As Good Gets Easier, Great Becomes More Valuable

There has never been an easier time to make something that looks pretty good.

You can build a polished website without knowing how to code, generate a credible identity direction in an afternoon, produce hundreds of images without a photoshoot, and get a reasonable first draft of almost any piece of writing by describing what you want in a prompt.

For most of the history of creative work, simply reaching that level required much more specialized knowledge. You needed access to people who knew how to design, write, photograph, animate, or build, and even fairly ordinary execution took time.

That is changing quickly.

AI is part of it, but so are templates, no-code tools, component libraries, design systems, better software, cheaper computing, and decades of accumulated knowledge becoming easier to access.

All of these things are lowering the difficulty of producing competent work.

It is tempting to conclude that this makes good design, writing, and creative execution less valuable.

We think something more interesting is happening.

The baseline is rising.

Good used to be harder

A lot of what we once interpreted as quality was partly a consequence of difficulty.

A website that loaded smoothly, worked properly on different devices, had thoughtful typography, and included sophisticated interaction was difficult enough to produce that the execution itself created separation.

The same was true elsewhere. Professionally produced photography required equipment and technical knowledge that most people did not have. Publishing something that looked convincingly designed required access to software and somebody who knew how to use it. Even getting an idea from a sketch into a polished piece of communication involved a significant amount of production work.

Better tools gradually removed some of those barriers.

This has happened repeatedly enough that it should feel familiar by now. Digital cameras made technically competent photography available to far more people. Canva made it possible for almost anyone to produce a reasonably polished graphic. Website builders and templates made layouts that once required specialists accessible to anyone willing to spend an afternoon learning a tool.

Each time, more people gained the ability to produce acceptable work.

What did not happen was the disappearance of exceptional work.

The standard simply moved.

The baseline is moving much faster now

Generative AI accelerates this because it does not only make production easier. It makes many forms of production available through ordinary language.

People no longer need to know exactly how something is made before attempting to make it.

That is an enormous change.

A founder can describe an interface and generate code for it. A marketer can produce dozens of visual concepts without commissioning them individually. Someone who has never written professionally can create competent copy, and a designer can explore directions that would previously have required a much larger amount of execution before they were visible enough to judge.

For creative work, this means the level that once looked impressive simply because it was difficult to reach becomes increasingly ordinary.

A clean layout is easier. Plausible copy is easier. Functional code is easier. An attractive image is easier.

None of those things have stopped mattering, but they carry less differentiation on their own when everybody has greater access to them.

Better tools also create more sameness

There is another consequence of lowering the production barrier.

When many people use similar tools, similar references, similar patterns, and increasingly similar models to solve similar problems, competent work can begin to converge.

You can already see this in certain kinds of websites and branding.

They are well made. The type is good, the grids are disciplined, the animation is smooth, and the words are perfectly reasonable. There may be very little to criticize from a production point of view.

There is also very little to remember.

This is not a problem unique to AI. Templates have always pushed people toward familiar structures, design trends spread quickly, and entire categories have spent years borrowing the same language from one another.

AI simply makes convergence cheaper and faster.

If thousands of people begin from roughly the same references and ask increasingly capable systems for a “modern technology brand,” we should not be particularly surprised when some of the answers feel related.

The question then moves beyond whether something is well produced.

You have to ask whether it could belong to anybody else.

The scarce part moves upward

As the mechanics of producing competent work become easier, the difficult decisions do not disappear.

Some become more important.

You still have to decide what deserves to exist in the first place, which idea has enough potential to pursue, what belongs specifically to this company, what people should understand, and what they should feel when they encounter it.

You need to know which references are useful and which ones are simply fashionable, what should be removed when the work contains too much, where a surprising idea strengthens the brand and where it becomes a distraction, and whether a technically excellent result actually says anything worth remembering.

Those are judgment problems.

They do not depend on whether the final execution was produced manually, with AI, through a no-code tool, or with some combination of all three.

A bad idea does not become better because it took somebody three weeks to execute by hand. At the same time, making that bad idea beautifully and instantly does not make it good.

The value sits in knowing the difference.

Taste becomes visible when there are more options

Creative tools are also increasing the number of possible answers available to us.

A writer can generate dozens of approaches to a headline before lunch. A team can test different structures, images, interactions, or bits of code before committing significant production time to any of them.

More options sound unambiguously useful, but they create another problem: somebody still has to choose.

That choice is where taste becomes more obvious.

Taste is sometimes described as if it were a mysterious instinct possessed by certain creative people. In practice, a lot of it comes from being able to distinguish between things that are superficially similar in quality and understand why one answer has more potential than another.

When producing options was expensive, the ability to generate them could itself be valuable. When options become abundant, the ability to recognize which ones deserve to survive becomes much more important.

This is one of the reasons we do not find the argument that AI “removes the need for designers” particularly convincing.

If anything, having a machine produce twenty ok answers makes the person deciding which answer is right more consequential.

Great work is specific

Specificity is probably one of the most important things that remains difficult.

A lot of competent creative work works because it borrows conventions people already understand. A fintech company looks recognizably fintech. An AI startup looks recognizably AI. A venture fund looks reassuringly like a venture fund.

There is value in familiarity because communication becomes difficult if everything is reinvented for the sake of originality.

The problem begins when familiarity is doing all the work.

A strong brand needs enough category recognition for people to understand what they are looking at, while having enough character that they can tell whose version of it they are seeing.

That requires knowing something specific about the company.

Who are these people, what do they believe, what do customers actually value about them, how do they compete, what should somebody feel when meeting them, and which parts of that are interesting enough to become visible?

You cannot answer those questions by asking for a better-looking version of whatever the category already does.

The more generic production becomes available, the more important those specific decisions become.

Emotional effect does not disappear because something is generated

Another strange version of the AI argument assumes that creative work matters because a human physically produced it.

That has never seemed like a particularly useful definition of quality.

People experience the result.

They respond to an image, an interaction, a story, a piece of music, a product, or a brand because of what it makes them notice, understand, remember, or feel. The tools used to produce it are part of the context, but they are rarely the entire reason the work matters.

Research into people's responses to AI makes this more complicated than a simple human-versus-machine argument anyway.

For example, experiments on algorithm aversion have repeatedly found that people can react more harshly to errors attributed to algorithms than to similar mistakes made by humans. One study found lower acceptance, more negative reactions, and stronger behavioral responses when an error was attributed to an algorithm rather than a person.

Research into AI-generated marketing communication also suggests that context matters considerably. A 2025 study found that when people believed emotional brand messages had been generated by AI rather than humans, measures including positive word of mouth and loyalty declined, with perceived authenticity helping explain the effect. The same penalty became weaker for factual communication or when AI was used primarily for editing.

That does not tell us to avoid AI. It tells us that people are sensitive to what they experience and to the context surrounding it.

The bar for something that wants to create trust, attachment, desire, or emotional connection may therefore become more demanding rather than less.

Craft is more than difficulty

Creative people sometimes make the opposite mistake and defend craft by defending effort.

Something took a long time, required a specialist, or was made manually, therefore it must be more valuable.

That argument becomes difficult to sustain when a new tool can achieve the same result faster.

The better reason to care about craft is that details change the experience.

The pace of an animation affects how something feels. The way type is set affects how easily people read and what kind of company they imagine behind the words. The transition between two sections can make a story feel continuous or make the website feel like a collection of unrelated blocks. A sentence can be technically correct and still sound exactly like every other company in the category.

Good creative work involves noticing those differences and deciding which ones matter.

Sometimes AI will help make them better. Sometimes a component system will save days of unnecessary work. Sometimes the right answer will still require someone spending an unreasonable amount of time adjusting something most people would struggle to describe afterward.

The method is secondary.

The standard is what matters.

Knowing when good is not good enough

This may become one of the most valuable creative skills as the baseline rises.

Modern tools are extremely good at getting work into the territory of plausibility.

The copy makes sense. The composition works. The website functions. The image looks good.

At that point, there is a strong temptation to stop because there is nothing obviously wrong.

Exceptional work often begins slightly beyond that point.

You look at something competent and ask whether the idea could be more specific, whether the obvious reference is doing too much of the thinking, whether the language has any character, or whether an interaction is there because it contributes something or simply because the tool made it possible.

The ability to recognize that technically good work still has somewhere to go is difficult to automate because the decision depends on what the work is trying to become.

You need a point of view before you can know whether you have reached it.

The gap between good and bad may shrink

This is where we think the creative market gets interesting.

Tools can reduce the gap between people who have access to production skills and people who do not. They can help inexperienced teams reach a level that previously required much more expertise, while allowing experienced teams to move through routine production much faster.

That is good.

There will be more competent work in the world.

At the same time, competence becoming common changes what competence is worth as a differentiator.

The meaningful gap shifts upward toward work that has something particular to say and knows exactly how it wants to say it.

This is why we think design, creativity, and craft become more important as production becomes easier.

The baseline is rising, the ceiling is not coming down.

What this means for brands

For brands, the implication is very straightforward.

Looking professional is the starting point.

A startup still needs a good website, clear communication, strong typography, functional technology, and a coherent visual identity. None of those requirements disappear simply because tools have made them easier to produce.

The difference is that meeting those requirements no longer guarantees that anybody will remember you.

When credible competitors can reach the same baseline, the work above it carries more weight: a sharper idea, a more specific point of view, better taste, stronger emotional direction, and the craft required to make all of those things visible in the details.

This is also why differentiation cannot be solved by adding originality at the end.

The distinctiveness has to come from understanding something particular about the company and carrying that through the positioning, language, identity, content, website, motion, and interaction so that the whole thing feels like it belongs together.

Otherwise, better tools simply help you make generic work more efficiently (aka AI slop).

We use these tools too

Refokus uses AI, no-code tools, reusable systems, and whatever else helps us make better work and shorten times.

We have no interest in keeping difficult production practices just because they used to be difficult.

If a tool lets us explore more ideas, remove repetitive work, build something more ambitious, or spend more time on the decisions that matter, we want it.

What we are less interested in is treating the output of those tools as the standard.

The important question stays exactly the same one we asked before these tools existed: is the work actually good?

And when the answer is yes, there is usually another question worth asking.

Can it be even better?

As good gets easier, that question becomes so much more valuable. That question, is what creates the real moat.

‍