Every manager makes decisions, but not every decision rests on solid ground. Some leaders trust their gut, others demand reports, and a few quietly run small tests before committing real money. The quality of a choice often depends less on intelligence and more on what it is built upon. In retail and broader business management, four foundations repeatedly shape how good decisions get made: past experience, experimentation, research and analysis, and decision tree analysis. Understanding when to lean on each one is what separates a confident manager from a lucky one.
Table of Contents
- Why every business decision needs a foundation
- Leveraging past experience
- When experience becomes a trap
- Turning experience into a reliable guide
- The role of experimentation
- The cost of finding out
- Research and analysis for rational decisions
- Navigating uncertainty with decision tree analysis
- How a decision tree works
- Reading the branches to find the best path
- Choosing the right foundation for the moment
Why every business decision needs a foundation
A decision is simply the act of selecting one course of action from several available alternatives. The hard part is not listing the options. It is justifying why one option deserves your capital, your people, and your time over the others. This is where a basis of decision making matters. The basis is the reasoning method you use to evaluate alternatives and pick the strongest one.
Management thinkers Harold Koontz and Heinz Weihrich grouped the classic approaches to selecting an alternative into experience, experimentation, and research and analysis. Decision tree analysis extends this thinking into situations clouded by uncertainty. Together these four form a practical toolkit. A store manager deciding stock levels for a festival season and a CEO deciding whether to enter a new city are using different combinations of the same underlying logic.
Leveraging past experience
Relying on past experience is the most natural basis for decision making, and it is especially common among executives who have been promoted from within an organisation. A manager who has handled a decade of seasonal demand swings carries a mental library of what worked and what failed. That accumulated judgement is genuinely valuable. In fact, one of the oldest schools of management thought, the empirical or case approach, studies management precisely by analysing real experience and past cases.
When experience becomes a trap
Experience turns dangerous the moment it is applied blindly. A pricing strategy that worked beautifully in a small town market may collapse in a metro where customer behaviour, competition, and rent structures are entirely different. The situation has changed, but the manager assumes the old answer still fits. This is the core risk: treating the past as a template rather than a reference. Past success does not prove that the same action will succeed again under new conditions.
Turning experience into a reliable guide
Experience becomes trustworthy only when it is analysed, not just remembered. The useful question is not “What did I do last time?” but “Why exactly did it succeed or fail, and are those reasons present now?” When a manager understands the underlying causes behind a past outcome, experience stops being a habit and becomes a diagnostic tool. Modern research on decisions from experience confirms that how people represent and recall past events strongly shapes the choices they make, which is why careful reflection matters more than raw memory.
The role of experimentation
Experimentation means testing an alternative on a small scale to see what actually happens before committing fully. It mirrors scientific enquiry: form a hypothesis, run a controlled trial, observe the result, and then decide. Instead of arguing about which option is better in a meeting room, the business lets reality answer the question.
A familiar example in retail is testing advertisement copies. Before a national launch, a company might run two or three different versions of an ad in selected regional markets, measure which one drives the most footfall or sales, and then scale up only the winning version. The same logic powers product placement trials, store layout pilots, and limited menu rollouts. Companies such as eBay have used experiments to cut tens of millions in advertising spend after discovering that certain ad budgets were not actually changing customer behaviour.
The cost of finding out
The strength of experimentation is also its weakness. Running real tests consumes capital, personnel, and equipment. Setting up a pilot store, printing variant packaging, or staffing a trial campaign all cost money before any decision is even made. For small businesses, an elaborate experiment may simply be unaffordable. There is also a cultural challenge. Research on experimentation in organisations notes that leaders must be genuinely willing to let data overturn their intuition, otherwise the test becomes theatre. If the boss has already decided the outcome, the experiment wastes resources without informing anything. Experimentation works best when the stakes of a wrong full-scale decision are high enough to justify the price of finding out first.
Research and analysis for rational decisions
For major decisions, research and analysis is the most widely used and most effective basis. Rather than testing in the field like an experiment, this approach solves the problem on paper first. It involves the systematic collection of relevant data, careful analysis, and interpretation in the specific context of the problem at hand. The manager builds a model of the situation, identifies the key variables, and studies their relationships before acting.
Consider a retail chain deciding where to open its next outlet. A research and analysis approach would gather data on local population density, income levels, footfall patterns, competitor locations, rental costs, and supply chain distance. Each factor is weighed, and the analysis points toward the location with the strongest combination of advantages. Because the decision rests on evidence and reasoning rather than instinct, decisions made this way tend to be the most rational. This is also the shift many modern firms describe as moving from opinion-led choices toward genuinely data-driven decision making, where the numbers, not the loudest voice in the room, guide the outcome.
The method does have limits. Good analysis depends on good data, and collecting reliable information takes time and money. Analysis paralysis is a real danger when managers keep studying a problem instead of deciding. Still, for high-value, complex choices, the discipline of researching before acting consistently produces better outcomes than guesswork.
Navigating uncertainty with decision tree analysis
The first three foundations work well when the future is reasonably predictable. But many business decisions involve genuine uncertainty, where the outcome depends on events the manager cannot control, such as whether a market grows, whether a competitor reacts, or whether monsoon-dependent demand holds up. Decision tree analysis is a diagrammatic technique built specifically for these situations. It is a structured way to confront risk explicitly rather than pretend it does not exist.
How a decision tree works
A decision tree maps out choices and uncertainties as a tree-like diagram that reads from left to right. It uses three building blocks. A decision node represents a choice the manager controls, such as “launch the new product” versus “do not launch.” A chance node represents an uncertain future event the manager does not control, such as “high demand” versus “low demand,” and each possible event is assigned a probability. A terminal node sits at the end of every path and carries the payoff, usually a monetary value, of that particular sequence of events.
Reading the branches to find the best path
The power of the tree comes from combining probability with value. To evaluate each option, you multiply the payoff of every possible outcome by its probability and add the results together. This gives the expected value of each decision path. The probabilities at any chance node must add up to one, since something must happen. The manager then works backward through the tree, comparing the expected values, and selects the path that offers the highest desirable return for the risk involved.
This makes the technique especially useful for staged decisions, where one choice leads to new uncertainties and further choices. The expected value calculation turns a fuzzy, anxious judgement into a clear comparison of numbers. It is widely applied in project management and investment analysis, where managers must justify large commitments under real risk. A decision tree does not remove uncertainty, but it does make the trade-offs visible, so everyone can see exactly what bet is being placed and what each outcome is worth.
Choosing the right foundation for the moment
These four foundations are not rivals. They are tools for different situations. Experience is fast and cheap but risky when conditions change. Experimentation gives real-world proof but costs money and effort. Research and analysis offers rigour for major decisions but demands good data and time. Decision tree analysis brings clarity to choices dominated by uncertainty and probability. A skilled manager rarely relies on just one. The strongest decisions often blend analysed experience with hard data, a small pilot to confirm assumptions, and a tree to weigh the risk before the final commitment.
What do you think? Looking back at an important decision in your own work or studies, which of these four foundations was it really built on, and would the outcome have changed if you had used a different one? When the future is genuinely uncertain, do you trust a structured tool like a decision tree more than your own experienced judgement, or less?
References
- https://en.wikiquote.org/wiki/Harold_Koontz
- https://pmc.ncbi.nlm.nih.gov/articles/PMC10790535/
- https://www.hbs.edu/faculty/Pages/item.aspx?num=57300
- https://arxiv.org/pdf/2102.05310
- https://www.smartgecko.academy/en/decision-trees-choosing-under-uncertainty/
- https://www.open.edu/openlearn/money-business/decision-trees-and-dealing-uncertainty/content-section-4.1
- https://www.pmi.org/learning/library/analyze-project-decision-trees-complexity-5027
Leave a Reply