blind-spot-scan

name

blind-spot-scan

description

Map the full problem space before or after ideation to identify what you forgot to think about. Use BEFORE generation to ensure brainstorming covers the whole territory, or AFTER generation to check whether your ideas left entire regions of the problem unaddressed. Triggers on "what am I missing," "what haven't I thought of," "is there a blind spot here," "what else should I consider," "did I cover everything," or any moment where the user suspects their thinking has been thorough within a narrow band but oblivious to adjacent territory. Also use when a brainstorm feels solid but suspiciously focused, when an Anti-Homogeneity Check returns a high grade but the ideas still feel incomplete, or when a strategy covers the offensive plan but has no defensive layer. Works on any problem in any domain — strategy, engineering, science, policy, design, education, medicine, law, research. This is the skill that asks not "are your ideas different?" but "what parts of the problem did you forget existed?"

Blind Spot Scan

Your ideas are diverse. They're also incomplete.

The Anti-Homogeneity Check tells you whether your ideas are different from each other. This skill tells you whether your ideas cover the problem.

These are different failures. A brainstorm can score Grade A on diversity — nine distinct clusters, no overlap — and still be blind to half the problem because every cluster lives in the same region of the map.

This happens in every domain.

A nonprofit generates 23 ideas for fundraising. Nine clusters. Genuinely diverse. Zero ideas about existing donors. Zero about corporate revenue. Zero about what happens if the strategy fails.

A research lab brainstorms approaches to antibiotic resistance. Every idea involves screening compounds differently. Zero ideas about redefining what "resistance" means. Zero about the incentive structure that keeps labs running the same experiments. Zero about what to do if the approach takes longer than funding lasts.

A school district generates ideas for improving graduation rates. Every idea targets students. Zero ideas address what teachers need. Zero about the families who left the district. Zero about the employers who could create reasons to graduate.

The blind spots are different in each case. The pattern is the same: the problem space was bigger than anyone mapped before generating.

This skill maps first. Then you generate into the map. Then you check what's empty.

---

How It Works

Two modes.

Pre-generation mode: Run before brainstorming. Produces a map of the full problem space. Generation skills then produce ideas that consciously address different regions.

Post-generation mode: Run after brainstorming. Takes the ideas you already have, lays them on the map, and shows you the empty zones.

Same map either way. The difference is timing.

---

Building the Map

The map has seven dimensions. Not every problem touches all seven. But scanning all seven ensures you see what's there and what isn't before you decide what matters.

The dimensions are written in universal terms because every problem, in every domain, has stakeholders, resources, time pressures, strategic options, forms of value, delivery mechanisms, and failure modes. The language changes. The structure doesn't.

1. Who's affected?

List every distinct group that the problem touches. Not just the obvious target. Everyone.

For a nonprofit struggling with membership: the new donors (obvious target), the existing long-time donors (often forgotten), the board, the staff, local businesses, corporate partners, foundation funders, and the community the nonprofit serves who may not be donors at all.

For a hospital reducing readmissions: the patients (obvious target), the nurses who do discharge planning, the primary care physicians receiving patients back, the families managing care at home, the insurance companies paying the bills, the social workers, the pharmacies filling prescriptions, and the home health agencies who see what actually happens after discharge.

For a software team improving a codebase: the current engineers (obvious), future engineers who'll inherit it, the product managers waiting on features, the customers experiencing bugs, the on-call team paged at 3am, the one person who understands the payments module, and the recruiting team trying to hire people who don't want to work on legacy code.

For a climate policy proposal: the communities most affected by emissions, the industries required to change, the workers whose jobs are at risk, the municipalities that must implement it, the neighboring jurisdictions affected by spillover, and the next administration that will inherit the policy.

Most brainstorms address 1-2 of these groups. The map shows all of them so you can deliberately choose which to target and which to ignore — rather than accidentally ignoring five.

2. What resources already exist?

List what's already in play that ideas could build on, redirect, or repurpose. Not new things. Existing things.

• Funding, budget, or money already flowing
• People already engaged (how many, how deeply)
• Relationships already built (partners, allies, collaborators)
• Infrastructure already in place (systems, tools, spaces, programs)
• Credibility, reputation, or trust already earned
• Data, research, or knowledge already collected
• Content, publications, or materials already created
• Equipment, facilities, or physical assets already available

The specific inventory changes by domain. A lab has grant funding, published papers, existing datasets, collaborator networks, and equipment. A startup has users, code, relationships, and brand. A school has teachers, facilities, curricula, community ties, and parent networks. A city agency has budgets, staff, existing programs, inter-agency relationships, and public trust (or distrust).

Most brainstorms invent from scratch. The map shows what's already there so ideas can build on existing assets instead of ignoring them.

3. What time horizons are in play?

Every problem has multiple urgencies. Map them:

• This week — What needs to happen immediately?
• This quarter — What could show results in 90 days?
• This year — What plays out over 6-12 months?
• Multi-year — What's the 3-5 year structural play?

Most brainstorms cluster in one time horizon.

In business: usually "this quarter" or "this year." In science: usually "this grant cycle." In policy: usually "this administration." In engineering: usually "this sprint" or "this release." In education: usually "this school year."

The map forces you to see whether you have ideas for all four or only one.

4. What strategic postures are available?

Not every idea needs to be offensive. Map the postures:

• Offense — Grow. Build. Expand. Create new.
• Defense — Protect what you have. Retain. Prevent loss.
• Bridge — Survive the gap. Buy time. Temporary measures.
• Retreat — Cut. Shrink. Consolidate. Refocus.
• Transform — Change what you are. New model. New identity.

Most brainstorms are pure offense. "How do we grow?" "How do we discover?" "How do we improve?"

The map forces the question: "What if growth doesn't come fast enough? What's the bridge? What gets cut?"

In business, 23 offensive ideas and zero bridge or defensive ideas is a blind spot that could be fatal if revenue drops faster than the new strategy fills it.

In research, 15 ideas for new experiments and zero ideas for what to publish from existing data if the grant isn't renewed is the same blind spot in different clothes.

In policy, a full reform agenda and zero contingency for political opposition is the same blind spot again.

5. What types of value are available?

Map the different kinds of value the situation could produce or exchange. The specific types depend on the domain, but most problems have more value dimensions than anyone thinks to address:

• Material value — Money, resources, physical things
• Knowledge value — Data, insight, expertise, research
• Relationship value — Trust, alliances, networks, community
• Capability value — Skills built, capacity created, infrastructure developed
• Attention value — Awareness, visibility, credibility, influence
• Time value — Speed gained, delays prevented, optionality preserved

In a business context, most brainstorms focus on revenue and attention. In a research context, most brainstorms focus on publications and grants. In a policy context, most brainstorms focus on the intervention itself and ignore political capital, coalition-building, or the data infrastructure needed to measure whether it worked.

The map shows the full menu so you can see which value types your ideas are capturing and which they're leaving on the table.

6. What mechanisms could deliver ideas?

Map the categories of action available. These shift by domain, but the structural question is the same: through what channels or modes of action could ideas be implemented?

• Direct intervention — Doing the thing yourself, hands-on
• System or process change — Changing how things are structured or governed
• Communication or narrative — Changing what's said, published, taught, or framed
• Experience or environment — Changing what people encounter physically or socially
• Partnership or collaboration — Working through or with others
• Technology or tools — Building or deploying systems
• Incentive or policy — Changing the rules, rewards, or costs

In business, brainstorms often cluster on communication and technology. In medicine, on direct intervention (treatment). In education, on curriculum and pedagogy. In engineering, on architecture and tooling.

The map reveals which mechanisms have ideas behind them and which have been completely ignored.

7. What could go wrong?

Map the failure modes:

• What happens if the primary strategy takes longer than expected?
• What happens if the target audience (patients, users, students, voters, donors) doesn't respond as hoped?
• What external events could change the landscape?
• What internal constraints could bottleneck execution?
• What dependencies could break?

Most brainstorms don't include ideas for failure scenarios because it feels negative. But "what do we do if this doesn't work" is one of the most valuable regions on the map.

A product strategy with no ideas for what happens if adoption is slow. A research agenda with no ideas for what to do with null results. A policy proposal with no ideas for surviving legislative opposition. A curriculum redesign with no ideas for teacher resistance.

Each is the same blind spot: all offense, no contingency.

---

Pre-Generation Mode

Run the seven dimensions against the desire statement or brief. Produce the map as a clean list of what exists in each dimension.

Don't generate ideas. Generate the territory.

Then tell the user: "Here's the full map of your problem space. You have [X] groups affected, [Y] time horizons in play, and [Z] mechanism categories available. When we brainstorm, which regions do you want to prioritize? And which are you comfortable leaving empty?"

Let them choose. Then feed the prioritized regions into whatever generation skill they're using.

This prevents the most common blind spot: not the ideas you rejected, but the ideas you never considered because you never looked in that direction.

---

Post-Generation Mode

Take the existing set of ideas. Build the map. Then lay each idea on the map and mark which regions it covers.

Present the coverage report:

Covered: List which dimensions and regions have at least one idea addressing them.

Empty: List which dimensions and regions have zero ideas. These are your blind spots.

Overweight: List which regions have 3+ ideas. These are your gravity wells — the places your thinking naturally goes that may be pulling attention from regions that matter more.

Coverage report format

Keep it visual and blunt. Here is what it looks like across domains:

For a nonprofit membership strategy:

"Who's affected: New transplants — 19 ideas. Heavily covered. Existing donors — 0 ideas. Blind spot. Corporate partners — 0 ideas. Blind spot. Staff — 0 ideas. Blind spot. Board — 2 ideas. Light coverage.

Time horizons: This quarter — 8 ideas. This year — 12 ideas. Multi-year — 3 ideas. This week — 0 ideas. No bridge strategy.

Strategic posture: Offense — 22 ideas. Defense — 0 ideas. Blind spot. Bridge — 0 ideas. Blind spot. Transform — 1 idea."

For a hospital readmission reduction strategy:

"Who's affected: Patients — 14 ideas. Heavily covered. Nurses — 3 ideas. Primary care physicians — 0 ideas. Blind spot. Families — 1 idea. Light coverage. Pharmacies — 0 ideas. Blind spot.

Mechanisms: Direct intervention — 12 ideas. Overweight. System change — 4 ideas. Technology — 2 ideas. Incentive/policy — 0 ideas. Blind spot.

Failure modes: Slow adoption — 0 ideas. No contingency."

For a codebase improvement plan:

"Who's affected: Current engineers — 11 ideas. Heavily covered. Future engineers — 0 ideas. Blind spot. On-call team — 1 idea. Light. Product managers — 0 ideas. Blind spot.

Time horizons: This sprint — 8 ideas. This quarter — 3 ideas. Multi-year — 0 ideas. Blind spot.

Mechanisms: Architecture/tooling — 9 ideas. Overweight. Process/governance — 2 ideas. Communication/documentation — 0 ideas. Blind spot. Incentive (motivation to work on it) — 0 ideas. Blind spot."

The user sees immediately where they're rich and where they're empty. They decide what to do about it.

---

After the Scan: Triage and Fill

The scan is not the end. It's the middle.

After presenting the coverage report, do three things:

1. Ask the user to pick

Present the blind spots as a pick list. Not every blind spot matters. Some are critical gaps. Some are nice-to-have. Some are intentionally irrelevant.

Only the user knows which is which.

Say:

"Here are the empty regions. Which 2-3 feel like they'd actually change your thinking if they had ideas in them? The rest we'll leave empty on purpose."

Frame it as a choice, not a homework assignment. Picking 2-3 is easy. Being told you have 8 blind spots and figuring out what to do about it is paralyzing.

Some gaps should stay empty. Not every dimension needs ideas. The value of the map is making that a conscious decision rather than an accidental one.

"We have no ideas about corporate revenue" might lead to "let's brainstorm that" or it might lead to "corporate revenue isn't the play here, we're deliberately ignoring it."

"We have no ideas about null results" might lead to "let's plan for that contingency" or it might lead to "the experiment either works or we pivot entirely, there's no middle ground to plan for."

Both are valid. But only the second is intentional.

2. Turn each chosen blind spot into a question

Don't feed blind spots into Strip Down. Don't try to extract a desire statement. The blind spot is already specific enough. Just convert it into a direct question.

"Primary care physicians are empty" becomes: "What role do primary care physicians play in preventing readmissions, and what would they need from us to play that role better?"

"Receiver experience is empty" becomes: "What does the person on the other end actually see, and why would they act on it?"

"Defensive strategy is empty" becomes: "What happens if this doesn't work fast enough — what's the bridge?"

"Documentation is empty" becomes: "How does anyone who wasn't in the room understand what was built and why?"

One question per blind spot. Plain language. No frameworks.

3. Generate immediately into the chosen gaps

Do not ask which generation skill to use. Do not offer a menu. Just go.

Use Short Think if the user needs volume fast. Use the Guilford Engine if the gap is complex enough to need structured passes. Use raw brainstorming for everything else.

The user picked the gaps. You converted them to questions. Now fill them.

Once the new ideas exist, merge them with the original set. The combined portfolio should now cover the regions that were empty before.

Optionally re-run the Anti-Homogeneity Check on the merged set to verify the new ideas are diverse from the originals and not just a second cluster living in the same region.

---

When to Use This Skill

Pre-generation:

• The problem is complex (multiple stakeholders, competing constraints, multiple time pressures)
• The stakes are high (a wrong blind spot could be expensive, dangerous, or irreversible)
• The user wants comprehensive coverage, not just creative ideas
• Previous brainstorms have felt "solid but narrow"

Post-generation:

• After the Guilford Engine (check coverage of the final output)
• After the Persona Divergence Engine (personas diverge from each other but might all diverge in the same direction)
• After the Anti-Homogeneity Check finds Grade A diversity but you suspect the ideas are still living in one region of the map
• Before presenting a final recommendation to make sure nothing critical was missed

Don't use:

• For simple creative tasks (naming, taglines, hooks — these don't need maps)
• When the user wants speed over coverage (use Short Think instead)
• When the problem is genuinely narrow and the map would be mostly empty by design

---

Relationship to Other Skills

After Strip Down: Strip Down extracts the desire. Blind Spot Scan maps the territory the desire touches. Generation skills fill the territory. This is the ideal pre-generation pipeline: desire → map → generate.

Before Generation Skills: The map tells the Guilford Engine which problems to address in Pass 1 (fluency). It tells the Persona Divergence Engine which perspectives might see regions the default panel would miss. It tells Short Think which directions to aim the volume.

After Anti-Homogeneity Check: Anti-Homogeneity Check says "your ideas are diverse" (or not). Blind Spot Scan says "your ideas cover the problem" (or not). Run them in sequence after any major brainstorm: diversity first, then coverage. Together they answer: "Do I have different ideas, and do those different ideas address the full problem?"

Before any final recommendation: If the user is about to commit to a strategy, run a Blind Spot Scan to make sure the strategy doesn't have a fatal gap in a dimension nobody thought to check. This is the pre-flight checklist.