I Thought Good Software Would Market Itself. I Was Wrong.

Mihai-Cristian Bâltac — Tue, 14 Apr 2026 10:42:23 +0000

For most of my career, I believed something a lot of technical people quietly believe:

If you build something genuinely good, people will notice.

Clean code. Good UX. Fast performance. Real value.

Surely that should be enough.

It was not.

The Belief That Broke

My background was mostly in frontend, UI, and later full stack work. For years, I thought like a builder first: make the product better, make the system stronger, and the results will follow.

That belief feels rational, especially when you spend most of your time around other technical people.

But the moment you try to build something outside your job, you run into a different reality:

good work does not automatically create attention.

Why I Started Questioning It

Then the market changed.

AI got better. Layoffs became more common. More teams started doing more with fewer people. More non-technical people started building tools and workflows with AI.

At the same time, I kept hearing the same kind of quiet anxiety from people in tech:

“Every quarter we have another meeting where they announce the team will shrink again.”

That pushed me into a question I had avoided for a long time:

If being a good developer is no longer enough, what else do I need to become?

The Wrong Assumption

When I started building things outside my job, I made a very common mistake.

I thought quality would create demand.

I thought if I made something useful, polished, and technically strong, people would come. Maybe through referrals, word of mouth, or “organic growth.”

It sounds reasonable.

It is also one of the biggest lies builders tell themselves.

A good product does not create traffic by itself.

A good website does not create attention by itself.

A good app does not create customers by itself.

What good software does is this:

once people arrive, it helps convert, retain, and impress them.

But first, people have to arrive.

That is where marketing starts.

What Frustrated Me About Marketing

When I started taking marketing seriously, it felt frustratingly vague.

As developers, we are used to practical steps:
install this, configure this, test this, fix this.

Marketing felt full of advice like:

post consistently
know your audience
provide value
build your brand

None of that is wrong.

But when you are starting, it can feel too abstract to act on.

So I approached marketing the same way I eventually learned programming:

learn the concepts
understand the mental models
test things in the real world
keep what works
stop confusing theory with progress

That changed the way I looked at it.

The Hardest Lesson

The hardest lesson was this:

building something good and getting attention for it are two different skills.

A lot of technical people overvalue creation and undervalue distribution. I did too.

We assume the hard part is making the product.

Sometimes it is.

But often the harder part is making the right people aware that it exists, understand why it matters, and trust it enough to try it.

I also had another wrong assumption:

I thought that if I wanted a problem solved badly enough, a lot of other people probably wanted the same thing.

Sometimes that is true.

Often it is not.

A lot of developer-built products are polished versions of our own frustrations. That is not automatically bad, but it is not the same thing as market demand.

To understand what other people actually need, you have to do something many developers avoid:

talk to people. A lot of them.

Where Developers Still Have an Advantage

The interesting part is that once I stopped expecting software to market itself, I started seeing a different advantage.

Developers are not automatically good at marketing.

But good developers do have one big advantage:

we are trained to think in systems.

And that matters more than I expected.

Because once you move past the surface-level idea of marketing, a lot of it becomes a question of structure:

what happens before attention
what happens after someone clicks
where people drop off
what gets measured
what creates friction

That was familiar territory.

I stopped seeing marketing as “making posts” and started seeing it as part communication, part systems design, part customer understanding.

That made it much easier to take seriously.

What I Believe Now

I no longer believe that good software markets itself.

I believe this instead:

good software matters once attention already exists
distribution is a real skill, not an afterthought
understanding business makes technical work more valuable
talking to customers beats guessing
building is important, but it is not the whole game

The market is getting noisier. More people can create. More tools reduce the gap. More companies buy outcomes, not technical elegance.

So the question is no longer just:

Can you build it?

It is also:

Can you understand who it is for, why it matters, and how it reaches the right people?

That is the shift I am trying to make now.

And honestly, I think more developers should make it too.

What changed your perspective more: building products, or trying to get people to care about them?

Has anyone managed to create a custom Gem in Gemini for image generation that actually works well?

Mihai-Cristian Bâltac — Tue, 24 Mar 2026 18:41:06 +0000

I’ve tried a lot of approaches: giving it JSON instructions, adding reference images, and describing the style in detail, but the results still don’t follow the intended look consistently.

What I’m trying to achieve is simple:
I want to upload one image and have Gemini generate a new image in the same visual style.

At the moment, it doesn’t seem to handle that reliably.

DEV Community: Mihai-Cristian Bâltac