LLenny's Podcast
← All frameworks
InnovationJenny Wen (head of design at Claude)

Legibility Framework for Spotting Frontier Ideas

Hunt for illegible ideas — the ones with real energy that nobody can quite articulate yet — and translate them.

Difficulty
Advanced
Time to result
~months to results
Steps
3
Confidence
85%

A 2x2 that crosses founders (legible vs illegible) with ideas (legible vs illegible). When both are legible the idea usually isn't novel and someone will already build it; the opportunity lives where the idea is illegible — on the frontier, not yet told in a way that makes sense. Jenny Wen adapts this so designers act like internal VCs: scan for illegible internal prototypes that generate real energy but aren't understood yet, then translate them into storytelling and form factor.

Origin

The legibility 2x2 is attributed to Evan Tana, a partner/VC at SPC, which Jenny saw on Twitter; she extends its VC logic to the internal role of a designer at a frontier lab.

Core principles

  • 01If both founder and idea are legible, the idea probably isn't novel — someone will already do it
  • 02The real opportunity is where the idea is illegible: on the frontier, not yet clearly articulable
  • 03People drawn to early ideas often can't explain why the energy is there
  • 04A designer's job at a frontier lab includes spotting illegible ideas and making them legible
  • 05Translate illegible energy via storytelling, UX, and form factor

How to run it

  1. 1

    Scan for energy you don't understand

    Watch internal prototypes, Slack, and research work for ideas that attract real energy from smart people but that you personally can't yet make sense of.

    Pro tip Treat 'I don't get it but there's clearly something here' as a buy signal, not a reason to dismiss it.

    Watch out Not every crazy-sounding idea is good; energy plus your own curiosity is the filter, not craziness alone.

  2. 2

    Dive deeper to make it legible

    Because early believers often can't articulate the value, it falls to you to investigate, understand the underlying mechanism, and figure out what's actually there.

    Pro tip Pull the specific elements people found genuinely helpful out of the rough prototype rather than judging the whole thing.

  3. 3

    Translate into form factor and story

    Transform the illegible idea into something people can grasp — through storytelling, UX, and a concrete form factor that surfaces its value.

    Pro tip Reuse the exact sub-interactions that already resonated internally instead of inventing from scratch.

In the wild

Claude Studio prototype into Co-work

A dense internal prototype called Claude Studio showed knowledge, skills, plans, to-dos and outputs; Jenny didn't understand it at first but sensed the energy, dove in, and later pulled elements like Claude's plan, to-dos, and context files into Claude Co-work.

The illegible prototype seeded both the skills/markdown-instructions framework and concrete Co-work interface elements.

Early joiners at breakout companies

Jenny references a study with VC Terrence Rohan finding that people who joined companies like Palantir, Stripe, Linear and Notion very early were drawn to ideas so crazy people laughed at them, plus visible excitement and top-1% founders.

Reinforces that illegible, laughed-at ideas with real energy are where outsized opportunities hide.

Common mistakes

Dismissing an idea because you don't get it

Illegible frontier ideas look confusing by definition; rejecting them for lack of immediate clarity means missing exactly the novel opportunities the framework points to.

Chasing craziness without energy

Not every idea that makes no sense is good; without real energy from credible people plus your own genuine curiosity, 'illegible' is just noise.

Is it for you?

Best for

Designers, PMs, and operators at frontier or research-driven orgs who need to spot and productize emerging internal ideas

Not ideal for

Execution-focused teams optimizing well-understood, legible products where novelty-hunting is a distraction

From the transcript

it was Evan Tana who is a partner at SPC

1:02:00

founders can be either illeible or legible

1:02:00

part of the role of the designer at least um at least at at a at a frontier lab

1:03:00

how can designers almost be more like VCs in this way internally

1:05:30

From the episode

The design process is dead. Here’s what’s replacing it.

Jenny Wen (head of design at Claude)