2. HELLO!
I am Subhankar Basak
Business Analyst
Here I will show you how to analyse sales data.
You can find me @ Linkedin
3. About the Data Used in this Analysis Example
This data is downloaded from IBM Website to download the data click the below link.
https://www.ibm.com/communities/analytics/watson-analytics-blog/sales-products-sample-data/
TOOLS USED FOR DATA ANALYSIS
Microsoft Excel
R Studio
TOOLS FOR PREPARATION OF SLIDE
Microsoft Power Point.
Slide Templates from Slide Carnival
4. “Sales Data for Multinational Company
can be overwhelming at first, however
when we start simplyfying the data
we get Information.
6. Take a close look at the Data Sample
Retailer
country
Order
method
type Retailer type Product line Product type Product Year Quarter Revenue Quantity
Gross
margin
United States Fax Outdoors Shop Camping Equipment Cooking Gear
TrailChef Deluxe
Cook Set 2012Q1 2012 59628.66 489 0.34754797
United States Fax Outdoors Shop Camping Equipment Cooking Gear
TrailChef Double
Flame 2012Q1 2012 35950.32 252 0.4742745
United States Fax Outdoors Shop Camping Equipment Tents Star Dome 2012Q1 2012 89940.48 147 0.35277197
United States Fax Outdoors Shop Camping Equipment Tents Star Gazer 2 2012Q1 2012
165883.4
1 303 0.28293788
United States Fax Outdoors Shop Camping Equipment Sleeping Bags Hibernator Lite 2012Q1 2012 119822.2 1415 0.29145017
88,475 11 144 21
DATA ROWS DATA COLUMNS PRODUCTS COUNTRIES
7. Now this Raw Data is useless in terms of any decision making, however we as a Business
Analyst will work with the data to find out crucial information about the revenue share
through analysis.
At first we are creating a Pivot Table in
MS Excel,
Row Level – we have “Countries”
Column level - we have “Order Type” &
“Products”
Values – we have “Revenue” i.e. Sum of
Revenue.
Sum of
Revenue
Column
Labels
E-mail Fax Mail Sales visit
Specia
l
Telephone Web Grand Total
Row Labels
United States 613855.53 9630394.82 4346724.58 38737642.45
24077
95.14
41463959.53 553610588.5 650810960.5
Japan 13960470.39 821487.69 3795285.41 9940937.8
36456
5.5
7876782.26 244905969.6 281665498.6
China 4453684.39 757563.02 10664498.65 11359416.7 221588053.8 248823216.5
Canada 31862614.15 195549.5 2392936.69 9180026.32 12262855.8 190993682.4 246887664.9
France 253033.53 3651122.85 16698553.17 20570160.62 178350772 219523642.2
PREVIEW PIVOT TABLE
8. UNDERSTANDING PIVOT TABLE DATA
From the Above table we get;
1. Revenue Total of each Countries
2. The segregation of the Revenue from differnent Order Types.
3. Total Revenue of each Countries.
This is an important information, however we are not done yet we cannot show this
pivot table in our presentation to the Management. We need graphs.
Sum of Revenue Column Labels
E-mail Fax Mail Sales visit Special Telephone Web Grand Total
Row Labels
United States 613855.53 9630394.82 4346724.58 38737642.45 2407795.14 41463959.53 553610588.5 650810960.5
Japan 13960470.39 821487.69 3795285.41 9940937.8 364565.5 7876782.26 244905969.6 281665498.6
China 4453684.39 757563.02 10664498.65 11359416.7 221588053.8 248823216.5
Canada 31862614.15 195549.5 2392936.69 9180026.32 12262855.8 190993682.4 246887664.9
France 253033.53 3651122.85 16698553.17 20570160.62 178350772 219523642.2
United Kingdom 1955831.77 1279532.73 458316.87 10345992.69 521382.36 5487268.82 199174762.4 219223087.6
Germany 17197904.03 118768.81 5674772.23 8132036.03 5358461.21 164291721.1 200773663.4
Korea 3381359.76 7944762 570265.85 5643233.32 141034511.6 158574132.5
Singapore 4619801.07 761688.62 1422191.4 5341842.49 3432525.85 140683764.7 156261814.2
Italy 13470599.28 2041782.44 5114211.56 289214.47 3928010.4 129786955 154630773.1
Lets go over to the next slide
9. DATA PRESENTATION USING DONUT CHART
STATES TOTAL REVENUE
United
States 650810960.5
Japan 281665498.6
China 248823216.5
Canada 246887664.9
France 219523642.2
United
Kingdom 219223087.6
Germany 200773663.4
Korea 158574132.5
Singapore 156261814.2
Italy 154630773.1
Finland 152221983.4
Netherlands 148936395.6
Mexico 139369605
Spain 136763768.7
Austria 113366615.7
Australia 109299969.1
Brazil 109065671.4
Switzerland 100731875.5
Belgium 96958569.91
Sweden 80635525.26
Denmark 47898531.93
TOTAL REVENUE,
United States,
650810960.5, 17%
TOTAL REVENUE,
Japan, 281665498.6,
7%
TOTAL REVENUE,
China, 248823216.5,
7%
TOTAL REVENUE,
Canada,
246887664.9, 7%
TOTAL REVENUE,
France, 219523642.2,
6%
TOTAL REVENUE,
United Kingdom,
219223087.6, 6%
TOTAL REVENUE,
Germany,
200773663.4, 5%
TOTAL REVENUE,
Korea, 158574132.5,
4%
TOTAL REVENUE,
Singapore,
156261814.2, 4%
TOTAL REVENUE,
Italy, 154630773.1,
4%
TOTAL REVENUE,
Finland, 152221983.4,
4%
TOTAL REVENUE,
Netherlands,
148936395.6, 4%
TOTAL REVENUE,
Mexico, 139369605,
4%
TOTAL REVENUE,
Spain, 136763768.7,
4%
TOTAL REVENUE,
Austria, 113366615.7,
3%
TOTAL REVENUE,
Australia,
109299969.1, 3%
TOTAL REVENUE,
Brazil, 109065671.4,
3%
TOTAL REVENUE,
Switzerland,
100731875.5, 3%
TOTAL REVENUE,
Belgium,
96958569.91, 3%
TOTAL REVENUE,
Sweden,
80635525.26, 2%
TOTAL REVENUE,
Denmark,
47898531.93, 1%
United States
Japan
China
Canada
France
United Kingdom
Germany
Korea
Singapore
Italy
Finland
Netherlands
Mexico
Spain
Austria
Australia
Brazil
Switzerland
Belgium
Sweden
Denmark
10. DATA PRESENTATION USING DONUT CHART
TOTAL
REVENUE,
United States,
650810960.5,
17%
TOTAL
REVENUE,
Japan,
281665498.6, 7%
TOTAL
REVENUE,
China,
248823216.5, 7%
TOTAL
REVENUE,
Canada,
246887664.9, 7%TOTAL
REVENUE,
France,
219523642.2, 6%
TOTAL
REVENUE,
United Kingdom,
219223087.6, 6%
TOTAL
REVENUE,
Germany,
200773663.4, 5%
TOTAL
REVENUE,
Korea,
158574132.5, 4%
TOTAL
REVENUE,
Singapore,
156261814.2, 4%
TOTAL
REVENUE, Italy,
154630773.1, 4%
TOTAL
REVENUE,
Finland,
152221983.4, 4%
TOTAL
REVENUE,
Netherlands,
148936395.6, 4%
TOTAL
REVENUE,
Mexico,
139369605, 4%
TOTAL
REVENUE,
Spain,
136763768.7, 4%
TOTAL
REVENUE,
Austria,
113366615.7, 3%
TOTAL
REVENUE,
Australia,
109299969.1, 3%
TOTAL
REVENUE,
Brazil,
109065671.4, 3%
TOTAL
REVENUE,
Switzerland,
100731875.5, 3%
TOTAL
REVENUE,
Belgium,
96958569.91, 3%
TOTAL
REVENUE,
Sweden,
80635525.26, 2%
TOTAL
REVENUE,
Denmark,
47898531.93, 1%
United States
Japan
China
Canada
France
United Kingdom
Germany
Korea
Singapore
Italy
Finland
Netherlands
Mexico
Spain
Austria
Australia
Brazil
Switzerland
Belgium
Sweden
Denmark
Our Analysis from the Chart:
Top 4 Countries for most revenue shares are:
United States – 17%
Japan, China, Canada – 7%
Bottom 2 Countries for least revenue shares are:
Sweden – 2%
Denmark – 1%
Now we got information.
“ Decision Time going in depth into
Sweden & Denmark,”
Lets go over to the next slide
12. PIVOT TABLE
ARRANGEMENTS
---------------------------
We arranged the pivot table to
get the total revenue of each
Retailer type, our goal is to see
each retailers performance to
find scope for improvement.
FOCUSING ON SWEDEN & DENMARK WE ALALYSE
Sum of Revenue Column Labels
E-mail Fax Sales visit Telephone Web Grand Total
Row Labels
Sweden 3880829.46 4205772.95 8082854.82 64466068.03 80635525.26
Department Store 16315108.7 16315108.7
Direct Marketing 78042.73 770136.51 848179.24
Eyewear Store 5464185.46 5464185.46
Golf Shop 1503872.49 1503872.49
Outdoors Shop 3051204.78 1102964.04 7901582.71 13654610.64 25710362.17
Sports Store 829624.68 3102808.91 103229.38 23463321.26 27498984.23
Warehouse Store 3294832.97 3294832.97
Denmark 175517.78 3070840.1 44652174.05 47898531.93
Department Store 15684025.18 15684025.18
Direct Marketing 806544.82 541924.67 1348469.49
Equipment Rental Store 1303052.22 1303052.22
Eyewear Store 7502647.55 7502647.55
Golf Shop 2479368.37 2479368.37
Outdoors Shop 2074894.91 9589527.09 11664422
Sports Store 189400.37 4702061.82 4891462.19
Warehouse Store 175517.78 2849567.15 3025084.93
Grand Total 3880829.46 175517.78 7276613.05 8082854.82 109118242.1 128534057.2
13. PRESENTATION FOR SWEDEN & DENMARK RETAILER WISE REVENUE
Total, Sweden
Department Store,
16315108.7
Total, Sweden
Direct Marketing,
848179.24
Total, Sweden
Eyewear Store,
5464185.46
Total, Sweden Golf
Shop, 1503872.49
Total, Sweden
Outdoors Shop,
25710362.17
Total, Sweden
Sports Store,
27498984.23
Total, Sweden
Warehouse Store,
3294832.97
Total, Denmark
Department Store,
15684025.18
Total, Denmark
Direct Marketing,
1348469.49
Total, Denmark
Equipment Rental
Store, 1303052.22
Total, Denmark
Eyewear Store,
7502647.55
Total, Denmark Golf
Shop, 2479368.37
Total, Denmark
Outdoors Shop,
11664422
Total, Denmark
Sports Store,
4891462.19Total, Denmark
Warehouse Store,
3025084.93
Row Labels Sum of Revenue
Sweden 80635525.26
Department Store 16315108.7
Direct Marketing 848179.24
Eyewear Store 5464185.46
Golf Shop 1503872.49
Outdoors Shop 25710362.17
Sports Store 27498984.23
Warehouse Store 3294832.97
Denmark 47898531.93
Department Store 15684025.18
Direct Marketing 1348469.49
Equipment Rental
Store 1303052.22
Eyewear Store 7502647.55
Golf Shop 2479368.37
Outdoors Shop 11664422
Sports Store 4891462.19
Warehouse Store 3025084.93
Grand Total 128534057.2
14. PRESENTATION FOR SWEDEN & DENMARK RETAILER WISE REVENUE
Total,
Sweden
Department
Store,
16315108.7
Total,
Sweden
Direct
Marketing,
848179.24
Total,
Sweden
Eyewear
Store,
5464185.46
Total,
Sweden Golf
Shop,
1503872.49
Total,
Sweden
Outdoors
Shop,
25710362.17
Total,
Sweden
Sports Store,
27498984.23
Total,
Sweden
Warehouse
Store,
3294832.97
Total,
Denmark
Department
Store,
15684025.18
Total,
Denmark
Direct
Marketing,
1348469.49
Total,
Denmark
Equipment
Rental Store,
1303052.22
Total,
Denmark
Eyewear
Store,
7502647.55
Total,
Denmark Golf
Shop,
2479368.37
Total,
Denmark
Outdoors
Shop,
11664422
Total,
Denmark
Sports Store,
4891462.19
Total,
Denmark
Warehouse
Store,
3025084.93
Analysis Report
From this chart we can conclude that;
In Sweden our poor performers are;
1. Direct Marketing Team
2. Golf Shop
In Denmark our poor performers are;
1. Equipment Rental Store
2. Direct Marketing Team
Now we found out which retailer are our poor performers,
So we will know move deep into each of this retailer to
check how well they are utilizing resource.
15. SWEDEN POSSIBLE CAUSE & SUGGESTED IMPROVEMENTS EXPLAINED
Analysis Report for SWEDEN
Here we conclude;
Direct Marketing Team & Golf Shop are not
utilizing all of the resource and are the
weakest of all in generating Revenue.
Moreover;
Department Store, Eyewear Store,
Warehouse Store are also not utilizing
resource to the fullest thus Sweden is the
overall poor performer.
Possible Cause:
Lack of knowledge. [Note this is a general
conclusion for better result feedback from retailers
are mandatory.]
Suggested Improvements:
Training in Email, Sales Visit & Telephonic
Marketing for Retailers in Sweden.
Retailer country Sweden
Sum of Revenue Column Labels
Row Labels E-mail Sales visit Telephone Web Grand Total
Department Store 16315108.7 16315108.7
Direct Marketing 78042.73 770136.51 848179.24
Eyewear Store 5464185.46 5464185.46
Golf Shop 1503872.49 1503872.49
Outdoors Shop 3051204.78 1102964.04 7901582.71 13654610.64 25710362.17
Sports Store 829624.68 3102808.91 103229.38 23463321.26 27498984.23
Warehouse Store 3294832.97 3294832.97
Grand Total 3880829.46 4205772.95 8082854.82 64466068.03 80635525.26
16. DENMARK POSSIBLE CAUSE & SUGGESTED IMPROVEMENTS EXPLAINED
Analysis Report for SWEDEN
Here we conclude;
Direct Marketing Team & Equipment Rental
Store are not utilizing all of the resource
and are the weakest of all in generating
Revenue.
Moreover;
Department Store, Eyewear Store,
Warehouse Store, Golf Shop, Outdoor
Shop & Sports Store are also not utilizing
resource to the fullest thus Denmark is the
overall poor performer.
Possible Cause:
Lack of knowledge. [Note this is a general
conclusion for better result feedback from retailers
are mandatory.]
Suggested Improvements:
Training in Email, Sales Visit & Telephonic
Marketing for Retailers in Denmark.
Retailer country Denmark
Sum of Revenue Column Labels
Row Labels Fax Sales visit Web Grand Total
Department Store 15684025.18 15684025.18
Direct Marketing 806544.82 541924.67 1348469.49
Equipment Rental Store 1303052.22 1303052.22
Eyewear Store 7502647.55 7502647.55
Golf Shop 2479368.37 2479368.37
Outdoors Shop 2074894.91 9589527.09 11664422
Sports Store 189400.37 4702061.82 4891462.19
Warehouse Store 175517.78 2849567.15 3025084.93
Grand Total 175517.78 3070840.1 44652174.05 47898531.93
17. DENMARK POSSIBLE CAUSE & SUGGESTED IMPROVEMENTS EXPLAINED
Analysis Report for All Countries Retailers
We can safely comment that all other Countries retailers are utilizing all the resource however training can be a scope for Improvements
Retailer country (All)
Sum of Revenue Column Labels
Row Labels E-mail Fax Mail Sales visit Special Telephone Web Grand Total
Department Store 21891035.28 3387670.1 6770557.83 24259359.44 330808.09 36252434.25 569553845.9 662445710.8
Direct Marketing 1821356.46 12870.04 417580.97 9626910.75 1738879.43 40749479.96 54367077.61
Equipment Rental Store 75009.5 118768.81 532122.09 5018669.64 20856.64 1325558.36 36343206.23 43434191.27
Eyewear Store 453745.4 1312644.59 6236725.59 3392836.54 160322203.1 171718155.3
Golf Shop 4866719.76 5626272.5 1655897.9 10013141.1 1278483.46 21213341.34 366216312.8 410870168.9
Outdoors Shop 25233882.49 20513926.14 7161324.06 93319892.46 9432540 52046910.24 1059662257 1267370732
Sports Store 29547732.2 11599044.5 3089877.73 43924810.12 799923.46 26105030.15 854814341.7 969880759.9
Warehouse Store 550766.86 175517.78 2384483.29 16722603.8 1582789.85 20117210.39 150802797.1 192336169
Grand Total 84440247.95 41434069.87 23324488.46 209122112.9 13445401.5 162192200.7 3238464444 3772422965
18. “This brings us to the end of our
Analysis, and finding out the poor
performer and also helped in
understanding the
SCOPE FOR IMPROVEMENTS
20. COUNTRY WISE PRODUCT LINES PERFORMANCE
Denmark
Sweden
Belgium
Switzerland
Brazil
Australia
Austria
Spain
Mexico
Netherlands
Finland
Italy
Singapore
Korea
Germany
21. COUNTRY WISE PRODUCT LINES PERFORMANCE
Denmark
Sweden
Belgium
Switzerland
Brazil
Australia
Austria
Spain
Mexico
Netherlands
Finland
Italy
Singapore
Korea
Germany
United Kingdom
France
Canada
China
Japan
United States
From this chart it is clear that the
“Outdoor Protection” product line is
our poor performer globally.
OUR GOAL
Finding out ways to increase revenue
globally from the product line
“Outdoor Protection”
Lets Begin.
22. ANALYSING OUTDOOR PROTECTION PRODUCT LINE
The above headers are provided in client’s data, along that client informed that they invest 5% of difference of Revenue &
Cost of Good Sold for marketing. [ *Note this is a hypothetic statement.]
For that information we would calculate the required Cost of Goods Sold (COGS) & Marketing Cost.
Calculation for COGS:
Retailer
country
Order method
type Retailer type Product line Product type Product Year Quarter Revenue Quantity
Gross
margin
Formula Gross Margin =
Revenue - COGS
X 100
Revenue
COGS = Revenue -
Gross Margin x Revenue
100
According to the above formula we will calculate the COGS and marketing cost.
23. ANALYSING OUTDOOR PROTECTION PRODUCT LINE
Now to start with, create a new excel file with the data for “Outdoor Protection” product line.
To do checklist:
Remove rows with Value in Revenue as zero “0”.
Remove all the “Non quantitative” column from the file, such as Retailer country , Order method type ,
Retailer type , Product line , Product type , Product , Year , Quarter. Now that we are left with Revenue ,
Quantity , Gross margin , COGS , Marketing Cost,
*NB : COGS & Marketing Cost are calculated see previous slide.
From here we will do regression analysis using R Studio.
24. ANALYSING OUTDOOR PROTECTION PRODUCT LINE
data <- read.csv("D:R - BIG DATAAnalysisSales Data
WAOutdoorProtectionProduct.csv", header=T)
# For Regression analysis we need Quantitative Value only
# Hence deleting the other columns.
data <- data[,-(1:8)]
# For regression analysis we need to check the following;
# Normality should hold,
# There should be no Multi colinearity
# Data should be Homoscedacity
# There should be no Autocorrelation
# NB. To make our life easy in R the linear regression code gives output
# where there is a p-value determining significance,
# here we will not go into all the above checks we will simply work with
# considering p-value and i will explain it along side.
lm.mod <- lm(Revenue~., data = data)
summary(lm.mod)
R CODE R CONSOLE OUTPUT
Call:
lm(formula = Revenue ~ ., data = data)
Residuals:
Min 1Q Median 3Q Max
-1.012e-05 -1.560e-08 -6.000e-09 2.200e-08 1.039e-05
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -1.891e-08 4.154e-08 -4.550e-01 0.64900
Quantity 8.607e-11 3.242e-11 2.655e+00 0.00794 **
Gross.margin 3.980e-08 6.521e-08 6.100e-01 0.54165
COGS 1.000e+00 1.789e-11 5.590e+10 < 2e-16 ***
Marketing.Cost 2.000e+01 2.489e-10 8.034e+10 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 9.581e-07 on 8349 degrees of
freedom
Multiple R-squared: 1, Adjusted R-squared: 1
F-statistic: 1.16e+23 on 4 and 8349 DF, p-value: < 2.2e-16
Note the p-value and the “*” beside p-value. From the above we should
remove Gross Margin as it is of no significance.
25. ANALYSING OUTDOOR PROTECTION PRODUCT LINE
# In addition we are writing the following code to generate linear
model.
lm.mod <- lm(Revenue~Quantity+COGS+Marketing.Cost, data =
data)
summary(lm.mod)
R CODE R CONSOLE OUTPUT
Call:
lm(formula = Revenue ~ Quantity + COGS + Marketing.Cost, data
= data)
Residuals:
Min 1Q Median 3Q Max
-1.012e-05 -1.520e-08 -6.100e-09 2.250e-08 1.039e-05
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 5.245e-09 1.265e-08 4.150e-01 0.67848
Quantity 8.520e-11 3.238e-11 2.631e+00 0.00852 **
COGS 1.000e+00 1.676e-11 5.968e+10 < 2e-16 ***
Marketing.Cost 2.000e+01 2.342e-10 8.539e+10 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 9.581e-07 on 8350 degrees of freedom
Multiple R-squared: 1, Adjusted R-squared:
1
F-statistic: 1.547e+23 on 3 and 8350 DF, p-value: < 2.2e-16
Note the p-value and the “*” beside p-value. No we find all the
variables are significant.
So lets create the linear equation
Y=mx1+mx2+mx3+b
Y = Predicted Revenue
m = Slope or the estimate from console result.
x1, x2, x3 = Quantity, COGS, Marketing.Cost
b = Intercept;
Hence the formula; by replacing ‘m’ with the coefficient.
Y = 8.52*x1 + 1*x2 + 2*x3 + 5.25
26. Prediction Revenue
Now that we have our regression
equation
Y = 8.52*x1 + 1*x2 + 2*x3 + 5.25
Here ‘Y’ being the revenue.
We need the substitute the x1, x2 & x3
with respective values to determine
revenue, where x1 – Quantity.
x2 – COGS, x3 – Marketing cost
Lets Analyze for situation where,
Quantity produced = 100 unit, COGS = 100 & MC=
100;
Y=8.52*100+1*100+2*100+5.25
Y=1157
Now changing 1 unit in Quantity, and considering
all other remain constant
Y=8.52*101+1*100+2*100+5.25
Y=1166
We have a increase in Revenue with 1 unit change
in Production.
27. Data Analysis - Conclusion
Based on Multiple Linear regression analysis we can make vital Decision for betterment
of the Business.
With this we conclude our presentation. Thank you for your time
Subhankar Basak