Using Value Chain Analysis to Discover Customers - amazonia.fiocruz.br

Using Value Chain Analysis to Discover Customers Video

Variants: Using Value Chain Analysis to Discover Customers

The top domain in which advanced analytics and data science is used is in understanding customers. And it is far more superior compared to any other domain — supply chain, IoT, finance etc.

Tale of the Mighty Duct

And the reason is obvious — the success of your business depends upon how well you understand your customers. All other things will fall in placeonce you have started to understand your customer in an effective way. Customer analytics is a very wide area. As well as every industry will analyse customers in a different way. However the top 3 customer analytics across all industries are the following. Now let us see what are these analytics as well as what are the data science techniques are involved. Customer segmentation, sometimes also called clustering, it allows you to make group of similar customers. So why is this useful? Say you want to Chqin a marketing campaign and want to send email to your customers. So either you can send a common email to all your customers.

What Is the Flywheel?

Both these methods would be ineffective. In the first case, a common message to all your customers may not be suitable as each customer has his or her specific needs. In the second case, you will be spending too much time in creating tailor made message for everyone.

So in order to solve this problem, customer segmentation is useful. It will group similar customer into segments.

Navigation menu

So now you can send an email message per segment. As customers in a segment are similar, you can tailor the message for each segment. Also as you save lot of timeas you are creating message for each segment and not for each customer. Here are some techniques which you can use for customer segmentation. Let us assume that you have a dataset of customer with different attributes as shown here.

One way of making segments is based some columns which you already article source in mind. For example, you can make segments based on Job and Marital status, and see which segment has most of the customers and which segment has least number of customers.

A heatmap visualisation is a good way to find out segments visually, as shown below. So as you can see that heatmap is a good way to segment the customers. Here you can see that most of the customers fall into segment of married and blue-collar. So you can tailor your marketing message to specific segments. However in many situations, you do not have specific columns in mind to make segments.

So you can try to make segments on multiple columns. In this wayyou let data tell you which segments are possible rather than any pre-determined segments. Clustering is a way to make segments using multiple columns. Clustering groups similar data records based on multiple columns into clusters or segments. Shown below is result of clustering shown as a scatter plot.]