Secret to Clear Thinking as a Data Analyst? MECE Principle!
As a data analyst, there can be a lot of clutter in the head regarding a specific problem you’re working on.
And, it can really effect you’re performance because of a feeling of overwhelm since there are a multitude of ways of going about that problem & devising a solution.
Thinking Clearly for yourself is one part of the equation.
However, equally important is communicating what you’ve built to non-technical users like Executives.
I’ve faced both of these issues for which the solution has been the MECE Principle.
MECE, pronounced “me-see”, stands for “mutually exclusive, collectively exhaustive”.
It was developed by Barbara Minto at McKinsey & Company. During her time there, she found that people were not able to communicate clearly in their writing.
Her synthesis was that the problem was the thinking, not the language.
Thus, the question became, how do we think clearly?
This is where MECE comes into play.
It is about separating a set of items into subsets that are mutually exclusive(ME) and collectively exhaustive(CE).
This leads to structure in your thinking while also being complete & avoiding any confusion regarding the problem at hand.
How MECE works?
Mutually Exclusive
In order to reduce complexity, its important to avoid overlaps. In our thinking, it is exactly this overlap that causes clutter.
Thus, mutually exclusiveness forces us to look at each option separately.
When you have a problem to solve, you need to make a list of issues.
Thus the question to ask here is each one a separate and distinct issue?
Collectively Exhaustive
Now, for each of the mutually exclusive points, we need to collectively exhaust the major alternatives or categories.
Thus, the question here becomes have you thought of all the factors relating to the issue?
Though we can come up with multiple ME & CE points, a general rule of thumb is 3 major Mutually Exclusive points and then 3 major Collectively Exhaustive points (if not, between 2 & 5).
Only the most important points with most leverage should be considered.
Impact
The MECE principle has been so effective at McKinsey that every new hire who joins the company is made aware on this principle and all documentation is based on this principle.
Some benefits of using MECE are:
- Structure your thinking with maximum clarity & completeness, minimum confusion.
- Effectively Communicate your Solution
- Break down a problem into clear chunks along with their analysis
- Collective Exhaustiveness ensures you do not miss any key part of the analysis
- High Quality ideas from Brainstorming which sees the bigger picture
Applications as a Data Analyst
- Mapping out factors for data inaccuracy
Data Accuracy Validation is among the most important parts of the work of a data analyst before deploying a dashboard or presenting insights from a certain data set.
Data inaccuracy while dealing with a complex metric can be a real headache for which MECE can be useful.
For instance, recently while deploying a dashboard for one of the business verticals of where I work, the records of 2 clients were missing. This was causing a 0.2% inaccuracy in the dataset.
In such instances, the query has been made with complexity yet the error is very small.
This makes the debugging process even harder.
Thus, mapping the entire possible issues & their causes out, the task becomes more manageable.
You can simply tick off the factors that are accurate and investigate on the missing ones of what went wrong.
2. Communicating to Non-Technical Users
In the same case, showing the image above to the functional manager, he was able to understand the scenario easier rather than me talking in a million different directions.
The same idea can be used in presentations as well to simplify and only focus on the most important content.
3. Brainstorming Data-Driven Strategies
Another key part of the job as a data analyst for me right now is creating data driven strategies that helps the different departments reach their goals faster and thus enabling the company’s goals to progress faster and be met.
But that’s way too much vague, isn’t it?
The process always starts with a specific ‘How might We’ statement, after considering the problem statement of the department.
In the example above while dealing with the Customer Experience team, the customer retention numbers weren’t looking pretty.
So the question to the analytics question became on how data could be used in increasing customer retention?
3 Key strategies were mapped out and for each, the data points of relevance was mapped out.
One more layer of collectively exhaustiveness can be mapped in this case to specifically define the data points.
This can help to quickly test on whether these data points can be directly used in customer retention strategies and thus lead to meeting the CX team’s goal of higher retention.
Overall, as a data analyst since we are dwelling in an ocean of data & information, it is important to not boil the entire ocean for some salt.
This is where frameworks like the MECE Principle are the secret sauce.
