Factors to consider for share question

Question

I have a question about frameworks: I came across a couple of questions where the instructions are saying "the share of usage" of something has declined. For example here bikes:

Austin is one of the biggest cities in Texas, USA. Austin is famous for its bike-friendly road

system. A bicycle can easily access every corner and spot in the city. However, in the last

decade, the share of bicycles in total city traffic has gone down. The city government has asked

McKinsey to find out the root-cause and to fix it.

What are some factors you would look at in order to help the city of Austin?

However, I am majorly struggling with finding buckets that are MECE in such share cases. Do you have any tips on how to approach such questions?

Answer 1

The key insight: share is always relative. If bike share declined, it means bikes got less attractive, alternatives got more attractive, or both. That anchors your entire structure.

The cleanest MECE split is to look at both sides of the equation:

1. Did cycling decline in absolute terms?

Supply side: fewer bikes, worse infrastructure, higher cost, safety concerns
Demand side: demographic shifts, more remote work, cultural change

2. Did alternatives grow faster? Even if cycling stayed flat, share falls if other modes explode, e.g. ride-sharing getting cheaper, car ownership becoming more affordable...

The principle for all share questions: always decompose into your own performance vs. alternative performance, then go one level deeper into supply and demand for each. That keeps your buckets MECE.

One last tip: always clarify the trend before diving in. Is the absolute number of cyclists flat or actually falling? That single data point tells you whether you have a relative problem, an absolute problem, or both.