and a WOW effect. Please provide a price and delivery timeline.”

Sometimes “premium” gets added. Sometimes “minimalist like Apple”. Sometimes “SEO-friendly”. One public job post genuinely asked for a “modern, premium, wow effect”, elaborate animations and “in-depth UI/UX”. Another wanted a website that was modern, minimalist, premium and conversion-focused at the same time. Somewhere else, the requirement was an “SEO-friendly”, fast, scalable website ready for future growth. Yes, these are real job posts.

Every one of those words can make sense on its own.

The problem starts when, together, they’re supposed to pretend to be a project scope, or a brief. They don’t.

“We want a modern website”

The first classic.

Modern meaning what, exactly?

Large typography? Animations? Background video? Gradients? A huge product photo? Glass cards? Elements moving as you scroll?

“Modern” is neither a functional requirement nor a business goal. It’s an aesthetic judgement, and it depends largely on who’s looking at the screen.

For one person, modern means a brutalist website without a single rounded corner. For another, it’s a pastel SaaS design with thirty cards. For a third, it’s anything with an animated headline and the word AI somewhere on the page.

Worse still, visual trends age quickly. So if the main strategy is “let’s build something that looks like websites in 2026”, we’ve just started building a website that, somewhere around 2028, will look exactly like a website from 2026. And we’re back where we started.

Don’t get me wrong – that doesn’t mean design is unimportant. It’s very important. It shapes the first impression, often builds credibility, creates information hierarchy and affects how the brand is perceived.

It just can’t be the only diagnosis of the problem.

“It needs to be more dynamic”

This is one of my favourite requirements.

More dynamic – what exactly is supposed to happen?

Should an element slide in from the left?

Should an image grow as you scroll?

Should the cursor leave a trail?

Should a section stick to the screen?

Should text appear one letter at a time?

A “more dynamic” look is about as precise a requirement as saying a car should be “more sporty”.

You can replace the engine.

You can improve the suspension.

You can fit a spoiler.

Or you can stick a red stripe on the bonnet.

All four can later be described as “more sporty”. Only one of them actually changes performance.

Websites work much the same way.

Animation can improve how information is communicated, show relationships, direct attention or simply make an interface feel nicer to use.

It can also weigh half a megabyte, block the main thread, trigger motion sickness for some users and move a button at exactly the moment someone is trying to click it.

But hey, it’ll be dynamic.

“WOW effect”

If there’s one phrase that should switch on a small warning light above a brief, it’s this one.

Real job posts manage to ask for both “premium” and a “wow effect”, leave the colour palette to the designer, and at the same time expect complex visual effects and animations.

But WOW for whom?

The business owner?

The CEO?

The designer?

The customer?

The CEO’s wife, when he shows her the website on his phone that evening?

Because those are not necessarily the same people.

If you sell exclusive events, a strong visual effect can be part of the product and the brand experience. If you sell environmental audits to industrial plants, the operations director probably didn’t come to your website to watch a headline spin around for the eighth time while scrolling.

He wants to know whether you can solve his problem.

And whether you can do it without creating three new ones.

“The website needs to feel premium”

The word “premium” in website briefs has become something halfway between a business expectation and a magic spell.

It needs to be premium?

Fine.

What makes your brand premium?

Price?

Service?

Quality of execution?

Process?

Expertise?

Exclusivity?

Materials?

Availability?

Reputation?

If we can’t answer that, the designer has to guess. In practice, that usually means lots of empty space, thin type, huge photos and fewer things on the screen.

Congratulations. We’ve just discovered the visual language of half the cosmetics industry and every other designer chair brand.

Premium is not a preset in Figma.

“Minimalist like Apple”

Another classic.

Apple can put a gigantic product image, four words and a “Buy” button on a page because users already know perfectly well what an iPhone is.

If you run a company offering fifteen technical services whose differences are unclear even to half of your prospective customers, removing ninety percent of the information from the website will not turn you into Apple.

It will turn you into a company whose offer nobody understands.

Minimalism does not mean having very little content.

It means having no unnecessary content.

That’s a big difference.

“The website needs to be conversion-focused”

This is where things get interesting.

Because conversion is not a visual style.

There is no button labelled “increase conversion”.

There is no colour that magically generates more B2B enquiries. There is no sales font called Lead Generation Sans Serif. And a scroll animation will not persuade the person responsible for procurement to sign a contract worth tens of thousands of euros.

If you want to increase conversion, you first need to know why people aren’t converting now.

Maybe they don’t understand the offer. Maybe they don’t know whether you work with companies like theirs. Maybe the website shows no proof that you can actually do what you say you can. Maybe the most important service is buried three levels deep in the menu. Maybe the form looks like a mortgage application. Maybe someone arriving from Google was looking for something completely different. Maybe the website takes five seconds to load. Yes, that’s a lot. Maybe the company replies to form enquiries four days later.

Or maybe the problem isn’t on the website at all.

Ordering a redesign to increase conversion without first checking why the current website isn’t converting is like taking a car to the body shop because it uses too much fuel.

It may look fantastic when you get it back.

It will still drink fifteen litres.

“UX/UI should increase conversion”

Sounds even more professional, because now we’ve added three fancy words.

UX.

UI.

Conversion.

Bingo.

Except UX is not about adding a CTA every screen and a half. Nor is it about a designer drawing six prettier cards in Figma.

One public job post was looking for a designer to rebuild an online shop so it would be more logical, more premium and generate better conversion. Interestingly, the ad did include at least one sensible element: an audit was supposed to happen before the design work. Paradoxically, applicants were still expected to quote the full project price upfront.

And that’s exactly where far more projects should begin.

First, we examine the problem.

Then we design the solution.

Not the other way round.

“SEO-friendly”

“SEO-friendly” is a wonderful phrase because it can mean almost anything.

One WordPress job post asked for a modern responsive design, fast loading, an SEO-friendly structure, room to expand the content, sitemap.xml and “basic SEO”.

And some of those requirements are perfectly reasonable.

But you still need to establish what “SEO” is supposed to achieve in this particular project.

Is the site supposed to build informational traffic?

Bring in local customers?

Generate leads for specific services?

Compete for a handful of major keywords?

Will it have twenty pages or twenty thousand?

Are multiple language versions planned?

Is there a content strategy?

Has anyone actually done keyword research?

Because installing an SEO plugin, generating sitemap.xml and filling in a meta description is not an SEO strategy.

That’s basic equipment.

A bit like ordering a car “suitable for driving” and listing four wheels in the specification.

“It needs to be fast”

Now this is actually a great requirement.

Until it’s followed by:

“PageSpeed minimum 95.”

On every device.

On every page.

With Google Maps, YouTube, chat, six trackers, three fonts, a huge slider and a 4K video in the hero.

That’s no longer a requirement.

That’s a negotiation with physics.

Website speed can and should be treated as part of the project. But you need to define the conditions: which pages are measured, which devices matter most, which third-party integrations have to stay, and which metrics actually matter to users.

The number in PageSpeed Insights is a diagnostic indicator.

It is not a moral certificate for a website.

“Expand the functionality where possible”

It sounds harmless.

But what exactly does that mean?

A calculator?

A client portal?

CRM integration?

A configurator?

A multi-step form?

Bookings?

Automatic proposal generation?

Chat?

A document area?

Filtering portfolio projects?

“We’ll expand the functionality where possible” is an excellent way to start a project whose scope will change every three days.

First comes “just one more form”.

Then “maybe a calculator”.

Later, “actually, it would be useful if the customer could log in”.

And in the end, the contractor discovers they’ve accidentally built half a SaaS platform for the price of a company website.

“We want more B2B enquiries”

Great.

From whom?

“B2B” is not a target audience.

A company with three employees is B2B. A corporation with three thousand employees is B2B too.

A small-business owner, a procurement manager, a CTO, a marketing manager and an operations director are all B2B customers.

Each makes decisions differently. Each needs different information. Each has different objections. They may even use completely different language to describe the same problem.

If the website is supposed to generate leads, the contractor needs to know who exactly we’re trying to persuade.

Otherwise we can build a very nice website for nobody in particular.

“Please provide a price and timeline”

Based on what?

This question regularly appears under briefs that tell us roughly one thing: the company has a website and would like a better one.

And now the competition begins.

Someone says €550.

Someone says €1,500.

Someone says €2,400.

Someone says €9,000.

And of course someone will still turn up with a price of €149.

The client sees wildly different prices for supposedly “the same thing” and concludes that the industry has completely lost its mind.

Except these are not four prices for the same thing.

One contractor priced a ready-made theme and text replacement. The second priced a custom design. The third priced analysis, information architecture, copywriting, UX/UI design, development, optimisation, analytics and testing. The fourth may have included an online shop because the brief never said there wouldn’t be one, so maybe there will be.

If you ask ten builders to quote for “a modern house of about 150 square metres, functional and energy-efficient”, you’ll also get very different prices.

And none of them will yet be a real quote.

“Budget to be decided”

That’s not always a problem.

The problem starts when the expected scope is also “to be decided”. Then both sides begin a very interesting game.

The client doesn’t want to reveal the budget because they’re worried the contractor will simply adjust the price to the maximum amount available.

The contractor doesn’t want to give a price because they don’t know whether they’re discussing a project worth three thousand, thirty thousand or one hundred thousand.

For the next forty minutes, everyone very politely avoids mentioning any number at all.

The budget doesn’t need to be exact.

A range is enough.

A budget of 3-5 thousand and one of 30-50 thousand lead to completely different project decisions. Knowing the budget isn’t only about giving the contractor a chance to “charge more”.

It allows them to design a solution that can actually be delivered.

“Inspiration: Apple, Tesla, Stripe”

Of course.

Budget: €1,000.

And before anyone gets offended – there is absolutely nothing wrong with a small budget.

The budget is not the problem.

The problem is the lack of any relationship between budget, scope and expectation.

Apple does not have a good website because somebody found the right WordPress theme, installed it, dropped in some random template copy and the site suddenly started printing money.

Stripe did not reach its current interface quality because a freelancer “made something similar” over a weekend.

You can take inspiration from a visual direction, structure, storytelling, or a specific interface solution.

But “I want something like Apple” is not a specification.

Just as saying “I want something like a Porsche” at a dealership does not tell the salesperson whether you’re interested in a 911 GT3, a Cayenne or a keyring.

“Technology: WordPress”

Why?

This is not an attack on WordPress. I use it myself. Yes, yes, the website you’re reading right now is built on WordPress. WordPress is a good solution for a huge number of projects.

But technology decisions should follow requirements.

Instead, the whole brief often starts with:

“We need a WordPress website.”

Only later does it turn out that the site needs user accounts, a complex dashboard, document generation, complicated workflows, integrations with four systems and business logic that looks far more like an application than a company website.

WordPress may still be the right solution.

But it would be nice to ask the question first.

The reverse happens too. A simple company website does not need to be built as a headless architecture spread across seven cloud services just because somebody really wanted to use the newest framework.

Technology should serve the project.

The project should not exist in order to use a technology.

The worst mistake: starting with the solution

A lot of companies come to a contractor with the diagnosis already made:

“We need a redesign.”

How do we know?

Maybe you need to improve the information structure. Maybe you need new content. Maybe the offer needs to be rebuilt. Maybe performance is the biggest problem. Maybe people don’t understand the difference between your services. Maybe campaign traffic is landing on the homepage instead of the right landing page. Maybe a potential customer does nothing because the site gives them no reason to make any decision. Maybe the current website doesn’t need a full redesign at all.

Companies often order a website the way a workshop customer might say:

“Please replace the gearbox.”

A good mechanic first asks what the car is actually doing. A good website contractor should do exactly the same.

So how should you commission a website?

You do not need to know how to build it. You do not need to know the technology. You do not need to know the names of every interface element. You do not need to arrive with a finished mock-up drawn in PowerPoint.

You really don’t need to know whether the CTA should be lime green or navy blue either.

That’s not your job.

Your job is to describe the business and the problem well.

Explain why the website exists. Who uses it. Which customers you especially want to attract. What the user should do. How sales work today. Where customers come from. What they ask most often before buying. Why they choose you, or what sends them to a competitor. Why they drop out. Which products or services are strategically important. What about the current website gets in the company’s way. Which functions are actually needed. Which systems need to be connected. What the budget is. What the constraints are.

And most importantly:

How will you know in six months that the new website is better?

Because if the only answer is “it looks more modern”, you didn’t order a business tool.

You ordered the walls repainted, possibly in the basement.

A good brief can be surprisingly simple

It doesn’t need to be forty pages long. It can sound like this:

“We work with mid-sized manufacturing companies. We’re especially interested in customers with more than 50 employees. Clients usually find us through Google and referrals. Service X has the highest value for us. Our current website generates around five enquiries a month, but most concern services we don’t particularly want to sell. Users often ask us about things that are theoretically already explained on the website. We want to increase the number of valuable enquiries and explain our offer better. We have project examples and data that we currently don’t show. The project budget is around X. We don’t know whether we need a full redesign and want the contractor to assess that first.”

You can work with that.

You can analyse the problem. You can propose a solution. You can define the scope. You can even price the project sensibly.

From “we want a modern, dynamic, premium, SEO-friendly website with good UX/UI, a WOW effect and a focus on conversion”, you can mainly build one thing: a modern, dynamic, premium, SEO-friendly website with good UX/UI and a WOW effect.

And then wonder together why nobody contacts you through it.

Don’t order a website

Order a solution to a problem.

Don’t tell the contractor to make the button bigger before you’ve checked whether anyone wants to click it.

Don’t order an animation before you’ve established what the user is supposed to understand.

Don’t demand “conversion-focused design” if you don’t know what conversion means in your business.

Don’t put WordPress in the brief just because the previous website used WordPress.

Don’t ask for a price for the whole project if you still can’t say what the whole project actually is.

And above all, don’t look for the contractor who answers, without asking a single question:

“Sure, I’ll do all of it. €1,000. It’ll be ready by the end of the week.”

You may just have found the cheapest offer.

That does not mean you’ve found a solution.