How to Write Website Copy That Converts

Most people write for the web the same way they write for print. That’s a mistake, and it’s one of the most common web copywriting errors I see.

I learned this the hard way. Coming from screenwriting, I was trained to think in sequences. That works brilliantly on screen, but on a webpage? People are gone before your setup lands.

Writing for the web is a different discipline. Here’s what I’ve come to rely on.

Start with context

Web copywriting starts with orientation, not information. Readers arrive at your page from somewhere, such as Google search, a social share, or a newsletter link. They don’t know where they are yet, and they’re deciding in milliseconds whether to stay.

That means your first job isn’t to inform. It’s to orient. Give them a clear signal that they’re in the right place, for the right thing, at the right time. Structure your page so the most important message appears first, not buried three paragraphs down after your company history.

Think of it like a film opening. You establish the world before you introduce the conflict.

Write short. Then cut it in half.

On screen, people don’t read — they scan. Long paragraphs feel like walls. Even if the content is genuinely good, dense formatting signals “this will take effort,” and effort is the enemy of engagement.

Keep sentences short. Break up paragraphs after two or three lines. If you can say it in ten words instead of twenty, say it in ten. Every unnecessary word is a small tax on the reader’s attention, and attention is the scarcest resource you’re working with.

This doesn’t mean dumbing things down. It means respecting your reader’s time.

Make your headings do real work

Headings aren’t decoration. They’re navigation.

A reader scanning your page is asking “Is this worth my time?”. Your headings answer that question. If they’re vague (“Our Approach,” “Learn More,” “Why Us”) you’ve given them nothing to hold onto. If they’re specific and informative (“Three reasons your homepage copy isn’t converting”), the reader knows exactly what they’re getting before they commit to reading.

Write headings that make sense out of context. Because for a scanner, they often are the context.

Use bullet points strategically

Bullet lists are useful when you have genuinely parallel items: a checklist, a list of features, a set of options. They make comparison easy and reduce cognitive load.

But bullets can also fragment content that should flow as prose. If you’re using bullet points because you can’t figure out how to connect your ideas, that’s a writing problem and not a formatting solution.

Use them when they earn their place. Not as a default.

Visual readability is a writing problem, not just a design problem

How a page looks affects how the text is received. That’s a writer’s concern, not just a designer’s.

Line length matters. Paragraphs that stretch the full width of a screen are exhausting to read. White space isn’t wasted space, as it gives the eye somewhere to rest. Images, when relevant, give the reader a mental anchor between blocks of text.

When I write website copy for a client, I think about how the text will sit on the page, not just what it says. Word count, paragraph length, and section breaks are all tools.

Write for everyone

Accessible web copywriting isn’t a niche consideration. It’s good writing.

Use plain language. Avoid jargon unless your audience genuinely uses it. Don’t rely on color or layout alone to convey meaning. Write link text that makes sense on its own. “Click here” tells the reader nothing, “Read the full case study” tells them exactly where they’re going.

Small detail. Real difference.

Social media is its own dialect

Web writing principles carry over to social, but the context shifts. On social, you’re competing with everything else in someone’s feed. The first line has to earn the next one.

I write social copy the same way I write headlines: what’s the one thing I want this person to take away? Everything else is support.

The 7 web copywriting principles I come back to

If someone asked me to summarize everything above into a checklist, it would look like this:

1. Orient before you inform — context comes first

2. Write short sentences and short paragraphs

3. Use headings that work out of context

4. Use bullet points only when items are genuinely parallel

5. Think about how the text sits on the page, not just what it says

6. Write link text that describes the destination

7. On social, the first line has to earn the second
   

The bottom line

Good web copywriting rewards clarity, structure, and a ruthless willingness to edit. It punishes assumptions about how much time or patience your reader has.

Get the structure right first. Then cut. Then make it scannable. Then read it out loud. If it sounds natural, it probably reads naturally too.

That’s the system I keep coming back to.

Every great brand has a story worth telling.

I became a copywriter because I was already a filmmaker. A screenwriter. Someone who spent years obsessing over the gap between what people say and what they actually mean. Over the small gestures that reveal character. Over why some stories stay with you for decades, and others dissolve before you’ve even closed the tab.

That’s a discipline, and it shapes every single word I write.

Trained by film. Shaped by people.

Screenwriting teaches you something most copywriters never learn: the most important thing in any scene is almost never what’s being said.

A character says “I’m fine.” What they mean is: I am holding myself together and I need you to see that without making me say it out loud.

That’s subtext. And I am obsessed with it.

Because good copy works exactly the same way. The best campaigns don’t tell you what to think. They create a moment of recognition with the reader. That recognition is what builds loyalty. Not a clever headline. Not a conversion-optimized CTA. A moment where a real person felt genuinely seen.

You can’t generate that. It requires someone who has been paying close attention to people for a very long time.

That’s what I bring to the table.

Thinks in story arcs. Writes in search-optimized prose.

Most copywriters start with the keyword. I start with the character. Who is this person, what do they actually want, and what are they really afraid of?

From there, structure follows naturally. The hook. The tension. The turn. The resolution. It’s not a framework I learned in a marketing course. It’s what happens when you’ve spent years writing scripts where every scene has to earn its place or it gets cut.

That discipline is rare in content right now. And it’s becoming more valuable by the day.

Because when execution is cheap the question of what those words should mean, and to whom, and why, becomes the only question that matters.

Story first. Strategy second. Always.

Interested in content that lands.

Scroll through any feed right now. The same minimalism that isn’t minimal, it’s just empty. Content that performs the idea of having something to say without ever quite getting there.

And then, occasionally, something cuts through. A piece of writing that makes you send it to someone with no comment except read this. A brand that makes you feel like the people behind it are actually alive and real.

Those moments are becoming rarer. And they are becoming more valuable.

Meaning over volume. Always.

The agencies and clients who still believe in this are out there.

I know you exist. You’re the ones who still lose sleep over whether a campaign means something.

You are going to win. Not because creativity is some romantic ideal worth protecting, but because humans are hardwired to respond to story, to honesty, to the feeling that the thing they’re reading was made by someone who actually cared.

The printing press didn’t end writing. Photography didn’t end painting. And AI won’t end the one thing it cannot replicate: a perspective that comes from a life genuinely lived and paid attention to.

Don’t buy the panic. Read a book. Watch a film. Spend time with people you love.

The future is bright, but only if we let it be.

Stop posting AI banners. It's not a flex. It's a red flag.

When I see a business post a social media banner that was clearly generated by AI I don't think efficient. I don't think innovative.

I think that they didn't care enough to try.

And if they didn't care enough to try with their own image, the thing they chose to represent themselves publicly, what does that tell me about how much they'll care about the actual product or service?

Nobody Cares What You Used. They Care What It Says About You.

There's a difference between using a tool intelligently and using it as a substitute for giving a damn.

An AI banner says three things simultaneously:

1. We don't think our visual identity is worth investing in. You couldn't spend two hours with a designer, or even find a decent template and customize it. You generated something in 40 seconds and called it done.

2. We assume our customers won't notice (meaning: we don't care if they do). This is the part that really gets me. It's not even cynicism. It's indifference. And customers feel indifference. They may not be able to name it, but they feel it.

3. We have no backbone. You chose to represent your brand with something that has no perspective, no voice, no trace of an actual human decision. A placeholder dressed up as an identity.
   

Your customers are not stupid.

The comments sections on posts like this are consistent. People notice. They're turned off. They're simply scrolling past and never coming back.

If a company can't be bothered to create real content, why should we be bothered to become a real customer?

That's the direction the whole market is moving. As AI-generated content floods every platform, authenticity (actual human intention behind a visual) is becoming a premium signal. A trust marker. The brands that understand this are already pulling ahead.

The irony is that it costs you more than it saves you.

You saved what? A few thousands kronor on a graphic? And in exchange you communicated to every person who saw it that your brand is lazy, cheap, and doesn't believe in itself.

That's not efficiency. That's a slow leak in your credibility.

If you genuinely can't afford a designer right now, there are tools that let a non-designer make something that looks considered and intentional. Use a template and make it your own. Show up with something that has a pulse. Put in the effort.

But please, stop mistaking automation for strategy. Stop confusing fast with good. And stop insulting your potential customers by assuming they won't tell the difference.

They can. They do. And they remember.

Why Your Click-Through Rate Collapsed in 2026

It's become a recurring conversation in 2026. A client calls their digital marketing agency. Conversion rates are down. Click-through rates have collapsed. Traffic looks like an ECG on a patient just taken off life support. The question is always the same: What's happening?

What's actually happening is three things at once. And you can't fix them with more blog posts, a new ad budget, or an A/B test on button colour.

AI Has Stolen the Front Page of the Internet

In September 2026, internal documents from Microsoft's lawsuit with The New York Times revealed something the industry already knew but had never seen in black and white: when users search via Copilot instead of regular Bing, clicks to external websites drop by up to 93 percent. A Microsoft researcher called it a "doom loop" for the web. An OpenAI engineer put it even more plainly back in 2023: "No matter how prominently we show the links, users won't click."

Google AI Overviews answers directly. Perplexity cites one source and never sends traffic onward. ChatGPT closes the question.

According to HubSpot's State of Marketing 2026, 92% of marketers are now using, or planning to use, optimisation for AI search engines. This is no longer a niche strategy. It's the baseline requirement. And yet most agencies are still counting organic sessions as if it were 2019.

The new metric isn't the click. It's the citation. The brand mentioned in the AI answer wins visibility, credibility and recognition, without generating a single session in Google Analytics. Marketers who understand this now track how often their brand appears inside AI-generated answers. Those who don't are still optimising for a world that no longer exists.

The Middle Class Has Essentially Disappeared as a Buyer

A global recession is underway. Disposable income has shrunk dramatically. People who used to click, compare prices and add products to digital shopping carts are now running a completely different calculation every morning: bills, food, rent. What used to be an obvious purchase is now an active decision not to buy.

The middle class, the group that for decades drove consumption of everything from subscription services and home furnishings to fitness equipment and digital tools, is in a financial position it never anticipated. Savings are gone. Margins are gone. And with them: the impulse purchases, the loyalty programmes, and the digital consumption that entire business models were built on.

Clients calling their agencies and asking about conversion drops are forgetting to ask the more fundamental question: Can my customer actually afford this right now?

It's not the campaign's fault. It's not the copywriter's fault. It's not even the algorithm. It's that the economic reality of your target audience looks fundamentally different in 2026 than it did in 2021.

The Agencies Don't Have the Answer

On Reddit, a Head of Marketing at a B2B company put it better than most analyst reports: "2026 feels like a completely different year. Not a weak quarter. Not one underperforming channel. Fundamental changes across the entire funnel."

What he's describing isn't an anomaly. It's the new normal.

And in the comment threads under that post, and hundreds like it, a pattern emerges: agencies that added a slide about "Generative Engine Optimization" to their sales deck and then kept doing exactly what they were doing in 2022. More blog posts. More backlinks. More AI-generated content they hope will rank — in a world where AI engines actively prioritise human voices, Reddit threads, third-party reviews and established brand authority over machine-produced articles.

One commenter put it plainly: "If an agency claims they can game Perplexity the same way they gamed Google in 2018, they're selling snake oil with a new buzzword on the invoice."

The Expertise AI Can't Replicate

In the middle of all this chaos, demand for one specific skill set is growing.

People who understand what communication is actually supposed to do.

AI cannot understand what a human needs to hear in 2026, when the economy is tough, the future feels uncertain, and trust in brands is at a historic low.

A copywriter with genuine expertise does three things no prompt can replicate. They understand the intent behind the search query, they write for passage indexing and AI citation with precision, and they choose the one that actually lands with a real person in a real situation.

What This Requires From You as a Brand

The clicks are not coming back the way they were. That's not a forecast. It's a statement of fact.

The new metrics are citation, trust and relevance at the right moment. It requires content specific and well-structured enough for AI systems to choose to cite it. It requires presence beyond your website: on Reddit, in industry publications, in third-party reviews, because that's where AI engines actually look when building their answers.

And it requires a copywriter who understands all of this. Not a prompt operator. A strategist who uses language as a tool.

That's a rare combination of skills. And it isn't cheap. But it's cheaper than continuing to pay for dashboards measuring a reality that stopped existing two years ago.

Access Is Not Expertise.

Why Your Brand Sounds Like Everyone Else

When did everyone become a copywriter?

When did a room full of people who had never written a campaign decide they had all it figured out?

I think it was around the time AI produced something that looked impressive on first read. Impressive enough to convince a room that the hard part was done. Not quite impressive enough to do the actual job.

What the tool actually gave us

Something happened when the AI tools became accessible to everyone at the same time.

Suddenly the client arrived at the briefing with three headline options they were already attached to. The marketing manager rewrote the copy themselves and wanted to know why my version was different. The founder told me, with genuine enthusiasm , that Claude had basically handled the About page. They just needed me to check it over.

I understand the logic. The output maybe looked professional. If you have never spent years thinking carefully about what makes communication actually work, it can genuinely be difficult to see what’s wrong with it.

What’s wrong with it is that it sounds like everyone else. Because it was trained on everyone else.

The brief that puts you in an impossible position

It is not just clients who show up with AI output and call it done. It is the brief that arrives asking you to use AI to make the process faster.

So you do. You use the tools carefully. You rewrite draft after draft, each one slightly closer to something real but still not quite there. Still carrying that particular quality of having been averaged from everything that came before it.

Nobody is satisfied. The client is frustrated and the team is frustrated. But somehow the question that surfaces is not whether the process was the problem. It is whether you are good enough at your job.

I have been doing this for years. And in that moment, sitting across from a revision note that questions your craft because the AI draft felt more efficient, it’s difficult not to feel like the ground has shifted under you.

Not because the criticism is fair. But because the room stopped asking whether the process was the problem.

Numbers Are Safe. Ideas Are Not.

And underneath all of this, something else is happening.

There is a slow, passive-aggressive erosion of the creative role itself. Not confrontational, but just a steady accumulation of moments where the question is implied rather than asked; do we actually need this? Do we need someone whose process takes time, whose best work cannot be scheduled, whose output is difficult to measure in a spreadsheet?

The people asking that question, quietly in budget meetings and capacity planning sessions, are rarely the ones whose roles are under the same scrutiny. The sales team can show their numbers. The account managers can show their pipeline. The data analysts can show their dashboards. These roles are legible in ways that creative work has never been. And in an environment where AI can now produce something that looks like creative work in thirty seconds, the argument for protecting the time and space that creative work requires becomes harder.

What I find most telling is who feels safe in this moment. The people pushing hardest for AI-first creative processes are almost never the ones whose jobs are being replaced by them. They are optimising someone else's role out of existence while quietly protecting their own.

What copywriting actually requires

Let me be direct about something most people who hire copywriters do not fully understand.

Copywriting is not filling in the text. It is not making things sound professional. It is the strategic application of language to produce a specific effect in a specific person at a specific moment in their relationship with a brand.

That requires knowing the person reading. What they are afraid of, what they want to believe about themselves, what would make them trust you versus what would make them leave. It requires knowing the brand, not the approved adjectives in the guidelines, but the actual specific truth of what makes it different.

It requires judgment. The ability to choose, from among many technically correct options, the one that is actually right. The one that earns attention rather than just receiving it.

That judgment is not a talent you either have or do not have. It is built through years of work. Through the accumulated understanding of what an audience does when they encounter something real versus something that resembles something real.

A prompt cannot replicate that. It can produce output that looks like the product of that judgment. But it cannot produce the judgment itself.

And when you ask a professional to work inside a process that removes the conditions for that judgment to operate, you are not getting faster access to their best work. You are getting a constrained version of what they could produce, filtered through a workflow designed for something other than quality.

The result is generic, and the professional knows it is. But nobody wants to say it out loud.

What it does to your brand over time

The specific quality that made a brand recognizable starts to disappear. The edges that gave it character get averaged away, and it begins to sound like a slightly different version of every other brand in its category running the same tools through the same prompts.

The audience always notices this before the brand does. People rarely articulate why a brand feels different than it used to, but they feel it. The sense that something has changed and that the communication is coming from somewhere less real.

Trust disappears slowly until the cumulative effect shows up in numbers that nobody immediately connects to the decision made a year earlier to handle the copy in-house with a tool.

The question worth sitting with

Before you decide the copy-problem is solved, ask this.

Does this really sound like us? Or does it sound like a very competent version of our category?

And before you question the professional sitting across from you, did you give them the conditions to produce something genuinely good? Or did you give them a tool, a tight timeline, a brief that was already half-answered by a model, and then express surprise when what came back felt like everyone else?

The process produces the output.

If the output is not good enough, the honest question is whether the process was ever designed to produce something good from the start.

Access is not expertise. It never was, and the brands that understand that difference are the ones still worth listening to.

AI Can Write Your Copy. It Can't Write Your Character.

Let me say something that might sound strange coming from someone who uses AI everyday.

I use AI in my work constantly to pressure test ideas, to structure thinking, to move faster than I could alone. But there is something AI cannot do. And right now, in a marketing landscape where everyone has access to the same tools producing the same output, that something is the only thing that actually matters.

AI cannot write your character.

What character actually means

In film, character is not personality. It is not tone of voice or a list of adjectives on a creative brief. Character is the specific set of beliefs, contradictions, fears and commitments that make a person, or a brand, fundamentally themselves.

You know a great film character not because they are consistent. You know them because they are specific. Because there is a way they see the world that could belong to no one else. Because when they speak, you recognise something true. Something that couldn't have been invented, only discovered.

That is what the best brands have always had. And it is exactly what AI cannot generate.

AI is trained on everything. That is its power and its fundamental limitation. It produces the most statistically likely version of any given thing. The average of all the voices it has absorbed. It optimizes toward what has worked before.

Character, by definition, is not the average. It is the specific. The unexpected. The thing that makes you unmistakably you.

What is actually happening right now

I read a lot of brand content. It is part of my job. And the pattern I see emerging is this: as more companies hand their content over to AI, the output becomes more technically correct and less worth reading.

The sentences are cleaner. The structure is tighter. The keywords are in the right places. And there is nothing in any of it that you would remember tomorrow.

Because memory requires specificity. And specificity requires someone, a real person with a real perspective, to have been brave enough to say something specific.

The brands disappearing into the noise are not disappearing because they are using AI. They are disappearing because they have outsourced the one thing that could never be outsourced. The decision about who they actually are.

What I learned from directing actors

When I directed films, the hardest part was never the technical execution. The hardest part was helping an actor find the specific truth of a character. The thing underneath the lines that made everything they said feel inevitable rather than performed.

You cannot tell an actor to be more authentic. Authenticity is not a direction. It is a result. It comes from specificity. From knowing exactly what this character believes, fears, wants and is unwilling to give up.

The same is true for brands. You cannot brief your way to character. You cannot A/B test your way to a soul. You have to do the harder work of actually knowing who you are, and then having the courage to say it out loud, consistently, even when it means not sounding like everyone else.

The competitive advantage nobody is talking about

In a world where every company has access to the same AI tools producing the same quality of technically correct content, character is the only differentiator left that cannot be replicated.

You can replicate a tone of voice. You can replicate a content format. You can replicate a posting frequency. You cannot replicate the specific accumulation of decisions, failures, beliefs and commitments that make a brand genuinely itself.

That is your moat. Not your product features. Not your SEO strategy. Not your content calendar.

Your character. The thing AI can describe but never become.

Where I come in

Helping brands find and hold onto what makes them specific is the work I care about most. If your content is starting to sound like everyone else's; technically fine, strategically sound and somehow forgettable, I am happy to have that conversation.