It felt like progress. Was it?

This week you bought the domain. You sketched the data model twice, and the second version was cleaner than the first. On Thursday you explained the idea to a friend, and they nodded in the right places. On your list, the same line has been sitting there for three weeks: login, payments, pricing page.

None of that was wasted. But it is worth asking what it was.

The feeling and the evidence

Planning, picking a name, choosing a stack, describing the idea out loud: all of it produces the same sensation as work, sometimes a stronger one. The difference only shows when you ask who else could see it. A domain nobody visits, a diagram in a notebook, a conversation that was over when the drinks were. Something changed in you, not in the project.

Someone in a discussion about finishing side projects described a rule they kept: the project went unmentioned to anyone until there was something to put in front of them. Talking about the plan was off the table. Running what already worked, right there in front of people, was allowed.

Of everything you did on the project this week, what could another person see today without you explaining it?

Time, or something else? Be honest about which

Two evenings can look the same from the outside. On one, the hour really wasn't there: the job ran long, someone needed you, the day was simply full. On the other, the hour was there, and it went somewhere else.

Both happen, and neither is a character flaw. They just need different answers. If the hour is missing, no method will create it, and the honest move is to plan with the hours you actually have. If the hour was there and went elsewhere, more time won't help either, because the next free evening will go the same way.

Did you have an hour this evening? And if you did: would you have spent it on the project?
Ask it on three evenings, not one.

The rest is a list, not a mood

When a project stalls near the end, what's left feels like one large grey block. Write it out and it turns into a list of unglamorous items: sign-up and login, password reset, the payment flow, a pricing page, what happens when something fails, the actual deploy. None of it is interesting. That is exactly why it waits.

Writing the list down doesn't make it shorter. It turns a feeling into items you can count and cross off, and it tells you whether you're looking at four evenings or forty.

"Done" keeps moving

Ask yourself what done means for this project, and notice whether the answer arrives in quotation marks. A "done" in quotation marks shifts every time you get close to it: one more feature, one more cleanup, one more redesign. That isn't dishonesty. A finish line that only exists in your head is allowed to move, so it does.

One developer compared this with games. A game spells out what winning looks like, so at any point you know how far off it is. Their own goals, by contrast, used to sound like a loose intention to build something along those lines, and an intention like that never says at what point the thing counts as built. What helped was a different kind of goal: one that names the moment the result leaves your hands and is out there.

A goal like that fits into one sentence that someone else could answer with yes or no, on a date you pick yourself. For example:

On the date I wrote down, a person who has never seen the project can create an account and reach the first real screen without asking me anything.

If your sentence can't be answered with yes or no, it is still a mood.

What helped, and what didn't

Among the things people describe as having helped when a side project did get finished: a smaller scope than they first wanted, fixed slots in the week instead of "whenever", someone who was genuinely interested in the result, and an end that was settled before they began.

The other side is just as real. For some, a date turns their own project into a second job, and they stop rather than start. One developer put themselves firmly on that side: they tie their effort to the project being complete, not to a deadline. Guessing durations isn't their strength, and a clock running in the background makes them anxious, so the goal is what they steer by. They also admitted that they don't always manage to live by this themselves.

Both sides describe something true about themselves. A date that carries one person freezes another. Which one you are is better found out on a small goal than on a big one. And if a date freezes you, write the sentence from above without one. The yes or no is the part that does the work.

Live is not the end either

Getting something live is a real line. It isn't the last one. A project that is out can go just as quiet as one that never left your laptop: nobody signs up, the next feature feels pointless, the evenings drift back to other things.

It helps to know that in advance. Not as a reason to stay put, but so the quiet afterwards doesn't land like a verdict. Nobody can honestly promise you a fix for that part.

Maybe you're already done

There is one more honest possibility.

Some projects exist to learn something: a framework, a language, how payments work. Others exist to solve a problem for someone. A learning project can be finished long before its first user, at the moment you've learned what you came for.

In the same kind of discussion where others swap tips for getting over the line, someone simply turned the question around: perhaps not finishing is fine. If the part that drew you in is behind you, the project may already have given you what it had to give. What remains is whether completing it would mean anything to you in its own right.

That question deserves a real answer, not a reflexive one. If the answer is no, closing the project isn't a failure, it's a decision. If the answer is yes, the list from above and the sentence with a yes or no are where to start.

Originally published on my own site, where the discussions referenced above are linked: mvpbuilder.io/blog/it-felt-like-progress. I build a product for developers in this situation, so I'm not neutral; this post doesn't pitch it. This article was written with the help of AI and edited and checked by me.