I wanted to make a poster generator for people who enjoy good typography but do not necessarily want to negotiate with a blank canvas.
Something playful. Something structured. Something that could help a curious person make a genuinely interesting poster without first completing a minor degree in layout systems.
So I built one.
It began with a grid, obviously.
The first version of Poster, Please started with a familiar designer's tool: the grid. I wanted enough structure to keep the composition intentional, then enough freedom for people to interfere with it.
They could change the type, shift the layout, distort the grid, move geometric elements around and choose a colour palette. The system would hold everything together while the user made increasingly suspicious decisions. (If you're reading this and would like to tinker with the microsite, please do share your posters with me and let me know how I can improve it. I would be eternally grateful.)
That was the idea, anyway.
Then my first user arrived.
I sent the tool to my childhood friend and watched him use it. He changed the words, moved things around and made a poster.
Then he asked:
There's only one design?
Rude. Accurate. Extremely useful.
He had found the difference between a working feature and a satisfying product in approximately one minute.
I had been thinking about how customisable the composition was. He was thinking about variety. To me, the grid could produce many outcomes. To him, it was still visibly the same starting point wearing different trousers.
One composition is a demo.
That comment changed the direction immediately. Poster, Please needed more than controls. It needed distinct systems with different personalities.
The next version introduced another poster composition and made the interface more flexible. Controls became collapsible. Geometric elements became movable. The poster stopped feeling like one template with several knobs and started becoming a small family of visual possibilities.
It is still growing. That is part of the point.
Constraints are not the enemy.
I do not want Poster, Please to become professional design software in a tiny browser window. We already have professional design software. It has seventeen panels and knows what it did.
The interesting part is deciding which choices belong to the user and which choices the system should quietly make for them.
Too little control and the result feels generic. Too much control and I have simply rebuilt the blank canvas I was trying to avoid.
The sweet spot is guided mischief: enough structure to create confidence, enough freedom to produce surprise.
Building changed how I was designing.
I came to this as a designer, not a developer. I was used to imagining how an interface should behave. Building the thing made me confront how it actually behaved when someone else touched it.
Buttons needed a reason to exist. Controls needed to feel understandable. Movable objects needed to move where people expected them to. Every interaction had to survive contact with a human who had not been sitting inside my brain during the design process.
That is what I enjoy about making small digital tools. The idea cannot hide inside a polished mockup. It has to work.
The first user is not the final verdict.
But they are very good at revealing the question you forgot to ask.
Poster, Please is still an experiment. It is also already more useful than the version I first shared, because one person looked at it without any attachment to my original plan and said the obvious thing.
Good feedback does not always arrive as a detailed critique.
Sometimes it arrives as six words and ruins your afternoon beautifully.