Categories
Crazy Ideas

I’m Stuck: A Random Design Idea Generator (and What I’d Build Today)

In 2010 I built “I’m Stuck”, a tiny web app that served a random design suggestion when you hit a creative wall. Here is the original idea, how I’d expand it, and how AI tools and Oblique Strategies fit in 2026.

Last updated: October 17, 2026

Every designer hits a wall. In 2010 my answer was a tiny web app called “I’m Stuck”. You clicked a button and it gave you one random design suggestion: change the color palette, try a serif, crop tighter, remove something. Most suggestions were useless for the project in front of you. Every so often one knocked something loose.

The original demo lived on josefresco.com and is no longer online, so this post is the record of the idea and where I’d take it now.

The original idea

The concept was simple: a list of short, specific design prompts and a button that picked one at random. No accounts, no settings. Just a nudge.

It works for the same reason a walk works. When you’re stuck, you’re usually stuck on one path. A random constraint forces you onto a different one.

Ideas for expanding it

These were my plans in 2010, and most still make sense:

  • Ratings. Let people vote suggestions up or down so the useful ones surface.
  • User submissions. Let other designers add their own prompts.
  • Filters. Narrow suggestions to typography, color, layout, or copy.
  • Community. Show what people made after following a suggestion.
  • Mobile. In 2010 this meant a mobile site. Today it would simply be a responsive page you can add to your home screen.
  • Beyond design. Random pick-up lines, economic policy, sitcom plots. The format works for any creative block. Some of these would work better than others.

The 2026 update: AI and an older idea

Two things have changed how I think about this.

First, I later learned the idea was not new. Brian Eno and Peter Schmidt published Oblique Strategies in 1975, a deck of cards with prompts like “Use an old idea” for musicians and artists who were stuck. If you like the concept of “I’m Stuck”, that deck is the original and still the best version of it.

Second, AI chat tools can now generate endless prompts on demand. That is useful, but it misses part of the point. A short, fixed list of human-written prompts works because each one is blunt and a little strange. An AI asked for “design ideas” tends to give you the safe, average answer, which is the opposite of what you need when you’re stuck.

If I rebuilt it today, I’d keep a curated list of prompts and use AI only for an optional second step: “Here’s my project and the prompt I got. Show me three ways to apply it.”

Try it without an app

You don’t need software. Write 30 short prompts on index cards, shuffle them, and draw one when you’re stuck. A few to start with:

  • Remove one element and see if anyone misses it.
  • Make the most important thing twice as big.
  • Use only one typeface.
  • Design it for a phone first.
  • Swap your main color for its opposite.
  • Write the headline before you touch the layout.

FAQ

Is the I’m Stuck web app still available?

No. The original demo on josefresco.com is no longer online. The prompts in this post and a deck like Oblique Strategies cover the same idea.

What are Oblique Strategies?

A deck of cards created by Brian Eno and Peter Schmidt in 1975. Each card has a short prompt meant to help artists and musicians break out of a creative block.

Can AI help with designer’s block?

Yes, as a second step. Pick a blunt, specific prompt first, then ask an AI tool for ways to apply it to your project. Asking AI for open-ended ideas tends to produce safe, average results.

How do I get unstuck on a design project?

Add a constraint. Limit yourself to one typeface, design for a phone first, or remove one element. Constraints push you off the path you’re stuck on.

Related posts

Tell me why I'm wrong...