Business Problem Solving
Simple analysis: Rules of thumb, summary statistics and heuristics
Dr. Stephen Hills
Learning objective
To undertake simple analysis that can answer research questions about the issues identified in your problem disaggregation and prioritization to inform problem solutions.
Simple analysis
The seven-steps process
How do you define a problem in a precise way to meet the decision maker’s needs?
How do you disaggregate the issues and develop hypotheses to be explored?
How do you prioritize what to do and what not to do?
How do you develop a workplan and assign analytical tasks?
How do you decide on the fact gathering and analysis to resolve the issues, while avoiding cognitive biases?
How do you go about synthesizing the findings to highlight insights?
How do you communicate them in a compelling way?
Step 5: Conduct critical analyses
Start with heuristics – short cuts or rules of thumb – to get an order of magnitude understanding of each component and assess priorities
Understand where there is a need for more work and for more complex techniques
Make frequent use of one-day answers
Simple analysis: Rules of thumb, summary statistics and heuristics
Good problem solvers have a toolkit for fact gathering and analysis.
Starting with rules of thumb, summary statistics and heuristics to understand the direction and magnitudes of relationships.
We can structure and resolve many analytic issues with rules of thumb, summary statistics and straightforward heuristics.
Rules of thumb are shortcuts in analysis that we can quickly apply to answer a question.
Summary statistics are calculations that provide a summary of data, e.g. Mean average.
Heuristics are any approach to problem solving or self-discovery that employs a practical method that is not guaranteed to be optimal, perfect, or rational, but is nevertheless sufficient for reaching an immediate, short-term goal or approximation.
All three help to size the different elements of the problem to determine the critical and efficient path in further analysis.
Rules of thumb
Rules of thumb: 80:20 rule and rule of 72
80:20 rule: Developed by the Italian economist Vilfredo Pareto, this rule of thumb describes the common phenomenon that 80% of outcomes come from 20% of causes. In other words, the 80:20 ration is a very common market structure, so it can be applied if your problem requires you to estimate a market structure. For example:
20% of consumers account for 80% of sales.
20% of the population account for 80% of healthcare costs.
Rule of 72: This rule is a shortcut for estimating compound growth or compounding effects, which can be applied if your problem requires you to estimate how quickly wealth might build, how quickly an enterprise might scale or how quickly a population might grow. The rule of 72 allows you to estimate how long it takes for an amount to double given its growth rate by dividing 72 by the rate of growth. For example:
If the growth rate is 5% an amount will double in 14.4 years: 72/5 = 14.4 years.
If the growth rate is 15% an amount will double in 4.8 years: 72/15 = 4.8 years.
If an amount doubles every 2 years, the growth rate is 36%: 72/2 years = 36% growth rate.
Rules of thumb: S-curve
The S-curve is a rule of thumb, which shows a common pattern of sales with a new product or a new market. It can be applied if your problem requires you to estimate the adoption rate for a new innovation.
This common pattern of the S-curve is that adoption starts slow and then picks up before starting to plateau.
Summary statistics
Simple risk and pricing calculations: Break-even point and marginal analysis
Break-even point: The level of sales where revenue covers cash costs, which can be applied to determine risk of a particular endeavour, e.g., how confident are we that we can cover our costs? Requires knowledge of marginal and fixed costs.
Break-even point = Fixed costs / unit price – unit variable costs
The unit price should be known and the fixed costs and variable costs can be quickly calculated, but the complexity comes in how these variables can change as you scale a business, e.g. step-fixed costs where to double volume involves significant investment.
Marginal analysis: Examining the cost or benefit of the next unit. Can be applied to inform problem solving around decisions on producing more, consuming more or investing more in an environment of scarce resources.
Where there are fixed costs of machinery and plant, marginal costs (i.e., the cost of producing one more unit) can fall very quickly, favouring more production until more machinery is required.
Example: Imagine that the total fixed costs (e.g., mortgage, staff, taxes) of a 200 room hotel are $10k per night or $50 per room. A last minute special offer of $30 per night for unsold rooms is below the average room cost of $50, but the marginal cost to sell one more room is very small (i.e., electricity, water, towels).
Simple probability calculations: Bayesian thinking and expected value
Bayesian thinking: Based upon conditional probability, which is the probability of an event given another event which also has a probability (i.e., a prior probability). Can be applied to predict probability on the basis of a given situation.
Example: The probability of raining is higher if it is cloudy than if it is sunny.
Expected value: The value of an outcome multiplied by its probability of occurring. It can be applied to set priorities and reach conclusions on whether to take a bet in an uncertain environment.
Example: The Australian research organization CSIRO decided on whether to defend its WiFi intellectual property in court using expected value in reverse. They calculated that they would win $100 million and their legal costs would be $10 million. This reflects a 10% probability of occurring, but CSIRO thought the probability of success was much higher, so it was a gamble worth taking.
Heuristics
Reasoning by analogy
Reasoning by analogy: When you have seen a particular problem structure and solution before that you think may apply to your current problem.
Example: Precedent in law is where conclusions in one court case are applied to another.
Distribution of outcomes
On the basis historical data about similar projects, you can estimate the distribution of outcomes, so to have contingency plans for budget cost overruns and possible delays.
Example: Companies planning large projects often add contingencies of 10% or more for cost overruns. However, research into comparable projects may show that similar projects as yours often need more than 10% contingencies.
Question-based problem solving
A deeper analysis after you have roughed out the scale and direction of your problem levers using rules of thumb and heuristics.
Paint a picture of the problem by asking questions, such as who, what, where, when, how and why.
A powerful root-cause tool to quickly focus problem solving.
Example: Should I have a knew arthroscopy?
A literature review established the key questions that informed the solution to the problem.
Root cause analysis
The technique of asking 5 Whys to get to the bottom of a problem was developed at the Toyota Motor Corporation.
The core idea is to dig deeper than the superficial or proximate causes of problems to uncover the source or root causes.
A root cause is one that when addressed removes the later problem.
The 5 Whys
A diagnosis of market-share loss, looking at superficial causes until the deeper source of the problem is found.
Forces us to push beyond the local or contributory causes of problems towards root causes by asking ‘why’ until there is no further ‘why’.
By forcing greater and greater specificity the ultimate source of customer loss is found in inadequate training of phone customer service agents.
Conclusions
Conclusions
Start all analytic work with summary statistics and heuristics that help you see the size and shape of your problem levers.
Rules of thumb can serve as useful short cuts.
Simple question-based analysis grounded in the literature can lead you to a solution.
Root cause and 5-Ways can help you identify fundamental causes of problems that then lead to a solutuion.
Workshop: Challenger Disaster Research