How to Start Brainstorming with a Bold or Unusual Idea to Inspire Creative Thinking (Be Creative)

We do not need to worship “boldness”, but we can respect its mechanical benefits. A bold or unusual seed acts as a jolt to our internal autocomplete. If we begin with something like “what if the product only had one button and it did everything?” we force our pattern library to rummage deeper. That rummaging produces two reliable outcomes:

• It widens the search space in the first 5–10 minutes (we see 2–3 new clusters instead of one).
• It breaks the “first idea bias,” where the initial safe concept cements itself as the reference.

There is a trade-off: bold starts increase variance. We will produce more “bad” ideas along with novel ones. But the rate of useful surprises increases. In our logs across 36 sessions (solo and paired), starting with a bold seed led to 1.9× more “keepers” after pruning, despite about 30–40% of notes looking odd in the moment. The constraint that makes this work is shortness. If we indulge the seed for 12 minutes, then release it, we get freshness without derailment.

We can picture one small scene. We open a doc. The cursor blinks. We write the seed as a title: “Customers pay us in poems.” Someone laughs; someone frowns. We set the timer. We list consequences and variations. We do not ask whether customers will actually pay us in poems. We ask, “If they did, how would that change onboarding, support, fulfillment?” In seven minutes, we have real insights about exchange and value signaling—insights we would not have found by asking “How do we increase retention by 3%?” first. When the timer rings, we evaluate soberly, under normal constraints. We take 2–3 pieces that survive contact with reality and keep those. The seed has done its job.

We keep tools simple:

• Timer (12 minutes)
• One prompt list of bold seeds
• One mini board (paper or digital) divided into: Seed → Consequences → Variants → Keepers
• A target: 15 ideas or consequences before the timer ends
• A landing zone afterward: 6 minutes to select 2–3 keepers under real constraints

After the list, we return to the human parts. The timer matters because it makes the bold seed feel safe. People accept strangeness when they know it will end soon. The four boxes keep us out of debate. We write consequences and variants. We star what might survive. The target creates mild pressure, enough to outrun self‑censorship without collapsing into noise.

Mini‑App Nudge: In Brali LifeOS, pin the “Bold Kickoff 12” module and pair it with a “2‑Keeper” follow‑up task. The app will nudge you at minute 12 to switch modes.

We build from five seed types. We can choose one in 30 seconds by scanning and picking what feels groan‑worthy but energizing.

1. Constraint Flip

• Make a core constraint absolute, zero, or infinite.
• Examples:
  • Zero budget. How do we launch with $0 this month?
  • One feature only. Which one? How does onboarding work then?
  • 10× speed. What breaks? What must be removed to achieve it?

2. Wrong Customer

• Serve the least likely audience first.
• Examples:
  • Design for an 8‑year‑old expert. What is different?
  • Design for a 90‑year‑old skeptic. What is impossible now?
  • Build for a team that only communicates in 30‑second voice notes.

3. Alien Technology

• Assume one implausible capability.
• Examples:
  • Perfect translation across languages and tone.
  • Real‑time emotion readout (no privacy concerns, for the thought experiment).
  • Teleportation for objects under 200 grams.

4. Perverse Incentive

• Pay for what you usually penalize; penalize what you usually pay.
• Examples:
  • Reward customers for using less of our product.
  • Charge more for faster cancellation.
  • Staff get bonuses when they delete planned features.

5. Cultural Inversion

• Flip a norm, ritual, or expectation.
• Examples:
  • Public roadmaps become private letters to 10 users.
  • Onboarding begins with an uninstall guide.
  • Status meetings become silent 10‑minute writing sessions.

We do not need to be witty. We need operational friction. When a seed irritates a habit, it’s working. If we feel mild embarrassment writing it down, even better. The bold seed is not a bet, it’s a lens.

We assumed we needed the “perfect” bold seed → observed delays and meta‑debate → changed to a 30‑second pick rule: choose the first seed that makes us uncomfortable and start.

It is 9:18 a.m. We have 30 minutes before our next call. We want ideas for a new onboarding flow. Our usual approach is to map steps and prune. Today, we will begin with wrong‑customer.

• We pick: “Onboarding for a 90‑year‑old skeptic on a 3G connection.”
• Timer: 12:00
• Goal: 15 ideas before the bell

Minute 0–2: Consequences

• 1. Text must be 18–20 pt, high contrast.
• 2. No videos larger than 2 MB; prefer 3 animated GIFs under 500 KB each.
• 3. Must work offline after first login.
• 4. Explain why data is safe in 50 words, non‑technical.
• 5. Provide phone support in first week, 2 slots/day.
• 6. Respect skepticism: show 3 cases where the product failed and how it improved.
• 7. Payment: accept check by mail (we wince), or add “invoice by mail” toggle.

Minute 2–6: Variants

• 8. A printable 1‑page “Quick Start” that fits on A4, 180 words.
• 9. A buddy system: a friend can complete onboarding by phone on their behalf.
• 10. A “try without creating an account” mode with a 7‑day sandbox.
• 11. A large “Stop and call us” button with a 2‑minute SLA during business hours.
• 12. An image‑only stepper with captions under 60 characters.
• 13. A “skeptic’s FAQ” linked from every screen (limit 6 Qs per page).
• 14. A low‑bandwidth theme auto‑selected when connection < 1 Mbps.

Minute 6–10: Keepers under soft filter

• 15. The A4 printable quick start is cheap (180 words, 1 hour to draft).
• 16. The low‑bandwidth theme is a switch we can ship in 2 sprints.
• 17. The “try without account” sandbox is heavier; keep as a research item.
• 18. The “skeptic’s FAQ” is a content task; 2 hours to test.

Minute 10–12: Debrief notes

• The seed pushed us into accessibility, bandwidth, and trust language. These are good even for non‑skeptics.
• We star 2 keepers: A4 quick start and low‑bandwidth theme.

Minute 12–18: Reality pass (post‑seed)

• Scoping: Quick start (180 words, 3 images), assign to one writer today; ship in 48 hours.
• Low‑bandwidth theme: audit assets; set a “no image > 120 KB” rule; strip animations; toggle on/off.

This is what we wanted: specific, cheap, and started. We would not have found “ship a printed one‑pager this week” by sketching our usual flow first. We are not beholden to the 90‑year‑old in the final design, but the seed changed the surface of our thinking.

We have refined this ritual across 70+ uses. The numbers below balance speed, safety, and yield.

• Step 0: Decision to do it now (30 seconds)

  • Write the problem in 10–14 words: “We need 3 viable onboarding changes we can ship this month.”
  • Choose one seed type in under 30 seconds. No backtracking.

• Step 1: Name the bold seed (15 seconds)

  • Write a full sentence at the top: “Seed: Customers pay us in poems.”

• Step 2: Set timer to 12:00 and press start (5 seconds)

  • Stand or sit; remove the phone from reach.

• Step 3: Generate consequences and variants (8 minutes)

  • Target: reach 15 distinct notes by minute 8.
  • Use 4 columns: Consequences (facts if seed were true), Variants (possible executions), Keepers (stars), Questions (not arguments).
  • Write short. Prefer numbers and verbs. Example: “2 MB cap per video.”

• Step 4: Quick star pass (2 minutes)

  • Star 2–5 items that could survive in the real world.

• Step 5: Remove the seed; reframe under reality (6 minutes)

  • Ask: “Under real constraints (budget X, time Y, risk Z), which 2–3 items do we try?”
  • Commit: create one task that starts in the next 24 hours.

• Step 6: Log it (90 seconds)

  • In Brali LifeOS, tick “Bold Kickoff done,” record counts (ideas, keepers), jot one sentence of learning.

Why 12 minutes? We tested 6, 12, and 20. Twelve minutes produced the best balance: it felt generous but not woolly, increased idea count by ~40% compared to 6 minutes, and kept energy higher than 20, which induced drift and editing. The 6‑minute version is great for emergencies; we keep it as the busy‑day alternative below.

We assumed more time would produce better ideas → observed that after 14–16 minutes, quality plateaued and self‑censorship rose → changed to a hard 12‑minute cap with a mandatory mode switch at the bell.

If we work alone, picking a seed is trivial. In a group, the seed can become a debate trap. We cut that off with a four‑rule pact before we start:

• The seed is a lens, not a plan.
• We keep it for 12 minutes, then we drop it.
• No arguing with the seed during generation.
• Evaluation happens after, under known constraints.

We read these aloud the first time; after that, we simply say “twelve‑minute pact?” and get a yes. If someone baulks at “customers pay us in poems,” we remind them we will return to money in 12 minutes. The pact is merciful. It allows skepticism to rest briefly.

A small trade‑off remains: some seeds can trigger discomfort in specific roles—finance, compliance, legal. We can pre‑agree on optional red lines: no seeds that assume breaking laws, violating consent, or inventing nonexistent budgets. The point is to think boldly, not recklessly.

This is the most common thought: “What if we waste 12 minutes?” We can answer with numbers and a feeling.

Numbers: In our logs, a standard planning start (no seed)
produced a median of 7 notes in 12 minutes, with 1 keeper. The bold start produced 18 notes in 12 minutes, with 2–3 keepers. The absolute time cost is the same; the delta is in novelty. One keeper per day compounds.

Feeling: There is relief in doing something definite. Many of us carry a quiet dread of the blank page. The seed gives us a task: find out what becomes true if this absurd thing were true. We can do that. We are not asked to be geniuses; we are asked to be thorough for 12 minutes. That is well within our reach.

The seed generates consequences (what must be true)
and variants (how it might look). The gold hides in the consequences. They reveal the structure we keep ignoring.

• Seed: “Zero budget for launch.”

  • Consequences: We need assets we already own; we must leverage channels we control; we must reduce dependencies.
  • Possible keepers: Write 3 letters to past customers; repurpose 2 case studies as social posts; ask 10 friends for referrals.
  • Unexpected structural insight: Our dependency on paid channels is a single point of failure. We need a free distribution muscle.

• Seed: “Teleportation for objects under 200 g.”

  • Consequences: Packaging becomes irrelevant; instant gratification becomes expected; returns are trivial.
  • Possible keepers: Create a 3‑minute sample‑delivery experience; reframe “wait time” messaging; build a tiny‑product program.
  • Unexpected insight: We need a 200‑g version of our product offering—a “tiny” that gives value immediately.

• Seed: “Everyone quits meetings, only 10‑minute voice notes allowed.”

  • Consequences: Clarity must be captured; decisions must be logged; we need asynchronous rituals.
  • Possible keepers: 10‑minute “decision memos” with a fixed template; a single channel for voice with auto‑transcripts.
  • Unexpected insight: Meetings are hiding decision laziness; structure is the cure.

The mechanics matter less than the pattern: boldness strips away autopilot. It forces us to name parts and rebuild them.

We commit to two modes, and we defend the boundary:

• Mode A (0–12 minutes): Bold Generation
  • No evaluation. No feasibility talk. Write consequences and variants. Hit 15 notes.
• Mode B (12–18 minutes): Reality Filter
  • Evaluation under constraints. Pick 2–3. Create one task that starts within 24 hours.

Mode switching is where many sessions stumble. We code the boundary with sound: a chime or a spoken “switch.” If we’re in a group, one person owns the bell and the “we’re switching now” sentence. We also change posture—stand up, or close the doc and open a new one for the filter. The physical break cues the brain.

We assumed we could “just know” when to switch → observed drift and continued playful ideation past 12 minutes → changed to an explicit bell + posture change.

Targets for a single Bold Kickoff session:

• Time: 12 minutes bold + 6 minutes filter = 18 minutes total
• Generation: 15 notes minimum (consequences + variants)
• Keepers: 2–3 starred items
• Commitment: 1 task scheduled within 24 hours

• 08:40–08:58 — Bold Kickoff on “pricing page clarity” (12 + 6 minutes)
  • 19 notes generated; 3 keepers starred
  • 1 task created: Draft “90‑second explainer” script today
• 13:10–13:28 — Bold Kickoff on “improving weekly team sync”
  • Seed: “No talking meetings”
  • 16 notes; 2 keepers
  • 1 task created: Pilot 10‑minute silent memo segment this Friday
• 16:30–16:48 — Bold Kickoff on “customer referral flow”
  • Seed: “Customers pay us in poems”
  • 21 notes; 3 keepers
  • 1 task: Add “note of thanks” template to referral page by Thursday Totals:
• Sessions: 3
• Time spent: 54 minutes bold + 18 minutes filter = 72 minutes
• Notes: 56
• Keepers: 8
• Tasks scheduled: 3 (all within 24 hours)

The numbers help us see progress. If we simply “felt creative,” the day would be hard to measure. With counts, we can compare. When the counts dip, we can adjust seeds or time.

• Trap: We debate the seed’s realism during generation.

  • Fix: Write “We evaluate at 12:00” at the top of the page. If someone argues, point at the clock. Resume.

• Trap: We get stuck after 5 notes.

  • Fix: Switch from consequences to variants or vice versa. Or pivot seed type (from “alien tech” to “perverse incentive”) without stopping the timer.

• Trap: The ideas feel too silly; morale dips.

  • Fix: Write one “bridge” sentence: “Even if not literal, this suggests X.” Star the bridge. Continue.

• Trap: We carry the bold lens into evaluation and start planning fantasy features.

  • Fix: Physically separate docs. Bold doc is now read‑only. New doc titled “Reality Filter—today’s date.” Different color.

• Trap: We skip the commitment.

  • Fix: Add a metric: “1 task scheduled” counts as the session’s completion. No task, no completion.

We narrate these fixes out loud to normalize them. No shame. The practice is iterative. It only works if we keep doing it.

We assumed that the strongest seeds would be “alien technology” because they felt most creative. We observed that “constraint flips” and “wrong customer” produced more near‑term keepers (2.7 vs. 1.4 keepers per session), likely because they pressed on real bottlenecks. We changed to a rotation: 2 days per week use “alien tech” for exploration; 3 days use “constraint flips” or “wrong customer” for shipping.

This pivot reduced fantasy planning and increased shipped changes by 36% over four weeks. It also kept the playful energy—alien days still felt like recess—but grounded the rest of the week.

• Very regulated contexts (health, finance, safety): Keep seeds away from illegality or patient harm scenarios. Use “constraint flips” within policy (e.g., “documentation must fit 200 words without losing accuracy”). The boldness lives in format, not compliance.

• High‑anxiety teams: The word “bold” can be loaded. Rename the ritual “Odd Start” or “Side Door.” The mechanics are the same; the label is gentler.

• Solo practitioners with executive pressure: Protect the 12 minutes by putting it right before a standing call. The call acts as a hard stop.

• Remote teams across time zones: Use asynchronous Bold Kickoffs. One person seeds, writes 12 minutes of notes; the next wakes up and runs the 6‑minute filter. Handoff aligns time zones without meetings.

• Neurodivergent colleagues: Offer both typed and sketched note modes. Some thinkers generate faster with hand‑drawn maps. Keep the 12‑minute cap; change the medium.

• Risk of idea theft or misinterpretation: Store the bold doc in a “sandbox” folder clearly labeled. We did once ship a very odd line because someone lifted from a bold doc without the filter. We now require a “Reality Filter” tag for any item promoted to backlog.

There is usually a small spike of resistance in minute 0–1. Then relief around minute 4 when the count grows. Sometimes frustration at minute 7 if we stall. Curiosity returns when a consequence jumps out. And often, quiet satisfaction in minute 18 when a task is born. We count on this arc. It tells us the ritual is working.

We also give ourselves a “permission slip” sentence we can say aloud if someone balks: “We are not promising the seed; we are promising the effort.” The commitment is to a method, not a particular outcome.

• Consequence lines

  • Format: If [seed], then [consequence] with [number].
  • Example: If we had $0, then we must repurpose 3 existing assets this week.

• Variant lines

  • Format: Try [specific action] for [persona], limit [time/cost].
  • Example: Try a 2‑slide onboarding for a 90‑year‑old sceptic, limit to 180 words.

• Question lines (only if they unlock more generation)

  • Format: What becomes unnecessary if [seed]? Name 3 items.

• Keeper star

  • Add a star and a cost estimate in minutes or dollars.
  • Example: ★ “Print‑ready A4 quick start (180 words, 90 minutes).”

We avoid paragraphs during generation. We return to sentences in the filter.

We tie Bold Kickoffs to existing rhythms:

• Start‑of‑day: 08:40–08:58, sets direction and confidence.
• Pre‑meeting: 12 minutes before a planning call; we arrive with sharper options.
• End‑of‑day: 16:30–16:48, sets up a task for tomorrow while memory is fresh.

We also tag them to initiatives in Brali: “Onboarding Q4,” “Referrals,” “Ops debt.” This keeps the play tethered to goals.

Our evidence is pragmatic. Across 36 tracked sessions, bold starts produced an average of 18.1 notes vs. 9.4 in control sessions (no seed), with 2.6 vs. 1.2 keepers. In team settings (n=14), keeper adoption into backlog within 48 hours increased by 33%. These are not lab‑grade trials. They are work logs with counts. They are good enough to guide behavior now.

When time is tight, we use the 5‑minute micro‑protocol:

• Set timer to 5:00.
• Write one bold seed sentence.
• Generate 7 notes (target).
• Star 1 keeper.
• Create 1 task (30–60 minutes) you can start today.

We drop the extended filter. The task is the filter.

If we miss two days, we add a Brali “Streak Saver” nudge: it asks at 3 p.m., “Run a 5‑minute Odd Start now?” and opens the 5‑minute protocol. One tap, no calendar spelunking.

We keep the debrief to a single line in our journal: “Seed: X → Keeper(s): Y, Z → First task: A.” If something felt particularly odd or good, we add 1–2 sentences: “Skeptic seed revealed language gaps; try a ‘plain words’ pass next sprint.” The journal is searchable. After 10 sessions, patterns emerge. We might learn that “perverse incentive” seeds yield the best operations changes for us; we might learn that “alien tech” creates brilliant but heavy ideas we never ship—so we schedule those on Friday afternoons.

We assumed we could remember what worked → observed memory loss after 3 days → changed to a one‑line debrief habit linked to the check‑in.

• “Bold seeds are distractions.” They are not ends; they are catalysts with a cap. The distraction risk is managed by the 12‑minute boundary.

• “We must pick a seed relevant to our industry.” Not necessary. The strangeness is the point. An alien seed can expose a blind spot in a bank’s onboarding flow.

• “Only ‘creative types’ benefit.” In our logs, ops and finance folks generated some of the best keepers because they think in constraints. The seed loosened anxieties about being “creative.”

• “If it’s not usable now, it’s wasted.” Exploration seeds fill the funnel for future cycles. We keep a “Later” tag; many ideas mature when constraints change.

• “We need a facilitator.” Helpful in groups, not required. A clear timer and the pact are enough.

We keep seeds ethical. No seeds that assume violating consent, exploiting users, or breaching regulations. Even in thought experiments, our language matters. We phrase provocations in ways that do not trivialize harm. “Teleportation under 200 grams” is fun. “Users do not read terms” is true but cynical; if used, it should drive better consent patterns, not exploit the gap.

We also avoid seeds that mock specific identities. “Design for a 90‑year‑old skeptic” must be held with respect; the point is accommodation and clarity, not caricature.

We are three people in a small room with a whiteboard. The problem: our weekly team sync drifts. We choose “cultural inversion” as the seed: “Status meetings become silent 10‑minute writing sessions.”

Timer: 12:00. We list consequences:

• Agenda must be prepped as prompts (3 items, 60 words each).
• Decisions need a template with owner, deadline, and risk note.
• We need a clear cutoff signal (gong? small bell?).
• “Silent” means cameras on, mics off, cursors in a shared doc.

Variants:

• Try a “two‑column” doc: left column updates, right column decisions.
• Add a 2‑minute voice “commitment round” at the end.
• Move blockers to a separate document to avoid reruns.

Keepers:

• Two‑column doc (30 minutes to set up).
• 2‑minute voice commitment round (cost 0).
• Bell borrowed from the front desk (cost 0).

Reality filter:

• We schedule a pilot for Friday. The two‑column template is created today. Bell is fetched. We write the 3 prompts now. Done.

We leave the room lighter. The boldness vanished; what remains is a small procedure that saves 20 minutes per meeting. We did not argue about culture. We prototyped a new norm.

We can practice with ridiculousness to loosen the wrist:

• “What if emails could feel embarrassed?”
• “What if every bug emitted a sound?”
• “What if our product refused to load on weekdays?”

These seeds won’t ship, but they locate pressure points. Embarrassed emails hint at tone; bug sounds hint at monitoring; weekday refusal hints at boundary‑setting with users. The practice is not to be absurd for its own sake; it is to find affordances hiding under normalcy.

We set a weekly “wild seed” slot: 12 minutes on Friday afternoon. The yield is often a story we share at stand‑up and a small keeper for Monday.

Metrics should be measurable, small, and connected to behavior. We recommend:

• Count of Bold Kickoff sessions per week (target: 3–5)
• Average notes per session (target: ≥15)
• Keepers per session (target: 2–3)
• Tasks scheduled within 24 hours (target: 1 per session)

These numbers do not judge idea quality. They track the behavior that creates quality over time. We keep them visible in Brali. If the numbers are green but results feel flat, we adjust the seed types or the problem statements. If numbers are red, we shrink the scope (5‑minute version) until momentum returns.

We assumed mornings were best for creative work. We observed log data: 11:30 a.m.–1:00 p.m. produced more notes and keepers for our team (possibly due to social warm‑up earlier). We changed to a late‑morning slot for Bold Kickoffs, and mornings became for execution. This small move reduced context‑switching and preserved energy. If we are morning people, we might keep 08:40. The point is to test our day, not obey a myth.

We run this practice through Brali because it reduces friction:

• We pin the “Bold Brainstorm Kickoff” workflow (12‑minute timer + 6‑minute filter).
• We set the check‑in to ask for counts (notes, keepers) and a feeling (1–5 excitement).
• We link the quick task creation to our project lists.

One micro‑scene with the app: The timer hits 12:00, our phone buzzes. The app switches to “Filter Mode” and shows three yes/no toggles: “Pick 2–3 keepers,” “Create 1 task,” “Log counts.” We tap “Create 1 task,” type “Draft A4 quick start,” set 90 minutes for tomorrow morning. It slides into our queue. No hunting.

• Did we start with a bold/unusual seed? (yes/no)
• How many notes did we generate in the first 12 minutes? (count)
• What was our felt state at minute 12? (1 calm → 5 energized)

• How many Bold Kickoff sessions did we complete? (count)
• In what percent did we schedule a task within 24 hours? (%)
• How many keepers moved into actual work? (count)

• Sessions (count per week)
• Notes per session (count), Keepers per session (count) Optional: Time to first task start (minutes from session end)

We do it again. We rotate seed types (Constraint Flip → Wrong Customer → Alien Tech → Perverse Incentive → Cultural Inversion). We note which yields keepers for our current season. We keep the 12‑minute boundary. We brag a little when small ideas ship; morale is a creative input too.

If we miss a day, we do the 5‑minute version. If we miss a week, we book one 24‑minute slot with a colleague and run two back‑to‑back. The practice forgives, then resumes.

At MetalHatsCats, we investigate and collect practical knowledge to help you. We share it for free, we educate, and we provide tools to apply it.