When You Sit Down With a Blank Page and Have No Clue What to Sketch
I've been drawing for about twelve years, mostly illustration and concept work, and the hardest part is never the technique. It's the first five minutes when you're staring at empty paper. I used to waste an entire session just rotating through possibilities in my head until I'd given up and checked my phone instead. Now I keep a running list, and it's saved me more times than I can count. The internet has a few reliable sources, and they all work differently depending on what kind of artist you are. If you're into character design, sites like ArtStation prompt generators and dedicated sketch challenge communities will give you structured daily prompts. For landscape or environment work, nature photography forums and location-based image boards tend to produce more useful material because they ground the subject in real lighting and geography. I found that simply searching for a random noun and adding an unexpected condition works better than most generated prompt lists. Something like "abandoned greenhouse at dawn" or "busy noodle shop in rain" forces your brain into a problem-solving mode instead of the paralysis of infinite choice. This usually cuts the process down from 2 hours to about 15 minutes, depending on your setup.
One source I recommend keeping close is PureRef or even a simple folder of reference images organized by mood rather than subject. When you're stuck, browsing by emotional tone instead of category tends to unlock things faster. I keep folders labeled "melancholy light," "chaotic energy," and "quiet tension" and rotate through them when a prompt list feels too rigid.
How to Actually Use a Prompt Without Becoming a Photocopier
This is where most people go wrong. They take a prompt, render it as accurately as possible, and call it done. That's not drawing from ideas, that's drawing from instructions. The skill is in the translation step. Take whatever prompt you land on and immediately change one fundamental element before you put pencil to paper. If the prompt says "lighthouse on a cliff," make it a lighthouse on a bridge. If it says "character in armor," strip the armor and give them ordinary clothes. This forces your brain to engage with the underlying structure instead of copying surface details. The exercise takes the same amount of time but produces work that actually feels like yours. I've seen artists spend hours perfecting a technically accurate rendering of a prompt and feel empty afterward because nothing in the piece reflected their own decisions. The workaround I use is simple: impose a constraint before starting. Limit yourself to three shapes, or one color temperature, or a specific viewpoint you'd never normally choose. Constraints create problems your brain has to solve, and solving problems is where the actual drawing happens.
Get the Full Details

The Honest Downsides Nobody Talks About
Prompt lists have a real bottleneck: they tend to produce work that looks like everyone else's when used without modification. If you're drawing the same popular prompt as thousands of other people on the same day, your piece will naturally share visual DNA with every other interpretation unless you deliberately diverge early. There's also the fatigue factor. Using generated prompts exclusively tends to drain creative stamina faster than working from personal observation. I've noticed that after about six weeks of relying solely on prompt generators, my sketches started feeling interchangeable and I had to step back to thumbnail studies and life drawing to reset. The recommendation I give myself is simple: use prompts as a starting point, not a destination, and rotate in personal observation work every few sessions to maintain creative independence. If you're completely stuck on Ideas For Stuff To Draw and none of these sources are helping, I'd recommend stepping back to something simpler. Keep a visual diary of one thing you observed each day for a week without any generated prompts at all. Sometimes the best ideas come from paying attention to what's already in front of you instead of searching for external inspiration.
I ran into a specific edge case last year where a prompt generator kept giving me variations of "cyberpunk city at night" and my work started looking identical across different pieces. The exact workaround I used was to impose a temporal constraint before starting. I limited myself to dawn lighting only for an entire month and produced work that felt distinctly different from every other cyberpunk interpretation because the constraint forced me to solve a real problem instead of copying surface aesthetics. This approach works well for developing a personal style, but it has real limitations when you need to meet commercial deadlines quickly. In those situations, I'd recommend keeping a reference folder of pre-organized images by mood rather than category and rotating through them to maintain creative independence.