Business leaders are often admired for making difficult decisions under pressure.

The story is familiar. A crisis emerges, time is limited and an apparently exceptional leader steps forward with the right answer. Their decisiveness saves the project, reassures the team or protects the organisation from failure.

That ability has value.

But the repeated need for heroic decision-making can reveal something less impressive: the organisation may not have prepared properly.

If every deadline depends on someone working late, the problem is not simply time management. If every client project requires the founder to remember what happens next, the problem is not merely workload. If one missing file, unavailable colleague or misunderstood instruction can stop the work completely, the organisation has developed a fragile system.

Strong strategy does not wait for pressure and then hope that someone performs brilliantly.

It asks a different question:

“What can we decide, document and protect now so that a difficult day does not become a damaging one?”

The Visible Mistake Is Rarely the Whole Cause

When something goes wrong, attention usually settles on the final visible event. Someone forgot to send the document. A deadline was missed. The wrong version was published. A booking was not confirmed. A client received information that had not been approved.

The individual closest to that event may be blamed. But the mistake may have started much earlier.

Perhaps responsibilities were never clearly assigned. The deadline existed in one person’s inbox. Several versions of the document had almost identical names. Nobody was required to complete a final check. One person held information that the rest of the team could not access.

The final error was real, but it was only the last link in a longer chain.

This is close to the principle represented by James Reason’s “Swiss cheese” model of failure. Protective systems are rarely perfect. Each may contain weaknesses, but serious failure becomes more likely when weaknesses in several layers align and allow a problem to pass through every safeguard.

This way of thinking has influenced safety-critical fields because it shifts attention from individual blame towards the conditions surrounding an error. The question becomes not only, “Who made the mistake?” but also, “Why was the system unable to catch it?”

Businesses, cultural organisations and creative projects can benefit from asking the same question.

A Protocol Is a Decision Made in Advance

A protocol is a defined way of handling a recurring activity. It does not have to be complicated. It might be a checklist, a project template, an approval sequence or a short set of “if this happens, do this” instructions.

A useful client-enquiry protocol might state:

1. Record the enquiry in the central system.

2. Confirm receipt within an agreed period.

3. Check whether the request fits the organisation’s services.

4. Clarify the client’s objectives, budget and timescale.

5. Identify any safeguarding, accessibility or cultural considerations.

6. Prepare the appropriate response.

7. Obtain approval before making commitments.

8. Record the outcome and next action.

The purpose is not to remove thought from the process. It is to prevent the same basic questions from consuming attention every time the activity occurs. The protocol protects mental energy for decisions that genuinely require judgement.

Research into “implementation intentions” supports this general principle. Plans that specify when, where and how someone will act—often expressed as “if this situation occurs, then I will take this action”—can help people translate intentions into behaviour.

Knowing what matters is not always enough. Progress becomes more dependable when intentions are converted into actions that can be recognised and followed.

Reliability Does Not Mean Rigidity

The word “protocol” can sound restrictive, particularly within creative and culturally responsive work. Artists, entrepreneurs and community organisations may worry that formal processes will make their work bureaucratic or prevent them from responding naturally to people and circumstances.

That can happen when protocols are poorly designed.

A useful protocol does not attempt to predetermine every decision. It distinguishes between what should remain consistent and what must remain open to judgement.

A community-engagement process, for example, should consistently protect informed participation, accessibility, appropriate communication and the respectful treatment of knowledge. But it should not assume that every community wants to participate in the same way.

A marketing protocol may require evidence, approval and accessibility checks. It should not force every audience into the same language, imagery or cultural reference points.

A strong system creates dependable foundations without pretending that different people, places and situations are interchangeable.

The goal is not standardisation for its own sake. The goal is reliable quality with enough space for context, conversation and human intelligence.

Protocols Also Contain Power

Every protocol reflects decisions about what matters. It determines who is consulted, who can approve an action, what counts as acceptable evidence and whose needs must be considered before work proceeds.

That means protocols are not culturally neutral.

A funding-development process that measures only income may overlook trust, participation and social value. A recruitment process based heavily on informal networks may reproduce existing patterns of exclusion. A consultation process may collect community perspectives without giving participants any influence over the final decision.

The system may operate exactly as designed while still producing an unfair or culturally damaging result.

Organisations should therefore examine their protocols through questions such as:

Who helped design this process?

Who benefits from the way it currently works?

Who may find it difficult to participate?

Where is human discretion allowed?

Who has the authority to make exceptions?

How can someone question or challenge a decision?

What cultural knowledge does the process depend upon?

What happens when the protocol produces an inappropriate outcome?

A process can be efficient and still be wrong. Strategic maturity requires both operational reliability and the willingness to examine the values built into the system.

Build More Than One Layer of Protection

A protocol helps people perform an expected process correctly. A safeguard helps detect a problem. Redundancy provides another route when the original method fails. These are related, but they do different work.

Consider an important client report:

The protocol defines how the report is researched, written, checked and approved.

The safeguard requires a final factual and accessibility review.

The redundancy keeps separate, recoverable copies of the working files.

The contingency plan explains what happens if the principal writer becomes unavailable.

Together, these layers make the work less dependent on one person, one device or one perfect day.

The principle has been demonstrated in far more consequential settings. The World Health Organization developed its Surgical Safety Checklist to help teams complete essential safety checks consistently. Its use has been associated with meaningful reductions in preventable complications.

A business checklist is not equivalent to a surgical-safety intervention. The stakes and contexts are different. But the transferable lesson is valuable: expertise does not make people immune to omissions, particularly when work is complex, familiar or pressured.

Memory should support a system. It should not have to be the system.

Margin Is a Strategic Resource

Many organisations plan around the best possible version of events. Every person will be available. Every task will take the expected amount of time. Technology will function. Clients will respond promptly. No revision will be required.

This creates plans that appear efficient because nothing seems to be wasted. It also creates fragility.

Time margins allow for delays, checking and reflection. Financial reserves create room to respond to unexpected costs. Capacity margins prevent every new opportunity from overwhelming the existing operation. Relationship margins ensure that trust is not dependent on one hurried conversation.

Not every part of a small business can have extensive backup capacity. Redundancy carries costs, and unnecessary procedures can create more work than they prevent. The answer is proportionality.

The greatest protection should surround the points where failure would cause the greatest harm:

Client commitments

Payments and financial records

Legal and contractual obligations

Personal and sensitive information

Safeguarding

Public statements

Valuable creative work and intellectual property

Funding deadlines

Accessibility requirements

Relationships involving community trust

The purpose is not to eliminate all risk. That is impossible. The purpose is to prevent avoidable weaknesses from gathering around the work that matters most.

Do Not Automate a Process You Do Not Understand

AI and automation can help organisations apply protocols consistently. They can organise information, generate reminders, prepare standard documents, identify missing fields and move work between stages. Used carefully, they can reduce repetitive administration and make small teams more capable.

But automation also makes it possible to repeat a poor decision more quickly and at a larger scale.

Before automating a process, an organisation should understand:

What the process is intended to achieve

Which decisions can be standardised

Which decisions require human judgement

What information the system uses

Where bias or exclusion could enter

Who checks the output

What triggers escalation

How the process can be stopped or corrected

Who remains accountable for the outcome

AI should strengthen a well-considered process. It should not conceal the absence of one.

Create a Minimum Dependable Standard

A protocol should not attempt to describe the perfect performance of a task. It should establish the minimum standard below which the work should not fall.

For each recurring activity, identify:

The trigger: What starts the process?

The owner: Who is responsible for moving it forward?

The essential steps: What must happen every time?

The decision points: Where is judgement required?

The evidence: What information must support the decision?

The approval: Who can authorise the outcome?

The record: What must be documented?

The safeguard: What check could catch an error?

The contingency: What happens if the expected route fails?

The review: When will the protocol be reconsidered?

Start with one process that is frequent, important and currently inconsistent. Test the protocol in real work before building a large operating manual.

Notice where people become confused. Remove steps that do not protect quality. Add missing decisions. Ask whether the process works for people with different responsibilities, access requirements and levels of knowledge.

A protocol becomes valuable through use, not through its appearance in a document.

Strategy Is What Still Works on a Difficult Day

A fragile organisation can perform impressively when its founder is fully available, the team has enough time and nothing unexpected happens. That is not the strongest test.

The stronger test is what happens when someone is tired, technology fails, a deadline changes or several demands arrive at once.

Does the organisation still know what matters? Can important information be found? Are responsibilities clear? Will someone notice before an error reaches a client or community? Can the work continue without depending on one person remembering everything?

Strategic intelligence is not demonstrated only through brilliant choices at critical moments. It can be seen in the quieter work of designing conditions in which fewer critical moments are allowed to develop.

Make the important decisions while there is still time to think.

Document what should remain consistent. Protect what would be difficult to replace. Build margin around what cannot be allowed to fail. Preserve human judgement for the situations that genuinely require it.

The strongest system is not one in which nothing ever goes wrong.

It is one that can recognise difficulty, absorb disruption and continue without losing its purpose, its standards or the trust of the people it exists to serve.