Asher's Blog

Many hard problems remain hard partly because everybody who sees them is waiting for somebody else to authorise the first attempt. People wait for a title, a budget, an invitation, a perfectly formed plan or an expert who will confirm that their idea is worth pursuing. The problem remains untouched, not necessarily because nobody is capable of solving it, but because nobody feels sufficiently qualified to begin.

I have found that hard problems often become easier when you stop asking for permission and start running responsible experiments. The first step does not need to solve the entire problem. It only needs to reveal something that was previously unknown. Once you act, the problem stops being an abstract object in your mind and begins to give you information. You discover what works, what fails, what customers actually care about and which assumptions were never true in the first place.

This is how my first business began. A friend of a friend had returned from the United Kingdom and needed to recruit people in Nigeria, but he was finding the process difficult. I was not running a recruitment company at the time, and nobody had appointed me to solve the problem. I simply accepted the challenge and started looking for a way to help.

He also needed some social-media work done, so we used PeoplePerHour to find people who could execute it. One request led to another, and what began as an attempt to solve a practical problem gradually became my journey into recruitment. I did not begin with a grand announcement that I was building one of Nigeria’s leading job platforms. I began by helping one person find the people he needed.

At the time, I was still employed, and I told my boss what I was doing. Looking back, I would probably describe that period as idea validation, but I did not use that language then. I was simply doing the work and paying attention to the response. Sometimes people follow the correct entrepreneurial process before they know what the process is called. They see a problem, try a solution, observe what happens and make another decision based on what they have learned.

That experience taught me that action can reveal a business before the founder fully understands that a business is being formed.

Permission often disguises a desire for certainty

When people say they are waiting for permission, they are not always waiting for someone to say yes. Sometimes they are waiting to feel completely certain that the idea will work. They want to know the size of the market, the correct product, the exact price, the perfect marketing channel and the final outcome before they begin.

Unfortunately, the answers to many important questions are located on the other side of action. Research can narrow the possibilities, but it cannot remove every uncertainty. You may interview customers and study competitors, but you will still learn something different when a real customer has to pay, change an existing habit or trust your product with an important responsibility.

This is why entrepreneurship can feel uncomfortable to people who are accustomed to environments where a correct answer already exists. Hard problems are not examinations with answers hidden at the back of a textbook. They are closer to investigations. You form a hypothesis, test it, examine the result and improve your understanding.

Waiting to feel ready can therefore become a respectable form of avoidance. You attend another event, read another book, revise the deck again and ask ten more people whether the idea is sensible. All of those activities can be useful, but they become dangerous when they replace the experiment that would produce the most valuable information.

You do not need permission to conduct a lawful, ethical and contained test. You need clarity about what you are trying to learn, an honest understanding of the risk and enough discipline to observe the result without manipulating it to support what you already believe.

Not asking for permission is not the same as being reckless

The phrase “do not ask for permission” can be misunderstood, particularly in industries where people’s money, data, health or safety are involved. It does not mean ignoring regulation, breaking agreements, spending resources you do not control or exposing customers to risks they did not accept. It also does not mean disregarding a leader whose approval is genuinely required within an organisation.

There are permissions you must obtain. If an activity requires a licence, informed consent, board approval or access to another person’s property, proceeding without it is not courage; it may simply be misconduct. Responsible founders understand the difference between unnecessary validation and legitimate authority.

The permission you should stop seeking is the informal permission to take yourself seriously. You do not need a famous investor to confirm that a customer’s problem is real. You do not need a large team before testing whether people want a service. You do not need everybody in your industry to understand an idea before you begin learning how to make it work. Most importantly, you do not need to know the complete answer before earning the right to discover the next part of it.

The best early experiments reduce risk rather than create it. They begin with a narrow group of users, a manual process, a prototype, a limited geography or a service that allows the founder to observe the problem closely. The point is not to behave as though consequences do not matter. The point is to design action that produces knowledge without making an irresponsible promise.

Experimentation turns a large problem into evidence

Hard problems often feel impossible because we consider them as a single object. “How do we fix marketing?” is too broad to produce a useful answer. So is “How do we solve unemployment?”, “How do we improve payroll?” or “How do we make this company grow?” Each question contains several smaller problems that behave differently and require different interventions.

Experimentation makes the problem more specific. Instead of asking how to fix marketing, you can ask which customer segment responds most strongly to a particular message, which channel produces qualified interest, what happens between a lead arriving and a sale closing, and how much the company can spend to acquire a customer without destroying the economics of the product.

Each experiment should answer a question. If you change several things at once without defining what you are learning, activity may increase while understanding remains poor. You may generate more leads without knowing which action produced them, or spend more money without discovering whether the apparent growth is repeatable.

One of the difficult problems I have had to work through concerned marketing. I needed more than activity; I needed a model that would explain how our effort contributed to the company’s overall objective. I began by making the objective clear, because a model cannot be intelligent if it is optimising for an ambiguous destination.

Once the objective was clear, I broke it into the things that would have to become true in order to achieve it. I eventually identified roughly three layers of inputs feeding that objective. Each layer contained measurable activities and outcomes, which meant I could begin to see where progress was being created and where resources were being wasted.

The first version of the model was not final. I kept adjusting it as the results exposed weaknesses in my assumptions. I am still adjusting it because a useful operating model should learn from reality rather than demand that reality obey it. However, our marketing is now better because I have a clearer view of where to allocate funds, where not to allocate them and what effect an intervention is expected to have.

I did not arrive at that understanding by waiting for someone to hand me a complete marketing formula. I arrived at it by breaking the problem down, building a model, testing it and repeatedly improving it.

Begin with the objective, not with the activity

People sometimes start experimenting without deciding what success means. They launch campaigns, build features, hire employees and enter partnerships because the activities appear entrepreneurial. When the results are disappointing, they do not know whether the idea was wrong, the execution was weak or the measurement was irrelevant.

A hard problem becomes more manageable when the objective is specific enough to guide trade-offs. If the marketing objective is simply “awareness,” almost any visible activity can be defended. If the objective is to generate a particular number of qualified opportunities within an acceptable acquisition cost, the company can evaluate channels more honestly. An event that produces applause but no meaningful opportunities may be pleasant, but it cannot be confused with success against that objective.

Clarity also protects founders from copying tactics that were designed for another company’s circumstances. A strategy that works for a consumer product seeking millions of low-value users may be unsuitable for an enterprise product that needs a small number of high-trust relationships. Once you know the objective, you can ask whether a proposed action is likely to move the relevant measure rather than whether the action is fashionable.

This is one reason hard problems get easier after they are properly defined. The original problem may still require significant work, but the founder no longer has to consider every possible action. A clear objective eliminates attractive activities that do not contribute to the required outcome.

Break the problem down, then rebuild it from first principles

One of the most useful approaches to a hard problem is to separate it into its fundamental components and question the assumptions holding those components together. This is commonly described as first-principles thinking. Instead of beginning with how something is conventionally done, you begin with what must actually be true for the desired outcome to occur.

The widely told story of SpaceX illustrates the principle. When the existing cost of acquiring rockets appeared prohibitive, Elon Musk’s team did not merely accept the market price as an unchangeable fact. They examined the underlying materials, engineering requirements and manufacturing process, then asked what it would take to build differently. The useful part of the story is not that every founder should imitate SpaceX; it is that the current price or process may contain assumptions that can be separated and tested.

The same thinking applies to ordinary business problems. If recruiting is expensive, ask where the cost actually enters the process. Is it sourcing candidates, screening them, scheduling interviews, evaluating competence or losing strong candidates before an offer is made? If customers are not adopting a product, ask whether the problem is awareness, trust, pricing, onboarding, poor timing or a weak underlying need. “Customers are not buying” describes an outcome, but it does not yet identify a problem that can be solved.

Once the components are visible, you can decide where to experiment first. You may discover that one small constraint is weakening the entire system. Removing that constraint can make a problem that appeared enormous suddenly become tractable.

Research must end in a decision

I believe strongly in research. When you know that you do not know, the intelligent response is to study the problem rather than pretend to possess an answer. Read what has already been discovered, speak with people who have attempted similar work, examine available data and understand the history of the system you want to change.

However, research must eventually produce a decision. It should help you choose an experiment, define a risk or eliminate an unsuitable direction. If research continues indefinitely without changing what you do, it has become a shelter from responsibility.

The cycle should be simple: research enough to form a credible hypothesis, act within responsible boundaries, measure what happens and return to research with a better question. The result of one experiment improves the quality of the next investigation. In this way, action and thought strengthen one another.

This is particularly important now that artificial intelligence can make research faster. A founder can explore markets, compare approaches, analyse data and prototype ideas far more quickly than before. Yet easier access to information can also create the illusion of progress. Producing a long report is not the same as solving the problem. The advantage belongs to the person who converts improved information into better and faster experiments.

Your first solution may look unimpressive

People sometimes delay action because the first version of the solution does not resemble the company they hope to build. It may involve spreadsheets, telephone calls, manual introductions or a service performed personally by the founder. They worry that the work does not appear scalable or technologically sophisticated.

My first recruitment engagement did not need to look like a national platform. It needed to solve the immediate hiring problem in front of me. By doing that work, I gained access to the realities that would later inform the business: how employers described their needs, where candidates were difficult to find, how trust influenced recruitment and what small businesses struggled to manage after hiring.

The manual beginning was not separate from building the company. It was how I earned the information required to build it intelligently.

This is a useful principle for founders: do not confuse the form of the first experiment with the form of the eventual business. A manual service can teach you what to automate. A small customer can expose the need shared by a larger market. A simple landing page can reveal whether a message creates interest before a team spends months building the full product.

The first solution should be judged primarily by whether it produces useful evidence and genuine value, not by whether it looks impressive.

Action creates permission of its own

Something changes after you have solved a problem for one real person. You no longer need to argue entirely from theory because you possess evidence. The first customer gives you a case study, the first working process gives you a model, and the first failure gives you a more precise understanding of the risk.

Competence also attracts resources. It is easier to persuade partners, employees and investors when you can show what you attempted, what you learned and how the result changed your next decision. People may refuse to authorise an abstract ambition, but they often become interested when they see an emerging solution gaining momentum.

This does not mean that every experiment will become a business. Some attempts should end because they reveal weak demand, impossible economics or risks the founder should not carry. That is still valuable. An inexpensive experiment that disproves an idea can save years of commitment to the wrong direction.

The goal is therefore not to force every idea to succeed. It is to shorten the distance between a question and reliable evidence.

Hard problems demand humility because you must admit that you do not know the complete answer. They also demand courage because uncertainty cannot become an excuse for permanent observation. You have to begin somewhere, focus long enough to understand what the problem is teaching you, and change your approach when the evidence requires it.

Do the research, but make a decision. Define the objective, but do not wait for a flawless plan. Break the problem into its building blocks, test the part you can responsibly test and allow each result to improve the model. Ask for permission where law, ethics and genuine authority require it, but stop waiting for the world to certify that you are qualified to begin.

The problem may remain difficult, but it will no longer be mysterious. Once you start acting, it begins to show you how it can be solved.