PaperPict
Home / Blog / Article

SEPTEMBER 8, 2026 · 3 MIN READ

One year of PaperPict: what we learned about figures

Eleven months, six tools, and a lot of figures. Some of what we learned we expected. Some of it we didn't.

Editing beat generation

We set out to build a generator and found out we had built an editor. Almost nobody exports the first draft untouched; people retype a label or two, nudge an arrow, sometimes regenerate one panel. The first draft matters, but the fact that it can be edited is the reason people keep using it.

Reference-to-figure surprised us

We expected text prompts to dominate. Instead, uploading the paper quickly became the usual way to start a graphical abstract, and a very common way to start a mechanism diagram. Given the choice, researchers would rather point at the document than describe it.

Palettes are about consistency, not beauty

People switch whole palettes far more often than they pick individual colours, and the reason is rarely taste. It's usually to make a second figure match the first, or to move a paper to a different journal after a rejection.

Co-authors drove PPTX

Editable PPTX was a side feature at launch. It turned out to be one of the most requested exports, almost entirely because of co-authors and talks.

What we got wrong

Our first poster templates were too dense, and our early defaults encoded categories by colour alone. Both are fixed: type minimums are enforced, and Okabe-Ito is the default when colour carries meaning.

Year two

Custom lab palettes are shipping to all Lab plans, the phylogenetic tool gains circular layouts, and we are working on figure checks against specific journals' author guidelines before export. As always, we will write about it here — and about figures in general, because the craft matters more than the tool.

Make your next figure in about 30 seconds.

Free to start. No credit card. Every element stays editable.