Annexure:1

LOVELY PROFESSIONAL UNIVERSITY

Course Code: MKT307 Course Title: Visual Analytics

Course Instructor: Dr. Ankur Kukreti

Academic Task No.: 1 Academic Task Title:

Date of Allotment: 05 February, 2025 Date of submission: 28 February, 2025

Student’s Roll no: RQ6E85A31 Student’s Reg. no: 12213379

Declaration:

I declare that this Assignment is my individual work. I have not copied it from any other
student’s work or from any other source except where due acknowledgement is made
explicitly in the text, nor has any part been written for me by any other person.

Evaluator’s Comments (For Instructor’s use only)

General Observations Suggestions for Improvement Best part of assignment

This dashboard provides a high-level overview of sales performance, customer

Total LTV (lifetime value): 505,902 - this is arguably the most important indicator. It
represents the total revenue a company expects to generate from its entire customer base
through its relationships with them - a high LTV indicates a healthy, valuable customer base
and effective customer retention. An LTV in excess of 500,000 with only 200 customers is
very large, meaning that the average LTV per customer is high. (Note: This LTV calculation
is highly dependent on the model used. It is important to understand how this LTV is derived
in order to fully trust its interpretation). The first insights from the top indicators: moderate
number of customers but very high total LTV suggests that the value per customer is high,
although the average purchase value is low. This implies strong customer loyalty, repeat
purchases or high value products and services. Given the value of the existing customer base,
focusing on slightly higher average purchases can significantly increase total revenue and
LTV.

2. Gender distribution (pie chart):

male: 51%, female: 47%, other: 2% - gender distribution is almost equal, slightly skewed
towards male customers.

Insight: the product/service seems to appeal to both genders almost equally. Marketing
campaigns can be adjusted slightly so that key messages are gender neutral or, if necessary,
emphasize aspects that resonate more with each group. The 'other' category is insignificant
and does not seem to require a specific targeting strategy.
3. Regional distribution (bar chart):

south: highest (approx. 61), north: second highest (approx. 56), east: medium (approx. 47),
west: lowest (approx. 36): customer density is highest in the South and North. Marketing and
sales efforts should be concentrated in these regions. In contrast, the Western region has the
lowest number of customers. This may need to be investigated: Is market penetration low in
the West? Are there regional differences in product attractiveness, competition and marketing
effectiveness?

4. Campaign engagement scores (bar chart):

online advertising: highest score (approx. 4000); social media: second highest score (approx.
3000); email: medium (approx. 2000); TV advertising: lowest (approx. 1000): digital
marketing channels significantly outperform traditional TV advertising in terms of
engagement. Online advertising and social media are the clear leaders. This strongly suggests
that marketing budgets and focus are shifting towards digital strategies. TV advertising is
underperforming and may need to be re-evaluated and funds reallocated; email marketing is
moderately effective and could be further optimized;

5. Distribution of income levels (bar chart):

medium: highest (approx. 67); high: medium: highest (approx. 67); high: medium (approx.
53); low: Lowest (approx. 12): the main client group is middle-income, followed by the high-
income group. The low-income group is quite small. Product positioning and pricing
strategies seem to resonate most with middle-income and high-income customers. If the
strategic goal is to penetrate lower income groups, product content and marketing techniques
may need to be adjusted.
6. Purchase behavior - total average purchase value (stacked bar chart):

This graph appears to show the 'total average purchase value' broken down by region.
Looking at the values: South: quite high (about 7066.04): medium (about 6467.42) North:
medium (about 6826.01) West: lowest (about 4828.34): The South region has the highest
number of customers (regional scatter plot) but is also dominant in terms of total average
purchase amount. This is a powerful combination: high number of customers and high
contribution in terms of purchase value. The Western region consistently performs the worst
in terms of both number of customers and purchase value. This reinforces the point made
earlier in the study about the underperformance of the Western region.

7 Average purchase (bar chart):

west: highest average purchase (approx. 16 000) north: second highest average purchase
(approx. 14 000) north: second highest (approx. 14 000) east: medium (approx. 12 000)
south: lowest (approx. 10 000) Insight This chart shows a nuanced perspective. Although the
Southern region has the highest total purchase value (due to higher volumes), the average
purchase value per customer is actually lowest in the Southern region and highest in the
Western region. This may indicate different customer segments and purchasing patterns in
different regions. Customers in the West may make fewer purchases and have higher
individual transaction values, while customers in the South may make more but fewer
purchases. Discrepancies and in-depth analysis Discrepancies between 'regional distribution'
(most customers in the South) and 'average purchase value' (lowest average purchase value in
the South and highest in the West) are key points for further analysis. Possible explanations
Product mix: different products and services are sold in different regions, which affects the
average transaction value. Arguably, higher priced products are more popular in the West.
Customer segmentation: customer demographics and needs vary widely across regions,
leading to differences in purchasing behavior. Marketing effectiveness: Do marketing
campaigns in the West attract different types of customers who spend more per transaction,
albeit in smaller numbers;

8. Correlations in sales (correlation matrix tables):

by year and gender: although these tables appear to show correlations, they are largely filled
with '1s' and '11s'. This is highly unusual and most likely indicative of a data presentation
problem or a simplified composite data set. Real sales data should have correlations between
different years and genders, as well as autocorrelations of '1'. A true correlation analysis will
reveal patterns such as: sales increasing or decreasing from year to year (e.g. correlation
between 2020 sales and 2021 sales) Differences in sales trends by gender (correlations
between genders). Current interpretation (based on the '1' value - assuming it is intended to
indicate something): despite the quirkiness of the data presentation, if we were to extract the
intended message, the '1' value may be trying to imply a positive correlation within each
category, but there is no comparative correlation between categories (year or gender).
However, this section cannot currently be interpreted in a meaningful analytical way due to
data presentation issues. It will require clarification and data modification to be useful.
9. Filters - income level, region and gender:

these filters are powerful tools for interactive analysis. By applying these filters, data can be
dynamically segmented and drilled down to see how certain groups of customers behave. For
example, filtering by 'high income' and 'Western region': analyze purchase behavior, average
purchase amounts and marketing channel activity, with a special focus on high-income
customers in the Western region.

Filtering by 'women' and 'social media': see social media marketing engagement scores
specifically for female customers. These filters are crucial for uncovering detailed insights
and understanding nuances in different customer segments.

strategic recommendations Leveraging high LTVs: high LTVs are a key strength. Focus on
customer retention strategies and initiatives that further increase customer lifetime value
(loyalty programs, personalized experiences, etc.). Strengthen digital marketing: it is clear
that online advertising and social media are the most effective marketing channels. Direct
marketing budgets to these digital channels and separate them from unprofitable

TV advertising; further optimize email marketing; increase the use of social media to
promote the brand's products and services; increase the use of social media to promote the
brand's products and services. Capitalize on the South and North: these regions lead in the
number of customers, while the South is also the revenue leader. Ensure
a strong sales and marketing presence in these regions. Study the Western
region: The Western region is an anomaly with the lowest number of
customers but the highest average purchase value. Investigate why this is
the case. Are there untapped opportunities? Can marketing strategies be
adjusted to increase the number of customers while maintaining high
average transaction values?
Address the average purchase value: LTV is high, but average purchase value is relatively
low. Evaluate strategies to increase basket size, such as product bundling, upselling, cross-
selling and minimum order value promotions. Clarify correlated data: The correlation table
should be reviewed and the data corrected to provide meaningful insights. Once this section
is corrected, significant year-over-year sales trends and gender-based purchasing patterns
may emerge. Use filters for detailed analysis: encourage regular use of dashboard filters to
explore specific customer segments and identify niche opportunities and challenges.