Before You Commit · 2 of 10Before You Commit
The Assumption Beneath the Idea
Why the most important part of a project may be the thing nobody has questioned—and how to identify and test its load-bearing assumptions.
Every idea is built on something that has to be true.
A new business might need customers to pay a particular price.
A cultural programme might need a particular community to participate.
An artist releasing a limited-edition collection might need enough collectors to value the work at the proposed price.
A community project might depend upon trusted relationships with local organisations.
A technology proposition might require people to change an established behaviour.
A new event might need a particular number of attendees before it becomes financially viable.
A funding strategy might depend upon an organisation being eligible for a programme that has not yet been examined closely.
These are not necessarily weaknesses.
They are assumptions.
And assumptions are unavoidable.
The danger begins when an assumption becomes so embedded within an idea that nobody remembers it is an assumption at all.
At that point, people stop asking:
“Is this true?”
and begin planning as though it is.
That is how an uncertain proposition becomes a budget.
A forecast.
A business model.
A programme.
A funding application.
A marketing campaign.
A six-month development process.
Eventually, substantial resources can be resting upon something nobody actually tested.
The challenge, therefore, is not to create an idea without assumptions.
It is to discover which assumptions matter enough to test before you commit.
Every Plan Contains a Theory About the Future
Consider a founder planning a subscription service.
The proposition may appear straightforward:
Develop the service.
Build the website.
Set the price.
Launch a marketing campaign.
Acquire customers.
Grow recurring revenue.
But beneath those actions sits an invisible chain:
People experience the problem we believe they experience.
They recognise that problem.
They care enough about solving it.
Our solution is attractive to them.
They understand the proposition.
They trust us sufficiently to try it.
They are willing to pay our proposed price.
They continue finding enough value to remain subscribers.
We can acquire them at an economically sustainable cost.
We can deliver the service reliably as demand grows.
The visible plan might contain ten actions.
The invisible plan contains ten assumptions.
And the success of the visible plan depends upon enough of the invisible ones being true.
This is why assumptions deserve serious attention.
A business plan is not merely a description of what an organisation intends to do.
It also contains a theory about how reality will respond.
Not All Assumptions Are Equal
Suppose a project contains twenty assumptions.
Trying to investigate every one with equal intensity would be inefficient.
Some will have relatively minor consequences if they prove wrong.
Others could force a small adjustment.
A few could threaten the entire proposition.
These are what we might call load-bearing assumptions.
Think about a building.
Not every wall carries the same structural weight.
Remove a decorative partition and relatively little happens.
Remove the wrong structural support and the consequences can be profound.
Ideas work similarly.
The most important question is therefore not:
“What assumptions are we making?”
It is:
“Which assumptions is the idea most dependent upon?”
Finding the Load-Bearing Assumption
Imagine an independent creative entrepreneur planning a premium physical product.
They might be considering:
branding;
packaging;
website design;
photography;
delivery;
production quantities;
social media;
payment systems;
launch dates;
collaborations;
and advertising.
All of these decisions matter.
But underneath them may sit a much more consequential assumption:
Enough customers will pay £175 for the product.
If customers will, many of the downstream questions become worth solving.
If customers value the product but will only pay £70, the economics may require redesign.
If customers are unwilling to purchase it at all, improving the packaging is unlikely to rescue the proposition.
The pricing and demand assumption is therefore carrying considerably more weight than the colour of the website button.
Yet organisations frequently spend enormous amounts of energy optimising relatively safe decisions while leaving their most dangerous assumptions untouched.
Why?
Because the safer questions are easier.
You can choose packaging.
You can design a logo.
You can build a website.
You can produce a brochure.
Testing whether strangers will actually pay is more uncomfortable.
It introduces the possibility that reality may disagree with the idea.
But that is precisely why the question matters.
Importance × Uncertainty × Consequence
A useful way to prioritise assumptions is to examine three dimensions.
Importance
How central is this assumption to the proposition?
If it proves false, does the idea still make sense?
Uncertainty
How much credible evidence currently supports it?
Are we highly confident—or largely guessing?
Consequence
What happens if we are wrong?
Would we make a small adjustment, or would the economics, audience, delivery model or entire proposition need to change?
An assumption that is highly important, highly uncertain and highly consequential deserves attention early.
We can express this conceptually as:
Assumption Priority = Importance × Uncertainty × Consequence
This is not intended to produce an artificially precise mathematical score.
It is a thinking tool.
Its purpose is to stop teams treating every unknown as though it deserves equal attention.
Ask: What Must Be True?
One of the simplest techniques for exposing assumptions is repeatedly asking:
What must be true for this to work?
Suppose a cultural organisation proposes an evening programme intended to engage younger audiences.
What must be true?
Young people must find the programme relevant.
They must hear about it.
The language used to describe it must resonate.
The venue must feel accessible.
The timing must work.
Transport may need to be available.
Ticket pricing must be appropriate.
Participants may need to feel that the space is genuinely for them.
The programme itself must deliver on the promise made by the marketing.
They must have a reason to return.
Each answer reveals another layer of the proposition.
Now ask again:
How do we know?
That is where assumptions begin separating from evidence.
Evidence and Assumptions Are Not the Same Thing
This distinction sounds obvious.
In practice, it is remarkably easy to lose.
An assumption is something we currently believe, expect or need to be true without having sufficient evidence to establish it.
Evidence is information, observation or behaviour that gives us a credible reason to increase or decrease our confidence in a claim.
Consider these statements:
“People will buy this product.”
That is an assumption.
“Twenty-seven people purchased the prototype at £95 during our four-week test.”
That is evidence.
Or:
“The community will support the programme.”
Assumption.
“Forty-three residents participated in structured consultation, and 31 said the identified issue was a high priority.”
Evidence of those participants’ expressed views.
Notice the qualification.
That evidence still does not prove that the entire community supports the programme.
Good evidence does not give us permission to claim more than it actually demonstrates.
That is one of the most important disciplines in evidence-led decision-making.
Evidence Changes Confidence — It Does Not Automatically Create Certainty
Evidence rarely gives us absolute certainty.
Instead, it changes how confident we should reasonably be.
Imagine a founder believes customers will pay £100 for a service.
Before testing, confidence is based largely on assumption.
Then ten people say they like the idea.
That provides some evidence of interest.
Then twenty people join a waiting list.
Confidence increases.
Then twelve people pay £100 for a pilot.
Now there is behavioural evidence of willingness to pay.
Then ten of those customers return and purchase again.
The evidence becomes stronger still.
At each stage, we learn something different.
The crucial point is that evidence accumulates.
It can strengthen a proposition.
It can weaken it.
It can reveal that the original assumption was partly right but incomplete.
It can also reveal an entirely different opportunity.
Evidence is therefore not the opposite of uncertainty.
It is one of the tools we use to reduce uncertainty intelligently.
Strong Evidence Matches the Claim
A common mistake is having evidence—but using it to support a claim it does not actually prove.
Suppose 5,000 people visit a website.
That is evidence of website traffic.
It is not automatically evidence that 5,000 people want to buy the product.
Suppose 300 people like an Instagram post.
That is evidence of engagement with the post.
It is not evidence that 300 people will attend an event.
Suppose 50 people tell you they like a business concept.
That is evidence of positive reaction.
It is not necessarily evidence of willingness to pay.
Suppose an organisation says it is interested in collaborating.
That is evidence of interest.
It is not yet evidence of a committed partnership.
The strength of evidence depends partly on how closely it corresponds to the claim being made.
If the claim is:
“People will pay for this,”
then actual purchasing behaviour is usually more informative than likes.
If the claim is:
“This community considers the issue important,”
then meaningful engagement with people from that community is more relevant than an internal management meeting.
If the claim is:
“We can deliver this reliably,”
then a successful prototype may be useful, but a real-world pilot may provide stronger operational evidence.
Ask:
Does this evidence actually support the claim—or merely something adjacent to it?
Evidence Has Different Strengths
It can be useful to imagine evidence as existing on a spectrum.
Opinion
“I think customers will like it.”
Anecdote
“Three people told me they liked it.”
Expressed interest
“Forty people joined the waiting list.”
Behavioural evidence
“Twenty people paid for the pilot.”
Repeated behaviour
“Sixteen purchased again.”
Independent corroboration
“Comparable behaviour appears in credible external market evidence.”
This is not a universal hierarchy.
Different questions require different forms of evidence.
Qualitative interviews may be extremely powerful when investigating motivations, language, trust or cultural meaning.
Quantitative data may be more useful when measuring scale, frequency or conversion.
A prototype may answer a technical question better than a survey ever could.
A community conversation may reveal something no spreadsheet contains.
The principle is not:
Numbers are stronger than stories.
The principle is:
Use evidence appropriate to the question.
Assumptions Are Not Bad
It is important not to turn the word assumption into an accusation.
Innovation would be impossible without assumptions.
Every new proposition involves beliefs about a future that has not happened yet.
The problem is not:
“We have assumptions.”
The problem is:
“We do not know which statements are assumptions.”
Once an assumption is visible, it becomes useful.
We can examine it.
Research it.
Challenge it.
Test it.
Monitor it.
Build scenarios around it.
Or consciously accept the risk.
An invisible assumption controls the decision.
A visible assumption can become part of the decision.
A Simple CIS Test
When examining an important statement about an idea, ask four questions:
1. What exactly are we claiming?
Make the statement specific.
2. What evidence supports it?
Identify actual information rather than confidence or repetition.
3. What does that evidence genuinely demonstrate?
Avoid stretching evidence beyond its limits.
4. What remains assumed?
Identify the gap between what the evidence establishes and what the proposition still requires us to believe.
For example:
Claim: Customers will pay £150.
Evidence: Fifteen potential customers said the service sounded valuable.
What the evidence demonstrates: There is some expressed interest in the proposition.
What remains assumed: That interest will convert into purchases at £150.
Now the next research question becomes obvious:
Test willingness to pay.
That is the practical value of separating evidence from assumption.
It turns vague uncertainty into something actionable.
Cultural Assumptions Can Be Load-Bearing Too
Some of the most consequential assumptions are not financial or technological.
They are cultural.
An organisation might assume:
“This community will trust us.”
“This language will resonate.”
“People will feel represented.”
“This location is accessible.”
“Younger audiences want digital experiences.”
“Older audiences prefer traditional experiences.”
“This community needs what we are offering.”
“These people are difficult to reach.”
Each statement may contain some truth.
But each can also flatten diverse people into a convenient category.
Communities are not homogeneous.
Age does not automatically determine behaviour.
Identity does not guarantee preference.
Geography does not automatically produce shared experience.
Representation is not identical to trust.
Access is not merely physical proximity.
Participation is not the same as attendance.
This is where cultural intelligence becomes strategically important.
A proposition can be financially plausible and operationally deliverable while still being built upon a weak understanding of the people it expects to participate.
Ask Who Had the Power to Define the Problem
There is an even deeper layer.
Sometimes the assumption is not inside the proposed solution.
It is inside the definition of the problem itself.
Consider a statement such as:
“Young people are not engaging with our organisation.”
That might be factually accurate.
But several interpretations are possible.
Young people may lack interest.
Or the programme may not feel relevant.
The communication channels may be wrong.
The language may feel institutional.
The venue may carry particular associations.
Opening times may conflict with people’s lives.
Prices may create barriers.
Young people may already be culturally active elsewhere, meaning the organisation is measuring engagement only through its own institutional lens.
The original problem:
“How do we get young people to engage?”
might eventually become:
“What would need to change about our organisation for engagement to become meaningful to more young people?”
That is a very different question.
Sometimes challenging an assumption does not merely improve the answer.
It changes the question.
Positive Feedback Can Hide Assumptions
Imagine presenting an idea to ten people.
Eight respond enthusiastically.
“Brilliant.”
“I’d definitely be interested.”
“You should do it.”
It feels encouraging.
And it is encouraging.
But what exactly has been established?
Perhaps the assumption being tested is:
People find the idea appealing when described to them.
The feedback provides some evidence for that.
But if the real assumption is:
People will pay £150 for this service, the evidence is considerably weaker.
Interest is not purchase.
Attention is not adoption.
Praise is not commitment.
Intent is not behaviour.
This does not make qualitative feedback worthless.
It means the evidence must match the claim.
The Assumption Ladder
One useful way of understanding this is to imagine a ladder of increasing commitment.
Someone sees the proposition.
Someone pays attention.
Someone expresses interest.
Someone asks for more information.
Someone joins a waiting list.
Someone provides contact details.
Someone books.
Someone pays.
Someone uses the service.
Someone returns.
Someone recommends it.
Each step provides different evidence.
The further someone travels up the ladder, the more behavioural evidence we obtain.
This matters because projects sometimes make decisions based on evidence from the bottom of the ladder while assuming behaviour at the top.
Ten thousand social-media views do not automatically equal customers.
Five hundred likes do not automatically equal ticket sales.
One hundred survey respondents saying they would consider purchasing do not automatically equal one hundred purchases.
The assumption remains until behaviour provides stronger evidence.
Assumptions About Capacity Matter Too
Founders frequently focus on external assumptions:
Will customers buy?
Will audiences attend?
Will partners participate?
But some of the most dangerous assumptions are internal.
Can we actually deliver this?
Do we have sufficient time?
Do we possess the necessary expertise?
Can the organisation support increased demand?
What happens if the founder becomes unavailable?
Can our technology handle growth?
Can our cash flow survive the gap between expenditure and revenue?
Does the team have capacity alongside existing commitments?
Are we depending upon unpaid labour?
Does the model still work once the real cost of delivery is included?
A proposition can have genuine demand and still fail because the organisation underestimated what fulfilling that demand required.
Demand validation and delivery validation are different questions.
Both matter.
Assumptions About Partners
Partnerships create another category of hidden dependency.
A conversation goes well.
An organisation expresses interest.
Someone says:
“We would love to explore this.”
The project plan subsequently lists them as a likely partner.
Then the funding application assumes their participation.
Then the delivery model depends upon their venue, audience or expertise.
But no formal commitment exists.
An expression of interest has quietly become infrastructure.
This does not mean informal conversations are unimportant.
It means they should be classified accurately.
Interest is evidence of interest.
It is not necessarily evidence of commitment.
Assumptions About Technology and AI
The rapid development of artificial intelligence creates another important distinction.
A technology being capable of performing a task does not automatically mean an organisation is ready to use it effectively.
An AI tool might be capable of:
research;
analysis;
content creation;
customer support;
automation;
forecasting;
personalisation;
or workflow coordination.
But implementation may depend upon:
data quality;
staff capability;
integration;
governance;
privacy;
cost;
human oversight;
reliability;
customer acceptance;
and organisational change.
The hidden assumption can therefore become:
“Because the technology can do this, we can do this.”
Those are not the same statement.
Technical capability is only one component of operational readiness.
Turn Assumptions Into Testable Hypotheses
Once an important assumption has been identified, the next step is not automatically to believe or reject it.
Turn it into something testable.
Instead of:
“People want this.”
try:
“We believe at least 20 people from our target audience will register for a paid pilot within four weeks when offered the proposition at £75.”
Instead of:
“The community supports the project.”
try:
“We believe the proposed participants will identify the problem as important and consider the proposed approach credible when we conduct structured engagement.”
Instead of:
“Partners will support us.”
try:
“We believe at least three relevant organisations will commit a defined resource after reviewing the partnership proposal.”
Specificity creates the possibility of learning.
Test the Dangerous Assumption Early
There is a natural temptation to test whatever is easiest.
But the most useful experiment often targets whatever is most consequential.
If the biggest uncertainty is willingness to pay, test price.
If it is participation, test participation.
If it is technical feasibility, prototype the difficult technical component.
If it is community trust, conduct meaningful engagement before designing the entire programme.
If it is partner dependency, seek actual commitment.
If it is delivery capacity, simulate or pilot delivery.
This principle can be summarised simply:
Test the assumption most capable of changing the decision.
The Cheapest Useful Test
Testing does not always require an expensive pilot.
Sometimes a useful test could be:
a customer interview;
a prototype;
a landing page;
a waiting list;
a pricing experiment;
a small workshop;
a pre-order;
a limited edition;
a sample event;
a partnership proposal;
a technical proof of concept;
a community conversation;
or a small-scale delivery simulation.
The objective is not to create a miniature version of everything.
It is to obtain decision-relevant evidence.
The question becomes:
What is the smallest credible action that could tell us whether this assumption is likely to be true, false or in need of revision?
The word credible matters.
A test should resemble the decision you are trying to understand closely enough to produce useful evidence.
If you need to know whether people will pay, asking whether they like the idea is not enough.
If you need to know whether a community will participate, collecting opinions from people with no connection to that community is unlikely to answer the question.
If you need to know whether the team can deliver the service, producing a presentation about the delivery model is not the same as attempting a small real-world delivery.
The cheapest test is not necessarily the test requiring the least effort.
It is the least expensive test capable of reducing the uncertainty that matters.
Define the Result Before You Run the Test
Experiments become much less useful when organisations decide what the results mean after seeing them.
Suppose a founder launches a waiting-list page.
Forty people register.
Is that encouraging?
Perhaps.
But compared with what?
How many people saw the offer?
Who were they?
Were the registrations from the intended audience?
Did people understand what they were joining?
Would forty registrations justify the next investment?
Without an agreed threshold, almost any result can be interpreted favourably.
Before running a test, define:
What are we testing?
What result would increase our confidence?
What result would reduce our confidence?
What result would remain inconclusive?
What decision might follow each outcome?
For example:
Assumption
People will pay £75 for a two-hour diagnostic workshop.
Test
Offer twenty suitable prospective customers the opportunity to book a real pilot at £75.
Confidence-increasing result
At least six paid bookings.
Confidence-reducing result
No paid bookings despite evidence that the offer was understood and reached the intended audience.
Inconclusive result
High interest but insufficient reach, unclear communication or technical problems affecting booking.
Now the organisation is not simply launching an activity.
It is creating a learning mechanism.
Do Not Move the Goalposts
Assumptions become difficult to challenge when the success criteria keep changing.
Before the test:
“If ten people pay, we know the idea works.”
After four people pay:
“Four is actually very encouraging for a first attempt.”
Perhaps it is.
But the original threshold should not disappear without explanation.
Changing an interpretation is sometimes justified.
New evidence can reveal that the original target was unrealistic or poorly designed.
But the change should be made consciously and recorded.
Otherwise every result becomes success and no assumption can ever genuinely fail.
This is confirmation bias disguised as experimentation.
A credible testing process must allow reality to say no.
A Failed Assumption Is Not Necessarily a Failed Idea
Suppose customers do not pay £175.
Several explanations remain possible.
They may not value the proposition.
They may value it but not at that price.
The intended audience may be wrong.
The explanation may be unclear.
The trust required for purchase may not yet exist.
The offer may contain too much.
Or too little.
The buying moment may be wrong.
The channel may be wrong.
The problem may be real, but the proposed solution may not fit it.
The purpose of testing is not to deliver a simple verdict as quickly as possible.
It is to reveal what the result allows us to conclude—and what it does not.
A failed test may suggest:
stop;
revise;
test a different audience;
change the proposition;
alter the price;
improve the evidence;
or investigate the problem more deeply.
The important discipline is not to protect the original idea from the result.
Let the evidence change the proposition.
Ask What Else Could Explain the Result
Every test contains alternative explanations.
An event may attract few registrations because demand is weak.
Or because the date is unsuitable.
A product may receive few purchases because the price is too high.
Or because the audience does not trust an unfamiliar seller.
A community consultation may receive limited participation because people are uninterested.
Or because the organisation chose inaccessible times, locations or formats.
A landing page may convert poorly because the proposition is weak.
Or because the page is confusing.
Good analysis therefore asks:
What else could explain what happened?
This does not mean dismissing inconvenient evidence.
It means interpreting evidence carefully enough to avoid drawing a stronger conclusion than the test can support.
Keep an Assumption Register
As an idea develops, assumptions can easily disappear into documents, conversations and decisions.
A simple assumption register makes them visible.
For each important assumption, record:
The assumption
What needs to be true?
Why it matters
Which part of the proposition depends upon it?
Current evidence
What supports or contradicts it?
Confidence
How certain are we, and why?
Consequence
What happens if it proves false?
Priority
Does it require attention now?
Test
What could generate better evidence?
Threshold
What result would materially change confidence?
Decision
Proceed, revise, pause or stop?
This creates a living map of uncertainty.
It also helps teams avoid repeatedly debating the same question without producing new evidence.
Record Contrary Evidence
Evidence systems often become collections of reasons the project should proceed.
That is dangerous.
An assumption register should preserve information that weakens the proposition as carefully as information that strengthens it.
If three potential partners decline, record it.
If customers misunderstand the offer, record it.
If the proposed price generates resistance, record it.
If a community challenges the organisation’s framing of the problem, record it.
If the prototype fails under realistic conditions, record it.
Contrary evidence is not an inconvenience to remove from the presentation.
It may be the most decision-useful information available.
The Evidence Ledger
Within the Cultural Intelligence Simulation Lab, this discipline can be developed into an Evidence Ledger.
The ledger separates statements into categories such as:
Known
Supported by sufficiently credible evidence for the current decision.
Inferred
A reasonable interpretation derived from evidence, but not directly established by it.
Assumed
Required or expected to be true without sufficient evidence.
Testable
Capable of being investigated through a proportionate research activity or experiment.
Unknown
Not currently understood and not necessarily easy to determine.
This classification matters because organisations often present all five categories with the same level of confidence.
A market statistic becomes a prediction.
An interview becomes proof of demand.
A successful prototype becomes proof of scalability.
A positive conversation becomes a partnership.
An aspiration becomes a forecast.
The ledger interrupts that movement.
It asks every important claim to reveal what kind of statement it actually is.
Assumptions Change as the Idea Changes
An assumption is not resolved permanently simply because it was tested once.
Markets change.
Costs change.
Audiences change.
Competitors respond.
Technology develops.
Cultural meanings shift.
A pilot with ten customers may provide useful early evidence.
It does not guarantee that the proposition will behave identically with one thousand.
A community consultation conducted two years ago may not represent current circumstances.
A technology that worked during a controlled test may behave differently after integration into a real organisation.
Testing should therefore be proportionate to the stage of commitment.
The larger the commitment, the stronger the evidence the decision may require.
Match the Evidence to the Size of the Commitment
This principle is central to responsible decision-making.
You do not need perfect certainty before conducting a £200 experiment.
You may need substantially stronger evidence before committing £200,000.
The question is not:
Do we know everything?
It is:
Do we know enough to justify this next level of commitment?
This creates a staged process.
A small amount of evidence may justify a conversation.
A stronger pattern may justify a prototype.
A successful prototype may justify a pilot.
A credible pilot may justify wider investment.
Each commitment earns the right to consider the next.
That is very different from treating early enthusiasm as permission to build the entire proposition.
Assumption Debt
Organisations can accumulate something resembling debt.
Assumption debt occurs when unresolved beliefs become embedded in increasingly important decisions.
A founder assumes the audience.
Marketing is built for that audience.
A website is written for that audience.
A product is designed for that audience.
Advertising is purchased to reach that audience.
Revenue forecasts depend upon that audience.
If the original assumption is wrong, every downstream decision may need to be revisited.
The later the assumption is challenged, the more expensive the correction becomes.
This is why early clarity has disproportionate value.
Testing a foundational assumption may feel like slowing the project down.
In reality, it may prevent months of accelerating in the wrong direction.
When Is There Enough Evidence?
There is no universal point at which uncertainty disappears.
The appropriate standard depends upon:
the scale of the commitment;
the reversibility of the decision;
the cost of being wrong;
the people affected;
the time available;
the quality of the evidence;
and the consequences of delay.
A reversible decision with limited consequences may justify acting with incomplete information.
A decision affecting vulnerable communities, substantial investment, legal responsibility or long-term reputation may require considerably stronger scrutiny.
The objective is not maximum research.
It is sufficient understanding for a proportionate decision.
Research can also become avoidance.
At some point, the next useful evidence may only emerge through action.
The discipline is knowing whether the current uncertainty calls for more investigation, a controlled test or a decision.
The Cultural Intelligence Simulation Lab and Assumption Mapping
The Cultural Intelligence Simulation Lab is designed to make these hidden structures visible before commitment becomes difficult to reverse.
During an Idea Simulation Review, the proposition can be translated into a map of the assumptions beneath it.
The process asks:
What must be true?
What evidence currently supports it?
What evidence challenges it?
Which assumptions are financial?
Which are operational?
Which concern audiences, partners or technology?
Which are cultural?
Which involve trust, representation, participation, access or power?
Which assumptions are carrying the greatest weight?
What could test them?
What would happen if they proved false?
This allows the review to concentrate on decision-critical uncertainty rather than producing a generic list of risks.
An idea does not need every assumption resolved before it can move forward.
It does need clarity about which unresolved assumptions it is carrying into the next decision.
That is the difference between accepting risk consciously and simply failing to see it.
Cultural Intelligence Studio Perspective
Cultural Intelligence Studio treats assumptions as material for investigation, not evidence of failure.
A strong idea can contain uncertainty.
A responsible decision makes that uncertainty visible.
This is particularly important when propositions involve communities, culture, technology or emerging markets.
The people developing the idea may be highly informed.
They may also be close enough to the proposition that certain beliefs have become difficult to see.
The purpose of structured challenge is therefore not to undermine imagination.
It is to create distance from attachment.
To ask:
What are we treating as true?
Why do we believe it?
Who sees the situation differently?
What evidence would change our minds?
What could happen if we are wrong?
And what is the smallest responsible action capable of teaching us more?
These questions strengthen an idea because they allow reality to participate in its development.
The Question Nobody Asked
Projects rarely collapse because somebody openly announced:
“We are going to base this entire proposition on an unsupported assumption.”
The process is usually quieter.
A belief enters a conversation.
The belief enters a document.
The document enters a plan.
The plan enters a budget.
The budget creates commitments.
Repetition gradually gives the belief the appearance of fact.
By the time reality challenges it, the organisation may have invested too much to respond easily.
That is why the most valuable question in a project may be the one that interrupts the chain:
How do we know?
Not to create paralysis.
Not to demand certainty where certainty is impossible.
Not to extinguish ambition.
But to distinguish confidence from evidence.
To find the load-bearing assumptions.
To test the dangerous ones early.
To match the strength of the evidence to the size of the commitment.
And to preserve enough flexibility for the idea to change when the evidence requires it.
Every promising idea begins with belief.
But belief should open an investigation, not end one.
The assumption beneath the idea may be the part nobody has questioned.
It may also be the place where the most important learning begins.
Before you commit to the visible plan, examine the invisible theory holding it together.
Ask what must be true.
Ask what you actually know.
Ask what would prove you wrong.
Then design the smallest credible test capable of changing the decision.
Because the strength of an idea is not demonstrated by protecting its assumptions.
It is demonstrated by discovering which of them can survive contact with reality.
Continue Exploring
Previous Article
Your Idea Is Not the Evidence
How to Separate What You Know, What You Believe and What You Still Need to Discover
Article One introduced the distinction between evidence, inference, assumption, testable hypotheses and unknowns—and explained how the Cultural Intelligence Simulation Lab uses those distinctions to support better decisions.
Next Article
The Smallest Test Before the Biggest Commitment
How to Design Experiments That Produce Evidence Before Resources Become Difficult to Recover
Article Three will examine how to translate important assumptions into proportionate real-world tests, define useful success thresholds, avoid misleading validation and determine what the results genuinely allow you to conclude.
Before You Commit
This article forms part of the Cultural Intelligence Studio series exploring:
evidence;
assumptions;
constructive challenge;
cultural intelligence;
scenario thinking;
experimentation;
risk;
human judgement;
and better decision-making under uncertainty.
Cultural Intelligence Studio
Human judgement. Cultural intelligence. Strategic clarity. AI-enabled capability.