What is top down thinking?
It’s a highly efficient method often used in consulting to make clear and evidence a hypothesis or recommendation.
For instance, AAB CEO, Graeme Allan would like to supply pastries to all AAB employees to fuel the ongoing ‘CommonHealth Games’ initiative using a newly acquired hypothetical subsidiary, AAB Bakery. Graeme doesn’t want to undo the good work of the games so he thinks the bakery should just send each AAB employee two pastries every week, or in total 10,192 lemon tarts, croissants, or cinnamon rolls over 14 weeks but the bakery has never met such demand before.
Graeme is a busy man (sampling the products we think) so he has asked the consulting team to work out whether this is possible and what it would take. Using a top down thinking approach to the problem, we need to start by considering what is the core question we need to answer? In this case: is AAB Bakery capable of supplying each employee with two pastries each week?
We then break it down into components – can we produce enough pastries and can we get them to each employee. We can then break each component down further into subcomponents which test the validity of each component. For example, if we have enough ovens and enough bakers and enough ingredients, we can say with a good level of certainty that we can actually produce the required number of pastries. In the same manner, if both distribution and production are operating effectively then we can say that it is likely that AAB Bakery can indeed achieve its goal of supplying each employee with two pastries every week.
Using a key part of the top down thinking approach, the central recommendation is then stated in the format of Situation, Complication, Question (where the question is implied), Answer - “To fuel each employee during the Commonhealth Games with 2 cakes per week, it is possible to use AAB Bakery’s current production and distribution infrastructure with only additional ingredients.” [We’ve glossed over a few key questions about the costs of make vs buy and alternative approaches to production and distribution for the simplicity of this example.]
Should we use top down thinking or should we just stick with what we know?
Top down thinking is structured
The core premise of top down thinking is asking what needs to be true for this overall statement to be true. In comparison, if we work from a bottom up perspective, where we gather information first and then try to reach a conclusion from it, we often miss key aspects because we get caught up in the complexity of the situation. In fact we might even end up answering the wrong question entirely.
If we are talking about top down thinking, we can’t forget MECE (mutually exclusive and collectively exhaustive). In theory, it means that two groups can’t overlap and the groups must cover all possibilities, in reality it means lots of thinking about the right words to neatly divide up the situation.
If we don’t use MECE then our facts will be linked to multiple sub categories, making it difficult to evaluate each factor seperatley.
Taking this structured approach provides lots of little independent questions, which is far easier to divide up when working in a team and avoids work being repeated.
It allows us to focus on what’s important
Through creation of a top down MECE structure, we should for each main point have 3-5 supporting points. Because these should cover all possibilities, we can start ranking what is the most important without having to worry about totally forgetting a critical element of the puzzle. From that, we can start refining what is the most important and what plays more of a supporting role and focus our effort on those high impact areas.
To break each problem down into its categories and subcategories, we have to make some simplifications. The primary one is that each subquestion only has one link to its parent question only. In reality, we know this is not true. Given the fairly small size of AAB Bakery, the bakers and delivery drivers may in fact be the same people meaning we can’t split them directly into production and distribution. Alternatively, by having better route planning software, we may need less delivery vans and so there is a dependency between the two questions where we can’t answer one without answering the other. Of course, this makes it difficult to evaluate each component individually.
Whilst top down thinking provides a quick way to structure a problem, we can’t solely rely on it. In doing so we may end up over simplifying to the point where the problem we are solving looks nothing like reality. It’s a tool that can sit in a suite alongside other models like systems thinking. By having lots of perspectives, we can see closer to the truth.