How Much Should You Charge for a Website? (What I've Learned After 100+ Projects)

Nobody taught me how to price a website. Not a single course, not a single mentor, not a single YouTube video that actually reflected reality. I figured it out the hard way: by charging EUR 250 for my first project and slowly realizing that wasn't going to pay rent.

Six years and over a hundred projects later, I still don't think I've fully figured it out. But I've gotten a lot better at it. And I've made enough mistakes to save you from a few of the worst ones.

This isn't a generic pricing guide with neat little tables. This is what I've actually learned about how much to charge for a website as a freelancer, from someone who still gets nervous every time he sends a quote.

The Uncomfortable Truth About Pricing

Here's the thing nobody tells you when you start freelancing: pricing is mostly emotional. You're not calculating some perfect number based on market research. You're guessing, then sweating, then hoping the client doesn't laugh at you.

I wrote about my first website project, the EUR 250 one. Looking back, the problem wasn't that I didn't know market rates. The problem was fear. Fear of losing the client. Fear that someone on Fiverr would do it for less. Fear that maybe I wasn't worth more.

Sound familiar?

That fear doesn't fully go away, by the way. After 100+ projects, I still hesitate before sending certain quotes. But here's what changes: you start basing your prices on data instead of anxiety. You start recognizing which clients will respect your rates and which ones are shopping for the cheapest option. And you start understanding that undercharging doesn't make you more competitive. It makes you broke.

Does Your Client Even Need a $10,000 Website?

Before we get into rate cards and pricing models, let me ask you something that might be uncomfortable: does your client actually need what you're selling?

I've been building websites for years. WordPress, Next.js, custom stacks, the whole spectrum. And here's what I've realized: most small business clients don't need a $10,000 website with 40 plugins, a page builder, and a developer on speed dial for every update. They need a fast, secure site that loads in under a second, ranks well, and doesn't cost them $50/month in hosting fees.

That site exists today. And it costs almost nothing to run.

I build static websites with tools like Astro, deploy them on Cloudflare, and the result is: zero hosting costs. Free SSL. Free database if they need one. Unlimited bandwidth. The client gets a site that's faster than 95% of WordPress installations out there, and they're not paying a cent to keep the lights on.

Most people have no idea this is possible. And I mean that literally. The average business owner still thinks "website" means WordPress, a shared hosting plan, and a developer who charges them every time they want to change a phone number. That's not their fault. Nobody told them there's another way.

Here's what frustrates me: the general public is still using AI like it's a fancy Google search. They type a question, get an answer, and think that's it. They have no concept of what's actually possible. You can use Claude to build an entire website, set up a content pipeline, automate SEO audits, generate reports. Not in theory. Right now. I do it every day.

And then there's the European angle. While the US and China are racing ahead with AI development and deployment, the EU is busy writing regulations. Don't get me wrong, some regulation makes sense. But the gap is widening. The US produced 40 foundation models. China built 15. All of Europe? Three. European companies are falling behind not because they lack talent, but because the regulatory environment treats innovation like a liability.

This is why I started offering AI consulting. Because what most clients actually need isn't a developer building them a website. They need someone to set up the right system, train their team, and get out of the way. I build these kinds of setups for clients: static sites, automated content workflows, CMS integrations, all running on infrastructure that costs them nothing. And after onboarding, they run it themselves.

If you're a freelancer reading this, think about what you're really selling. Maybe it's not hours. Maybe it's not even a website. Maybe it's showing a client that the thing they were about to spend $15,000 on can be done for a fraction of that, and they'll trust you more for being honest about it.

Hourly vs. Project-Based vs. Value-Based: What Actually Works

There are three ways to price your work, and most freelancers cycle through all three as they gain experience.

Hourly billing is where most people start. It's simple: you work, you track time, you send an invoice. Great for maintenance work, consulting calls, and projects where the scope is genuinely unclear. The problem? It penalizes you for getting faster. The better you get, the less you earn per project. That's backwards.

Project-based pricing is where most freelancers land after a couple of years. You estimate the hours, multiply by your target rate, add a 15-20% buffer for the inevitable scope surprises, and quote a flat fee. Clients love the predictability. You love not having to justify every hour. The risk is underestimating the work, but you get better at that with experience.

Value-based pricing is the one everyone talks about but few actually pull off. The idea: price your work at 15-25% of the client's expected first-year revenue from the project. If you're building a website that will generate $200,000 in its first year, charging $30,000-$50,000 makes sense. The catch? You need to understand the client's business well enough to quantify the value. That takes trust and a proper discovery process.

Here's what the data shows: value-based freelancers earn a median of $96,000 versus $58,000 for hourly freelancers. That's a 66% gap. Among freelancers earning over $150,000, only 8% use hourly billing. The shift away from hourly tends to happen around the $80-100k income mark.

I'll be honest: I use a mix. Hourly for small maintenance tasks. Project-based for most website builds. Value-based when the project clearly ties to revenue and the client gets it. Most experienced freelancers do the same.

Real Numbers: What Websites Actually Cost

Let me give you the numbers I wish someone had given me when I started.

Here's what freelancers are charging right now:

• Simple brochure site (5-7 pages): $1,500-$5,000
• Custom business site (with integrations, forms, CMS): $5,000-$15,000
• E-commerce or complex web app: $10,000-$35,000+

Agencies charge more because they have more overhead:

• Boutique agency (2-5 people): $6,000-$12,000
• Full-service agency: $12,000-$35,000+

Hourly rates by experience level:

• Beginner (0-2 years): $25-50/hr
• Mid-level (3-5 years): $50-100/hr
• Expert (6+ years): $120-200+/hr

Geography matters too. A senior developer in the US charges $78-125+/hr. In Western Europe, it's $64-108/hr. In Eastern Europe, $25-65/hr. I'm in Germany serving mostly US clients, which puts me somewhere in between.

I wrote about website costs from the client's perspective a while back. This is the flip side of that conversation. What clients pay and what you should charge are two different discussions, and understanding both makes you better at the pricing conversation.

The Race to the Bottom (And Why I Almost Joined It)

Early in my career, I competed on price. I thought that was how you won clients.

I spent years on Fiverr, which I've written about in detail. The platform takes a 20% cut. After that and taxes, a $5,000 project doesn't feel like $5,000 anymore. It feels like $2,800. And at $2,800 for a custom website, you're working for less than minimum wage once you factor in the actual hours.

Here's the math that changed my mind: over 50% of freelancers admit to undercharging by at least 30%. The top 20% earn $80-200/hr while the average sits at $20-50/hr. That's not a skill gap. That's a pricing gap.

The undercharging cycle works like this: low prices attract price-sensitive clients who are often the most demanding. You need more volume to make ends meet. More volume means less time for learning and improving. Less improvement means you stay in the low-value bracket. Repeat.

Breaking out requires something uncomfortable: quoting a number that scares you a little. Not by 5%. By 30-50%. And seeing who still says yes.

When to Raise Your Rates (And How to Do It Without Losing Sleep)

The income progression I've seen, both in my own career and across the industry, looks roughly like this:

• Starter (0-2 years): ~$52,000/year
• Growing (3-5 years): ~$82,000/year
• Established (6-10 years): ~$112,000/year
• Expert (15+ years): ~$172,000/year

A good target: increase rates by 10-20% every 12-18 months. Not because inflation demands it (though it does), but because you're genuinely getting better and delivering more value.

Here's how I do it: test higher rates on new inquiries first. If people still say yes, the rate is right. If you're getting 100% acceptance on your quotes, you're probably too cheap. Aim for a 60-70% close rate. That means you're pushing the boundary without pricing yourself out.

For existing clients, give 30-60 days notice. Frame it as an investment in quality: "I'm allocating more time per project to ensure better results." Most will accept. The ones who leave were probably price-shopping anyway.

Two things that let you charge significantly more: specialization (specialists command 2-5x what generalists charge) and AI proficiency (freelancers who mention AI skills in their profiles charge about 25% higher on average). If you're good at something specific and you're leveraging AI tools to deliver faster, you have pricing leverage. Use it.

The Pricing Conversation: How I Talk Money With Clients

The question I dreaded most as a new freelancer: "So, how much will this cost?"

These days, I don't dread it. Here's why: I stopped leading with the price and started leading with the value.

My approach: before quoting anything, I run a proper discovery call. Understand what the client needs (not what they think they need, which is often a different thing). Then I ask the question that changes everything: "How much is a single new customer worth to you?"

If a new customer is worth $500 and the website will bring in 20 new customers per month, that's $10,000/month in value. Charging $8,000-$12,000 for the website is a fraction of the first-year return. Now we're talking value, not cost.

When they say "that's too expensive," I don't panic. I ask what their budget is. Sometimes the gap is negotiable. Sometimes it's not. Either way, I'd rather lose a project than do it for a price that makes me resent the work.

One more thing: retainers. 78% of agencies now use retainer-based pricing, up from 64% a couple of years ago. For freelancers, a retainer means predictable monthly income. For clients, it means ongoing support without the hassle of new proposals every time something needs updating. I pitch retainers after every website launch, and about half my clients take them. It's the most stabilizing thing I've done for my cash flow.

If you're curious about how to frame ongoing work like SEO as a retainer, I've written about that too. And if you want to protect yourself from scope creep during handoffs, I'd start there before quoting your next project.

Mistakes I've Made (So You Don't Have To)

I'll keep this honest. These are real mistakes from my real career:

Not factoring in business expenses. Software subscriptions, hosting for dev environments, hardware, insurance, marketing, accounting. That's $2,000-$8,000 per year in overhead. If your rate doesn't cover it, your take-home is lower than you think.

Basing my rate on my old salary. If you earned $60,000 as an employee and charge the equivalent as a freelancer, you're losing money. Freelance rates need to be 2-3x your salary equivalent to cover taxes, overhead, non-billable time, and the fact that only 50-70% of your hours are actually billable. The rest goes to admin, prospecting, and professional development.

One-size-fits-all pricing. A brochure site for a local bakery and a web app for a funded startup should not cost the same. It took me too long to understand that the client's budget and the project's business impact should influence the price, not just the hours.

Not tracking time on fixed-price projects. You can't improve what you don't measure. I once realized a "quick" $5,000 project had absorbed 20 extra hours of unbilled scope creep, dropping my effective rate from $125/hr to $83/hr. Now I track everything, even on flat-fee work.

Not questioning what the client actually needs. This is the big one. I used to just build what was asked. WordPress with WooCommerce, custom theme, dozen plugins. Now I ask: do you really need all this? Sometimes the answer is a static site that costs nothing to run and does exactly what the business needs. I stopped using WordPress for a reason, and part of that reason is watching clients pay for complexity they never needed. WordPress keeps adding features nobody asked for while the basics still feel clunky.

Here's the formula that finally made it click for me: to take home $80,000/year after taxes, with $5,000 in overhead, 1,200 billable hours, a 10% platform fee, and a 30% effective tax rate, you need to charge approximately $112/hr. When I first did that math, I understood why I'd been struggling.

I still get it wrong sometimes. Last month I quoted too low on a project because the client seemed nice and I wanted the work. Three weeks in, I was underwater. Pricing is a skill, not a formula. You keep developing it.

If you're figuring out your own pricing, or if you're a business owner trying to understand what fair rates look like, let's talk. I've been on both sides of this conversation enough times to help you find a number that works.