I Built a Tiny Tool for the Hardest Part of Drawing: Starting

TomKeller — Fri, 18 Sep 2026 12:51:51 +0000

I have a slightly embarrassing habit: I can spend longer choosing what to draw than actually drawing it.

The problem usually is not a lack of interest. It is the moment just after opening a sketchbook, when every possible subject is available and none of them feels concrete enough to begin.

“Draw something interesting” sounds inspiring, but it is not a very useful instruction. It leaves too many decisions open: what kind of subject, from what angle, at what level of detail, and how much time should the exercise take?

That small problem is what led me to build Drawing Suggestions.

A blank page is a decision-making problem

When I am stuck, I do not really need an endless stream of ideas. I need one idea with enough shape to get moving.

A useful drawing prompt should give me a subject I can see in my mind, while leaving enough room for interpretation. “A bird” is too broad. “A small bird balancing on a bent garden fence after rain” gives me something to work with without telling me exactly what the final drawing must look like.

That balance became the main design question: how do you make a prompt specific enough to start, but open enough to still feel like the artist’s own work?

The first version was too much of a list

My first instinct was to make a large collection of drawing ideas. More ideas felt like more value.

It did not take long to notice the flaw. A long list can create a second kind of blank page. Instead of asking “what can I draw?”, you start asking “which of these hundreds of things should I choose?” The choice has simply moved one step earlier.

So I started treating each suggestion as a small creative exercise rather than a database entry. The goal is not to impress someone with an unusual phrase. The goal is to make the next ten minutes easier.

The current version lets you browse ideas by theme and difficulty, then use Text to Prompt when you want to turn a rough starting point into a more developed creative direction. The result should feel like a nudge from a thoughtful sketching partner, not an instruction manual.

The four-step routine I keep coming back to

When I use the tool myself, I try not to overthink the prompt. This simple routine works better for me:

Pick a broad mood or subject area.
Choose one concrete suggestion instead of opening ten more tabs.
Set a short limit, usually 15 or 20 minutes.
Change at least one detail so the result becomes unmistakably yours.

That last step matters. A prompt is a starting line, not a finish line. The best outcome is not copying the wording perfectly; it is noticing an angle, texture, gesture, or relationship that you would not have chosen on your own.

What I am still learning

Building a creative tool has made me more cautious about the word “inspiration.” Inspiration is often presented as a sudden feeling, but for many people it is also a practical design problem.

You need a first action that is small enough to take. You need constraints that reduce noise rather than reduce freedom. And you need permission to make an imperfect first mark.

That is why I care more about whether a suggestion helps someone begin than whether it sounds clever in isolation. A quiet, ordinary subject can be more useful than a spectacular one if it gets the pencil moving.

If you have a sketchbook nearby, try choosing one subject and drawing it before you decide whether it was a good idea. Sometimes the idea becomes interesting only after you spend a few minutes with it.

That is the kind of beginning I wanted Drawing Suggestions to make easier.

DEV Community: TomKeller

I Built a Tiny Tool for the Hardest Part of Drawing: Starting

A blank page is a decision-making problem

The first version was too much of a list

The four-step routine I keep coming back to

What I am still learning