MECE Principle

No Rep, No Missing: How to unpack a problem into actionable chunks?

MECE principle stands for “Mutually Exclusive, Collectively Exhaustive.” It is a technique pioneered by McKinsey to structurally breakdown a client problem. It’s an effective way to approach a complex problem, and develop a more layered response. It aims to unpack a complicated issue into actable small chucks of subgroups. By doing so,

It reduces the scope of a complex problem into how to effectively unpack that problem.

There are two scenarios where MECE principle is often used:

1. To divide a targeted group or a market so that the standard of partition ensures “no repeating, no missing.”
2. To reason the deciding factors and steps in a problem to ensure “No repeating, no missing” on each of the hierarchical levels.

First scenario:

Ex: Estimate Airbnb’s market size in China.

At first glance, this thought experiment is not easy to get started because it hard to measure how many people in China use Airbnb as substitutes for hotels.

Conventional approach:

Generally people tend to analyze market share by first identifying the the target groups of Airbnb. And the market size:

total revenue=sum(number of people in each group x the average money spent by each group)

For example the target market of Airbnb could be divided into:

• 18–24 college students
• 20–30 young females who are more receptive to Western values.
• 20–35 white collars with overseas background
• 30–40 family group

The problem with this approach is that it’s prone to missing and repetition. Apparently, the first three group overlap: a young female could be both a college student as well as with overseas background. And without using MECE, this approach will overestimate the market size of Airbnb.

MECE Approach:

A better approach is to use conventional demographical metrics like age, sex, geographical location, income level to partition the market because it effectively avoids double counting the groups in partitioning the market. However, which demographical metrics to use also depends on the market and the business model of the company.

For example:

With a hypothesis-driven mindset, I estimate that the target group is young people, who are reliant on smart phones, and live in first-tier or second -tier cities, who likely expose to new things. The reason I select metrics of age, the permeability rate of smart phones, and geographical location are predicated by my understanding of Airbnb market and product. And I estimate the market share from here.

Second scenario:

Ex:How to improve profits of of Baidu.

Common approach:

It’s easy to fall into the trap of solely focusing on one standout business of Baidu like adversing, and spend inordinate amount of time on how Baidu can improve its advertising revenues.

MECE Approach:

However, let’s not forget other sources of revenue for Baidu. The blindspot of this approach is that people get the question without first reason a good way to unpack it. With MECE principle, it’s better to first to break its sources of revenue down into

revenue from existing business. (Assets in place)

potential sources of revenue in the future. (Growth assets)

For Baidu, its main existing businesses are advertising, e-commerce transaction, and video streaming. And its potential future businesses are Baidu Cloud, and AI.

By first establishing an holistic framework, it help me to develop a more layer response from here.

• I may go out of my way to analyze ways to improve the existing revenue streams of Baidu and brainstorm cost control measures by analyzing its operating leverage. (fixed+ variable)
• Zoom in into the revenues of its ads business by analyze the sales volumes, and price.
• Analyze the advertising performance, and relative portion of different modules be it medical, education.

Conclusion:

The first step in approaching a problem, especially a complex one from a unfamiliar domain is think the way to reason the way to divide up the the problem into actionable chunks. And I have to convince myself why I choose that particular way.

Key Takeaways:

1. Market size question: divide up the market using demographical metrics, in accordance with the traits of the market of product.
2. Improve profits question: categorize the business into current business, and new business in the future, and develop a layered response from there.