You are on page 1of 11

Running head: [SHORTENED TITLE UP TO 50 CHARACTERS] 1

Demographic segmentation Psychographic segmentation Behavioral segmentation

Outcome Marketing mix.


[SHORTENED TITLE UP TO 50 CHARACTERS] 2
[SHORTENED TITLE UP TO 50 CHARACTERS] 3

Abstract

1. Demographic segmentation: Refers to the classification of an audience based on

demographic factors such as age, race, gender, family size, income, or education. Demographic

segmentation is used to target specific demographic groups. You can't communicate effectively

with an audience if you don't know anything about them. To effectively manage your advertising

budget, you must also take a personalised, targeted approach. Demographic segmentation allows

you to target your marketing strategies more precisely. It clarifies your vision, provides more

guidance for future advertising plans, and assists you in optimising your resources, time, and

budget. If you have 85% of your clients between the ages of 20 and 35.

Examples:

The most fundamental factor of all is age, but it also has the most bearing because consumer

tastes vary as people become older. A lot of marketing initiatives target particular age group.

psycographic segmentation: A market segmentation strategy called psychographic

segmentation divides consumers into groups based on psychological traits that affect their

purchasing patterns, which are influenced by their lifestyles and preferences. It is heavily

influenced by "how" individuals think.

Examples:

Psychographic market segmentation, along with demographic, regional, and behavioural

segmentation, is one of the most efficient segmentation approaches. Examples of these

characteristics include social standing, daily routines, eating habits, and viewpoints on particular

subjects.

3.Behavioral segmentation : 3. Using behavioural segmentation, it is possible to group clients

according to the way they make purchases.


[SHORTENED TITLE UP TO 50 CHARACTERS] 4

Examples:

1. Purchase habits

The segmentation of purchase behaviour looks at how customers act differently while making

choices. It aids organisations in understanding:

Hierarchial clustering

Data is organised into a clustering tree using the hierarchical clustering approach. First, each data

point is treated as an independent cluster in hierarchical clustering. The following actions are

then repeatedly carried out: Find the two clusters that are most connected to one another, and.

Combine the two groups that are most similar.

The process of "hierarchical clustering” requires arranging clusters in ascending order. For

instance, a hard drive’s data and directories are all arranged in a hierarchy. Divisive and

agglomerative hierarchical clustering are the two categories under which they fall.

A dendrogram An illustration of the hierarchy of relationships between objects is called a

dendrogram. As a result, it often comes from hierarchical clustering. A dendrogram is most

frequently used to decide how to cluster objects. The height at which any two elements are

connected to one another is the secret to understanding a dendrogram. The link between E and F

in the aforementioned image is the shortest, emphasising their greatest resemblance. They then

create customised marketing tactics using this data. An insurance company, for instance, may

utilise clustering to find groups of drivers with high

K- Means clustering
[SHORTENED TITLE UP TO 50 CHARACTERS] 5

The iterative Kmeans algorithm aims to divide the dataset into K unique, non-overlapping

clusters that each contain a single data point. It seeks to maximise the comparability of the intra-

cluster data points while maintaining the clusters as diverse (far) apart as possible.

The following is how the kmeans algorithm functions:

The kmeans algorithm works as shown below:

The K-th cluster count.

Set up the centroids by selecting K data points at random, replacing none of them, and then

rearranging the dataset.

Continue iterating until the centroids stay the same. i.e., the clustering of the data points does not

change.

Implementation

To illustrate just a few concepts, we’ll use a simplistic implementation of kmeans in this case.

Then, we will use a better Sklearn implementation to carry out a number of tasks for us.

The standard for linalg import is to import numpy from numpy as np.

Putting the Kmeans algorithm, class Kmeans, into practise.

With demographic segmentation, you may more accurately focus your marketing plan. .

We must choose the right number of clusters for our dataset when using clustering techniques

like K-Means clustering. As a result, the data will be precisely and efficiently segregated. The

number of clusters, or “k,” should be selected at a suitable amount in order to provide adequate

cluster granularity and to maintain a reasonable balance between cluster accuracy and

compressibility.

Let's consider two cases:

Case 1: Use a single cluster for the entire dataset.


[SHORTENED TITLE UP TO 50 CHARACTERS] 6

Case 2: Consider every data point to be a cluster.

The form and scalability of the data distribution can be used to help us with this difficult issue.

1. Elbow Method: It’s approach is on the finding that the sum of within-cluster variance for

each cluster can be decreased as the number of clusters increases. Having more clusters makes it

possible to extract smaller, more homogeneous groups of data elements. For each value of k > 0,

the curve's initial turning point gives the appropriate value. Let's use Python to implement the

elbow method.

Bringing the libraries in first

Importing the libraries in Python 3

import matplotlib as pd and pandas.

import KMeans

Second one

The Mall Customer dataset, which is available at this URL, was used.

2. Silhouette Score: The quality of clusters generated by clustering algorithms like K-Means is

evaluated using the silhouette score, which assesses how well data points are clustered with other

data points that are comparable to each other. This method can be used to get the optimal value

of "k".
[SHORTENED TITLE UP TO 50 CHARACTERS] 7
[SHORTENED TITLE UP TO 50 CHARACTERS] 8
[SHORTENED TITLE UP TO 50 CHARACTERS] 9
[SHORTENED TITLE UP TO 50 CHARACTERS] 10
[SHORTENED TITLE UP TO 50 CHARACTERS] 11

You might also like