HackerEarth SIGMATHON 1.

0 Hackathon

DATA ANALYSIS AND INSIGHTS PRESENTATION

PRESENTED BY
TEAM
PyFlow ++
 INTRODUCTION

This is about the analysis made on the dataset obtained from DataCo
supply chain
Supply chain analytics brings data driven intelligence to the business,
reducing the overall cost to serve and to improve the service levels.
Descriptive analysis is carried out by using the data visualisation
techniques like charts and graph to uncover patterns and generate
insights.
On analysing this dataset, we have understood and acknowledged about
consumer behaviour and marketing strategies.
 TYPES OF ANALYSIS
The dataset includes a total of 180520 rows of customer purchase history,
details, demographics spanned to 53 columns.
There were many features like total sales, profit ratio, customer segment,
transaction details, customer region, order status, delivery status which are vital
to determine the sales analysis.
Based on these features, we have made several type of analysis like
 Monthly analysis
 Market analysis
 Financial analysis
 Delivery analysis
 Customer segment analysis
 Department analysis
 Shipping analysis
 MONTHLY ANALYSIS
4000000

3500000  The sales orders received is

3000000
highest during the month of July.
 Total sales done and profit
2500000
incurred is almost same in the
2000000 month of August comparative to
1500000
others.
 Reasons for decline in sales
1000000
maybe due to competitive
500000 market.
0  Price can also have a big impact
y y h il y e ly st r r r r
nuar uar arc Apr Ma Jun
r Ju ugu mbe tobe mbe mbe on sales. High price can maximize
Ja Feb M A e Oc ve ce
pt short term profit while low price
Se No De
products can maximize long term
profit.
 MARKET
ANALYSIS
 Order items placed is highest in
Europe(29.6%) Europe(demand high).
 This is because the shipment facilities are
excellent in Europe and customer shave
LATAM(27.9%) purchased more number of technological
products like camera and laptop which have
high budget
Pacific asia(22.5%)  Order items placed is least in Africa , analyzing
the cause can increase the profit per order
 The order item total is low in Africa because
USCA(13.8%)
the customers have purchased more number
of low budget goods rather than high budget
Africa(6.2% goods.
)  Hence the total amount gained by the
company is low.
 FINANCIAL ANALYSIS
36,786,000
 There is a decrease in order item profit
36,785,000
ratio from January 2015 to January
2018.
36,784,000
 The profit in the year 2016 is low when
36,783,000 compared to the other 2 years. This may
be due to low sales earnings and low
36,782,000 merchandise cost.
 The data is not complete for the year
36,781,000
2018. Hence the graph shows a steep
line, but the profit for the starting
36,780,000 month of 2018 is also high as 2017.
36,779,000
36,778,000
2015 2016 2017 2018
GEO MAP
 This is a world chart showing the
location of orders in a world map
 The number of orders is high in the
region of Europe followed by
America and Africa.
 The orders in these regions are
high because of good delivery
facilities and easy access to get the
products ordered through online.
 It indicates that there is a large
customer base and global presence
due to the availability of products
according to customer needs.
 MARKET ANALYSIS
Using Country and City

Top 10 Countries in Sales Top 10 Cities in Sales

25000 2500

20000 2000

15000 1500

10000 1000

5000 500

0 0
ia lia il a os cia a lia o do n s ua ila ty ty ia o go a
an as in di Ita xic ni do ele Ci Ci ph isc alp
m stra Br Ch nid
r an In © U n ng nag an o r k el nc m
in ig
e Au
U F Ã Lo sA
a M ic Yo d a
Do uc
Al os M in
o M ex ila Fr eg
d Re Lo M w h n to T
Es
ta Ne P Sa
Sa
n

• The TOP 10 Countries are mixed with both Developed and Developing Countries, with
majority of market available in developing countries.

• However, most of the cities are technically developed best cities in the world, mostly
concentrated towards American and European continents.
 CUSTOMER ORDER STATUS

 Orders received among all three

segments of customer is almost
Home Office complete, indicating customers
SUSPECTED_FRAUD have increased their purchasing
PROCESSING frequency.
PENDING_PAYMENT
 This also indicates prompt
Corporate
PENDING
execution of order.
PAYMENT_REVIEW

ON_HOLD
 A large number of orders are in the
state of pending payment mode,
the reason may be due to the
COMPLETE

CLOSED
issues in transaction.
CANCELED
Consumer
 Therefore we need to resolve the
technical glitches so that the order
0 2000000 4000000 6000000 8000000
status become complete .
ORDERS( in Millions)
 REGIONAL ANALYSIS
7000000  Europe being the top region in terms of higher
orders placed, risk of late delivery is equally
6000000 high.
5000000  Latin America is also among the highest late
Advance delivery region.
4000000 shipping
 This is due to the fact that more number of
Late de-
TOTAL SALES

3000000 livery orders are placed from these regions and the
shipment may get delayed due to these large
Shipping
2000000 on time
number of orders.
 Even though the company had large sales from
1000000
these regions, the company failed to deliver
0
the products on time for many products.
Africa Europe LATAM Pacific USCA
Asia
 As far as customer lifetime value is
concerned ,the number of orders may
ORDER REGION decreases because of late deliveries.
 DEPARTMENT ANALYSIS

Total
3.4% 0.1% 2.8% Apparel
0.3%  The products sold under the Fan Shop
Book Shop
department has high demand among
21.7%
Discs Shop the customers with 46.5 %
12.5% Fan Shop
 Among the 11 departments, Book
0.0%
0.6%
Fitness shop, Disc shop, Technology, Health &
10.9%
Footwear Beauty, Pet shop seem to have very
Golf
low sales.
Health and  The sales can be increased by
1.1% Beauty
providing incentives such as offers,
Outdoors discounts, free shipping etc.
46.5% Pet Shop

Technology
 SHIPPING ANALYSIS

9000000
 This is the analysis based on the
8000000
shipping mode.
7000000
 The standard class has more
6000000 number of advanced shipping as
Advance
5000000 shipping well as late deliveries compared to
other classes.
TOTAL SALES

Late de-
4000000 livery
Shipping  The late delivery can be minimized
3000000 canceled
Shipping by implementing extreme return
2000000 on time policies. This can be applied for
1000000 same day shipping as well.
0
First Class Same Day Second Standard
Class Class
SHIPPING MODE
 SHIPPING ANALYSIS
(CUSTOMER SEGMENT)

12000000
 The occurrence of late delivery is
higher among all the segments of
10000000
customer.
 The reason for more number of late
TOTAL SALES(MILLIONS)

8000000
Advance deliveries are because of the
shipping
6000000 Late de-
geographical features and mode of
livery shipment.
Shipping
4000000 canceled  The order city and the delivery city is of
Shipping long distance for many customer orders
on time
2000000
and hence the shipment is late.
 If this continues over a period ,the
0 customer retention rate of the business
Consumer Corporate Home Office will be affected , hence steps need to
CUSTOMER SEGMENT
be taken to meet customer satisfaction.
THANK YOU