Quadrant Analysis : One of simple method to set up prioritize and action for business strategy.
Sometimes we have many possibilities to determine to decide something for business strategy. For example, when we have two metrics, and want to determine which unit we want to prioritize, like :
- Which product that we want to boost is more important, if we have data on (1) revenue, and (2) number of sales.
- Which users will be given the promo, if we have (1) Retention, and (2) number of booking
- Which hotels can be boosted to be used as highlights on the homepage, if we have data on (1) number of sales, and (2) number of users visit.
This time, we will discuss one simple method that can be used to decide which one we can do first for our company’s business decisions.
Required data :
- Unit granularity, could be users, location, products, etc.
- 1st Metrics
- 2nd Metrics
From the table above, we can try to create a quadrant graph, where :
- X = Metrics 1
- Y = Metrics 2
With the final graph as below:
Some may ask, how do we determine the horizontal and vertical lines above?
We can use several methods, such as:
- Look at the value of the center tendency of the data, be it the median or the mean.
- Determine something with business needs, such as company target
From the graph above, it is found that, for each quadrant, we have :
- Q1 : A% users
- Q2 : B% users
- Q3 : C% users
- Q4 : D% users.
From the data above, we need to remember that our final goal is :
- How to change users in Q2 and Q4 to move to Q1 group, and
- How to change users in Q3 to move to Q2 or Q4 and then move to Q1 group.
If in terms of %, the Q1 group is already high in terms of volume of users, then there are not many things that need to be a concern because we can say that there are already quite a number of users who are our best users. But if the opposite happens, we need to do a deep analysis so that we can determine the right action for each user.
Apart from simply upgrade the users group status, we also need to pay attention to the costs needed. We can’t expect 90%++ of our users to be the best users (in the Q1 group), but at least when we do monitoring (example: daily monitoring, we see that there is a positive movement of users group.
