Delay the Primal Mark
Stay in low-fidelity blocks and conversation as long as possible before drawing anything that looks real.
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 4
- Confidence
- 94%
The first realistic-looking picture (the 'primal mark') anchors everyone and collapses the solution space to a first-order idea. Baxley's method is to deliberately resist high-fidelity mockups and AI prototypes early, working in words and chunky block diagrams so the team can reach second, third, and fourth ideas before committing. The hard work is thinking, not rendering.
Origin
Baxley's concept, borrowing 'the primal mark' from fine-art practice (the first mark on a canvas that everything after responds to).
Core principles
- 01The first mark on the canvas sets a baseline everything else reacts to
- 02People are so uncomfortable with ambiguity they treat the first realistic sketch as 'the thing'
- 03High-resolution output invites feedback on color and shape, not usability and value
- 04The heavy lifting is conceptual thinking; high-res comps are fast once thinking is done
- 05Getting to the 2nd-4th idea can happen inside a single meeting if you refuse to jump
How to run it
- 1
Work the idea in words and conversation first
Explore the concept verbally and conceptually, resisting the urge to render anything that looks real.
Pro tip When you get a promising idea, say 'that's interesting, let's table it, what else have we got?' to force more options.
Watch out As soon as a realistic picture appears, everyone gravitates to it and keeps doubling down, even if it's the shallow first-order answer.
- 2
Use block-frame diagrams, not wireframes or comps
Represent screens as big chunky blocks showing rough layout, deliberately too low-fidelity for anyone to comment on look and feel.
Pro tip Low fidelity forces the conversation to stay conceptual because there's nothing visual to critique.
Watch out Expect the product team to get nervous sitting in block frames for weeks asking 'when do we see the comps?'
- 3
Only render high-fidelity once the concept is locked
After the block frames and concept are agreed, produce the visual comps; with a robust design system this is fast, even a single day.
Pro tip Baxley's team could hand locked block frames to an agency and get full high-res comps back in a day.
- 4
Reserve interactive prototypes for expression, not exploration
Use AI/prototyping tools as production tools once you know what you want and can write a robust prompt, since an interactive idea needs to be expressed interactively.
Watch out Pushing a fragile early idea into a high-res prototype exposes it to premature critique and squashes it.
In the wild
The team sat in block-frame diagrams and wireframes for weeks, making the product team anxious about seeing comps. Because the concept was fully worked out and the design system robust, the final high-res comps came back from an agency almost overnight.
→ PMs were shocked at the speed; Baxley's point was the high-res stuff was never the hard part, the thinking was.
A PM insisted a homepage link 'has to be blue' and went around Baxley to engineering to make it so. The real problem was the link wasn't prominent enough, and blue was just the first, most obvious solution from a non-designer.
→ Illustrates jumping to a premature concrete solution instead of stating the problem and letting design explore the many ways to solve it.
Common mistakes
Going straight to prototypes / 'demos before memos'
Cranking a UI from a first-order prompt narrows possibilities to what existing-solution-trained tools already know, skipping the more interesting later ideas.
Handing designers a fully-drawn sketch
A PM's finished drawing tells the designer exactly what NOT to build; it removes the space to reach a better solution.
Is it for you?
Best for
Product and design teams tackling genuinely new problems who feel pressure to prototype immediately.
Not ideal for
Well-understood, production-ready ideas where the concept is settled and the goal is fast execution.
From the transcript
“you should wait as long as possible to draw a picture”
“I have this idea from art that I call the primal mark”
“block frame diagrams, which were even simplified versions of wireframes”
“an interactive idea needs to be expressed interactively”
From the episode
35 years of product design wisdom from Apple, Disney, Pinterest, and beyond
Bob Baxley