Framework ⚡ Low

Crazy 8s

Eight ideas in eight minutes — the rapid sketching exercise that breaks the tyranny of the first idea.

Ideation Design Sprints Rapid Sketching Creative Thinking Workshop Facilitation Divergent Thinking

01 — TL;DR

Two sentences.

Crazy 8s is a structured rapid ideation exercise — developed by Jake Knapp at Google Ventures as a core component of the Google Design Sprint methodology and documented in "Sprint" (2016) — in which each participant folds a sheet of A4 paper into eight equal panels, sets a timer for eight minutes, and sketches one distinct solution idea per panel at a rate of one minute per sketch, using the combination of extreme time pressure and forced quantity to override the creative fixation that causes teams to anchor to the first plausible solution and stop exploring alternatives. For product design teams, Crazy 8s solves the specific problem that arises when ideation sessions devolve into discussion — when the loudest voice, the most senior person, or the first idea proposed shapes the direction before the full range of possibilities has been explored — by making parallel individual generation mandatory before group discussion begins.

The exercise's power comes from the deliberate mismatch between the time available (one minute) and the cognitive effort required to produce a fully considered design solution (much longer) — this mismatch forces participants to sketch instinctively rather than analytically, which paradoxically produces more creative output because it bypasses the self-censorship and critical evaluation that prevents people from committing ideas to paper. The one-minute constraint also calibrates participant expectations: no one expects a one-minute sketch to be polished, which removes the social anxiety of sharing imperfect work and creates psychological safety for proposing ideas that feel risky or unconventional. After eight minutes, a team of five people has produced 40 sketched ideas — a creative inventory that almost always contains solutions that would not have emerged from an open discussion.

02 — When to Use

Apply this when…

→ A design sprint is running and the team has selected a HMW question — Crazy 8s is the standard ideation exercise that follows HMW selection in the Google Design Sprint methodology

→ An ideation session is being planned and the facilitator wants to ensure that every participant generates ideas independently before group discussion begins — Crazy 8s prevents early anchoring to a dominant idea

→ A team meeting is stuck on a specific design decision and a 10-minute exercise that generates multiple alternative directions quickly would be useful

→ A team is experiencing groupthink — converging on solutions that feel safe or expected — and needs a structured intervention that forces creative risk-taking through time pressure

→ A cross-functional group includes non-designers who are uncomfortable with open ideation but comfortable with a structured, time-boxed exercise with clear rules

→ A product team needs to explore a design problem quickly — Crazy 8s can be run in under 15 minutes including briefing and sharing

When NOT to apply it

Skip it when the design problem requires careful specification rather than rapid generation — writing complex form logic, specifying accessibility requirements, documenting technical constraints. Skip it when the team has genuinely already explored the solution space thoroughly and further divergence would be wasteful. Skip it when physical materials are not available and digital sketching tools would slow the exercise below the speed required — the exercise depends on the physical speed of pen on paper. Skip it when the session has fewer than 2 participants — the value of Crazy 8s comes partly from the diversity of output across multiple people.

03 — How It Works

The mechanism

Crazy 8s works through the interaction of three deliberate constraints. The time constraint (one minute per idea) forces instinctive generation rather than analytical evaluation. The quantity constraint (eight ideas required) forces participants past their first and most obvious ideas toward genuinely novel territory — ideas 5, 6, 7, and 8 are almost always more creative than ideas 1, 2, and 3 because the obvious options have been exhausted. The individual constraint (everyone generates independently and simultaneously) prevents the social dynamics of group discussion from anchoring the team to any single idea before the full creative space has been explored.

The underlying practice — Google Design Sprint and the five-day methodology

Jake Knapp developed Crazy 8s at Google Ventures as part of the Google Design Sprint — a structured five-day product design and testing methodology that compresses months of iterative design work into a single week. In the sprint's structure, Crazy 8s follows Lightning Demos (examples of solutions from other domains) and the HMW question selection, and precedes the Solution Sketch (a more detailed single-screen design). The exercise's placement in the sprint is deliberate: Lightning Demos load participants with diverse input material, HMW defines the creative challenge, Crazy 8s forces rapid parallel generation, and the Solution Sketch allows participants to develop their single strongest idea in more detail.

What it means in practice — running the exercise

The exercise requires minimal materials: one sheet of A4 or letter paper per participant, folded into eight equal panels (fold in half three times), and a pen. The facilitator sets a timer for one minute and calls "go" — participants sketch one idea per panel, responding to the HMW question or design challenge. When the timer sounds, participants move to the next panel regardless of whether their sketch feels finished. After eight minutes and eight panels, participants share their sketches by posting them on a wall and briefly explaining each idea — typically 30 seconds per idea. The sharing and discussion phase is as important as the generation phase: ideas from other participants frequently spark new directions or combinations that become the strongest solutions.

The counter-intuitive nuance — the worst ideas are often the most valuable

The most practically important and most counterintuitive Crazy 8s insight is that the exercise's value is not in identifying the best idea from the eight but in using the full inventory of ideas — including the weak ones — to identify unexpected directions and combinations. Ideas 6, 7, and 8 are where the creative territory gets genuinely interesting, because participants have exhausted their obvious options by then. Teams that evaluate Crazy 8s output by trying to identify the "winner" miss the real value: a weak idea in panel 6 might contain one element that, combined with a direction from panel 2, produces the strongest solution. The post-Crazy-8s discussion should look for patterns, combinations, and surprising elements across all 40+ sketches rather than trying to crown a single winner.

How to measure it — solution diversity and ideation velocity

Crazy 8s produces two measurable outcomes. Solution diversity measures whether the 40+ sketches generated by a team of five span meaningfully different solution types — different mechanisms, different interaction patterns, different conceptual approaches — rather than clustering around variations of the same idea. High solution diversity indicates that the exercise has successfully broken creative fixation. Ideation velocity measures how quickly the team moves from HMW question to a set of solution candidates worth developing further — a well-run Crazy 8s session should produce this inventory in under 20 minutes including briefing, exercise, and sharing.

Crazy 8s vs brainstorming — different mechanisms

Crazy 8s and brainstorming are both ideation techniques but they work through fundamentally different mechanisms. Brainstorming is group-simultaneous and verbal — ideas are proposed and discussed sequentially, which means each idea is influenced by the previous one and the loudest voice shapes the session. Crazy 8s is individual-parallel and visual — everyone generates simultaneously without seeing each other's work, which means each participant's output reflects their own creative thinking rather than a response to the group's direction. Research on group brainstorming consistently shows that nominal groups (individuals working separately whose output is later combined) produce more diverse and higher-quality ideas than interacting groups of the same size. Crazy 8s is structured to capture the benefits of individual generation while still enabling the cross-pollination that comes from sharing and discussing output.

04 — Real Example

Slack's notification redesign sprint and the panel 7 idea that shipped

Slack's product team used a Google Design Sprint format — including Crazy 8s — to address a recurring user complaint: notifications were overwhelming and difficult to manage, but the team's initial design direction (a more granular notification settings panel) was not meaningfully different from the existing interface. The Crazy 8s session was run with a HMW question: "How might we help users feel in control of their attention without having to think about notification settings?"

In the Crazy 8s session, the first four panels of most participants' sheets contained variations on the settings panel concept — more toggles, better organisation, smarter defaults. By panel 6 and 7, participants were sketching more radical concepts: a "focus mode" that temporarily silenced all notifications with a single action, a "do not disturb schedule" that learned from calendar events, and — in one participant's panel 7 — a simple status indicator that let users communicate their availability to their team as a way of managing incoming notification behaviour at the source rather than at the receiver. The status indicator concept, which became Slack's "Active/Away/Do Not Disturb" status system, emerged specifically because the Crazy 8s constraint forced the participant past their initial obvious solutions into genuinely novel territory.

Slack · notification redesign sprint

Time pressure forces past obvious settings redesign to the status concept

Panel 7 sketch → shipped Slack status system

05 — In the Wild

Test yourself & see real examples

No examples yet — be the first.

Spotted a product feature that feels like it came from a genuinely creative ideation process — where the solution is surprising and non-obvious in a way that suggests the team explored far beyond the first idea — or one that feels like the first idea proposed and never challenged? Submit what you observed.

✓ Reviewed before publishing ✓ Your name on every example you submit ✓ Violation or fix — both welcome

Seen a feature that clearly came from deep ideation — or one that is obviously the first idea that was never questioned? Help grow the evidence base.

06 — Common Mistakes

Where teams go wrong

✗ Sketching one idea eight ways instead of eight different ideas. The most common Crazy 8s mistake is treating the eight panels as eight iterations of the same concept rather than eight genuinely different directions. A participant who sketches variations of a navigation redesign in all eight panels has not completed Crazy 8s — they have created eight versions of one idea. The exercise requires eight distinct conceptual approaches: different mechanisms, different entry points, different interaction models, different metaphors. Brief this explicitly: "Each panel must represent a meaningfully different direction. If you cannot describe each sketch in a different sentence, you are not doing it right."

✗ Spending more than one minute on any single panel. The time constraint is not a suggestion — it is the primary mechanism through which the exercise produces creative output. Participants who spend two minutes on panels they find interesting and thirty seconds on panels they find difficult are defeating the purpose of the exercise. The one-minute constraint forces commitment: you must move to the next panel when the timer sounds, even if the sketch feels unfinished. An unfinished sketch that represents a genuinely new direction is more valuable than a polished sketch of an obvious idea. Use an audible timer and call the transition between panels firmly.

✗ Evaluating output during the generation phase. A common failure mode — especially in teams with strong senior voices — is participants commenting on each other's work during sharing in evaluative rather than generative terms: "that won't work because of X," "we already tried that," "that's not technically feasible." The sharing phase should be strictly observational: each participant explains their sketches without interruption, and the group looks for patterns, combinations, and surprising elements across the full set. Evaluation and selection happen in a separate subsequent phase. When evaluation enters the sharing phase, participants censor their remaining ideas and the exercise loses its psychological-safety value.

✗ Running Crazy 8s without a HMW question. Crazy 8s run without a specific creative challenge — "just sketch ideas for the product" — produces unfocused output that is difficult to evaluate and compare. The exercise depends on a shared creative challenge that every participant is responding to: without it, participants generate from different starting points and the inventory of 40+ sketches has no common frame for comparison. The HMW question does not need to be from a formal HMW session — any well-formed design challenge will work — but it must be specific enough that participants' sketches are recognisably responses to the same question.

07 — Variations & Related Principles

Connected ideas

Crazy 8s is the rapid ideation engine in the design sprint toolkit. Its closest relationships are with the frameworks that surround it in the design sprint sequence.

The most important pairing is Crazy 8s with structured sharing and pattern identification. The 40+ sketches generated by a five-person session are not valuable as individual ideas — they are valuable as a collective inventory that reveals patterns, combinations, and unexpected directions. Teams that run Crazy 8s well but rush the sharing phase lose most of the exercise's value. Allocate at least as much time to sharing as to generation, with a facilitator actively looking for cross-pollination opportunities between different participants' sketches.

08 — 10-Min Exercise

Run it right now

⏱ 10 minutes · Solo or with your team · Needs paper and a pen

The Solo Crazy 8s

1. Take one sheet of A4 or letter paper and fold it in half three times to create eight equal panels. Number each panel 1-8 lightly in the corner.

2. Pick a design problem you are currently working on — or use this prompt: "How might we make it easier for a new user to understand the value of our product in their first session?"

3. Set a timer for eight minutes. Sketch one idea per panel, one minute per panel. The sketch does not need to be polished — a rough wireframe, a flow diagram, or even a labelled box layout is fine. The only rule: each panel must represent a genuinely different approach from the others.

4. When the timer sounds, stop. Look at all eight panels. Which panel surprised you most? Which idea would you not have considered without the time pressure forcing you past your obvious options? This is usually panel 5, 6, or 7 — the ideas that only appeared because the earlier panels consumed your obvious responses.

5. If you are running this with a team: post all sketches on a wall after the exercise, take three minutes for each person to explain their eight panels without interruption, then spend five minutes as a group identifying the ideas or elements that appear across multiple participants' sketches — these convergence points are often the strongest starting points for further development.

10 minutes