Name - Jagadish Mali

Task 1 :-

Data preparation and customer analytics

Conduct analysis on your client's transaction dataset and identify customer purchasing behaviours to generate insights and provide
commercial recommendations.

The background information for this task :-

♦ I am part of Quantium’s retail analytics team and have been approached by our client, the Categor
y Manager for Chips, who wants to better understand the types of customers who purchase Chips and their
purchasing behaviour within the region.
♦ The insights from my analysis will feed into the supermarket’s strategic plan for the chip catego
ry in the next half year.

Here is task :-

♦ I need to present a strategic recommendation to Julia that is supported by data which she can then
use for the upcoming category review however to do so I need to analyse the data to understand the curr
ent purchasing trends and behaviours. The client is particularly interested in customer segments and th
eir chip purchasing behaviour. Consider what metrics would help describe the customers’ purchasing beha
viour.

• Examine transaction data - check for missing data, anomalies, outliers and clean them
• Examine customer data - similar to above transaction data
• Data analysis and customer segments - create charts and graphs, note trends and insights
• Deep dive into customer segments - determine which segments should be targetted

Importing Necessary Libraries

In [1]: import pandas as pd
import numpy as np
​
# for data visualization
import matplotlib.pyplot as plt
%matplotlib inline
import seaborn as sns
​
import warnings
warnings.filterwarnings('ignore')

Importing Dataset

In [2]: purchase_data = pd.read_csv('QVI_purchase_behaviour.csv')

purchase_data.head()

Out[2]:
LYLTY_CARD_NBR LIFESTAGE PREMIUM_CUSTOMER

0 1000 YOUNG SINGLES/COUPLES Premium

1 1002 YOUNG SINGLES/COUPLES Mainstream

2 1003 YOUNG FAMILIES Budget

3 1004 OLDER SINGLES/COUPLES Mainstream

4 1005 MIDAGE SINGLES/COUPLES Mainstream

In [3]: transaction_data = pd.read_excel('QVI_transaction_data.xlsx')

transaction_data.head()

Out[3]:
DATE STORE_NBR LYLTY_CARD_NBR TXN_ID PROD_NBR PROD_NAME PROD_QTY TOT_SALES

0 43390 1 1000 1 5 Natural Chip Compny SeaSalt175g 2 6.0

1 43599 1 1307 348 66 CCs Nacho Cheese 175g 3 6.3

2 43605 1 1343 383 61 Smiths Crinkle Cut Chips Chicken 170g 2 2.9

3 43329 2 2373 974 69 Smiths Chip Thinly S/Cream&Onion 175g 5 15.0

4 43330 2 2426 1038 108 Kettle Tortilla ChpsHny&Jlpno Chili 150g 3 13.8
 Data Exploration
In [4]: # Basic Information of dataset(QVI_purchase_behaviour)
purchase_data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 72637 entries, 0 to 72636
Data columns (total 3 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 LYLTY_CARD_NBR 72637 non-null int64
1 LIFESTAGE 72637 non-null object
2 PREMIUM_CUSTOMER 72637 non-null object
dtypes: int64(1), object(2)
memory usage: 1.7+ MB

In [5]: # Basic Information of dataset(QVI_transaction_data)

transaction_data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 264836 entries, 0 to 264835
Data columns (total 8 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 DATE 264836 non-null int64
1 STORE_NBR 264836 non-null int64
2 LYLTY_CARD_NBR 264836 non-null int64
3 TXN_ID 264836 non-null int64
4 PROD_NBR 264836 non-null int64
5 PROD_NAME 264836 non-null object
6 PROD_QTY 264836 non-null int64
7 TOT_SALES 264836 non-null float64
dtypes: float64(1), int64(6), object(1)
memory usage: 16.2+ MB
In [6]: # Statistical Summary of QVI_purchase_behaviour data
purchase_data.describe().T

Out[6]:
count mean std min 25% 50% 75% max

LYLTY_CARD_NBR 72637.0 136185.93177 89892.932014 1000.0 66202.0 134040.0 203375.0 2373711.0

In [7]: # Statistical Summary of QVI_transaction_data data

transaction_data.describe().T

Out[7]:
count mean std min 25% 50% 75% max

DATE 264836.0 43464.036260 105.389282 43282.0 43373.0 43464.0 43555.00 43646.0

STORE_NBR 264836.0 135.080110 76.784180 1.0 70.0 130.0 203.00 272.0

LYLTY_CARD_NBR 264836.0 135549.476404 80579.978022 1000.0 70021.0 130357.5 203094.25 2373711.0

TXN_ID 264836.0 135158.310815 78133.026026 1.0 67601.5 135137.5 202701.25 2415841.0

PROD_NBR 264836.0 56.583157 32.826638 1.0 28.0 56.0 85.00 114.0

PROD_QTY 264836.0 1.907309 0.643654 1.0 2.0 2.0 2.00 200.0

TOT_SALES 264836.0 7.304200 3.083226 1.5 5.4 7.4 9.20 650.0

Checking missing values

In [8]: ### Checking missing values of QVI_purchase_behaviour data
sns.heatmap(purchase_data.isnull())
plt.show()

In [9]: purchase_data.isnull().sum()

Out[9]: LYLTY_CARD_NBR 0
LIFESTAGE 0
PREMIUM_CUSTOMER 0
dtype: int64
In [10]: ### Checking missing values of QVI_transaction_data
sns.heatmap(transaction_data.isnull())
plt.show()

In [11]: transaction_data.isnull().sum()

Out[11]: DATE 0
STORE_NBR 0
LYLTY_CARD_NBR 0
TXN_ID 0
PROD_NBR 0
PROD_NAME 0
PROD_QTY 0
TOT_SALES 0
dtype: int64

•♦• As we can see there is no missing values in both dataset.

 Analyzing and Removing Outliers

In [12]: ### Merging both dataset

merged_data = pd.merge(purchase_data, transaction_data, on = 'LYLTY_CARD_NBR', how = 'right')
merged_data.head()

Out[12]:
LYLTY_CARD_NBR LIFESTAGE PREMIUM_CUSTOMER DATE STORE_NBR TXN_ID PROD_NBR PROD_NAME PROD_QTY

Natural Chip
YOUNG
0 1000 Premium 43390 1 1 5 Compny 2
SINGLES/COUPLES
SeaSalt175g

MIDAGE CCs Nacho

1 1307 Budget 43599 1 348 66 3
SINGLES/COUPLES Cheese 175g

Smiths Crinkle
MIDAGE
2 1343 Budget 43605 1 383 61 Cut Chips 2
SINGLES/COUPLES
Chicken 170g

Smiths Chip
MIDAGE Thinly
3 2373 Budget 43329 2 974 69 5
SINGLES/COUPLES S/Cream&Onion
175g

Kettle Tortilla
MIDAGE
4 2426 Budget 43330 2 1038 108 ChpsHny&Jlpno 3
SINGLES/COUPLES
Chili 150g

•♦• We can see "DATE" column is not in proper format, so we will change it.

In [13]: print(len(merged_data))
print(len(transaction_data))

264836
264836
In [14]: ### Basic Information of merged_data
merged_data.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 264836 entries, 0 to 264835
Data columns (total 10 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 LYLTY_CARD_NBR 264836 non-null int64
1 LIFESTAGE 264836 non-null object
2 PREMIUM_CUSTOMER 264836 non-null object
3 DATE 264836 non-null int64
4 STORE_NBR 264836 non-null int64
5 TXN_ID 264836 non-null int64
6 PROD_NBR 264836 non-null int64
7 PROD_NAME 264836 non-null object
8 PROD_QTY 264836 non-null int64
9 TOT_SALES 264836 non-null float64
dtypes: float64(1), int64(6), object(3)
memory usage: 22.2+ MB

Date column is not in proper format. so, date column should be datetime format

In [15]: from datetime import date, timedelta

​
start = date(1899, 12, 30)
new_date_format = []
for date in merged_data["DATE"]:
delta = timedelta(date)
new_date_format.append(start + delta)

In [16]: merged_data["DATE"] = pd.to_datetime(pd.Series(new_date_format))

print(merged_data["DATE"].dtype)

datetime64[ns]

Analyzing the product name column (PROD_NAME) to make sure all items are chips
In [17]: merged_data['PROD_NAME'].unique()

Out[17]: array(['Natural Chip Compny SeaSalt175g',

'CCs Nacho Cheese 175g',
'Smiths Crinkle Cut Chips Chicken 170g',
'Smiths Chip Thinly S/Cream&Onion 175g',
'Kettle Tortilla ChpsHny&Jlpno Chili 150g',
'Old El Paso Salsa Dip Tomato Mild 300g',
'Smiths Crinkle Chips Salt & Vinegar 330g',
'Grain Waves Sweet Chilli 210g',
'Doritos Corn Chip Mexican Jalapeno 150g',
'Grain Waves Sour Cream&Chives 210G',
'Kettle Sensations Siracha Lime 150g',
'Twisties Cheese 270g', 'WW Crinkle Cut Chicken 175g',
'Thins Chips Light& Tangy 175g', 'CCs Original 175g',
'Burger Rings 220g', 'NCC Sour Cream & Garden Chives 175g',
'Doritos Corn Chip Southern Chicken 150g',
'Cheezels Cheese Box 125g', 'Smiths Crinkle Original 330g',
'Infzns Crn Crnchers Tangy Gcamole 110g',
'Kettle Sea Salt And Vinegar 175g',
'Smiths Chip Thinly Cut Original 175g', 'Kettle Original 175g',
'Red Rock Deli Thai Chilli&Lime 150g',
'Pringles Sthrn FriedChicken 134g', 'Pringles Sweet&Spcy BBQ 134g',
'Red Rock Deli SR Salsa & Mzzrlla 150g',
'Thins Chips Originl saltd 175g',
'Red Rock Deli Sp Salt & Truffle 150G',
'Smiths Thinly Swt Chli&S/Cream175G', 'Kettle Chilli 175g',
'Doritos Mexicana 170g',
'Smiths Crinkle Cut French OnionDip 150g',
'Natural ChipCo Hony Soy Chckn175g',
'Dorito Corn Chp Supreme 380g', 'Twisties Chicken270g',
'Smiths Thinly Cut Roast Chicken 175g',
'Smiths Crinkle Cut Tomato Salsa 150g',
'Kettle Mozzarella Basil & Pesto 175g',
'Infuzions Thai SweetChili PotatoMix 110g',
'Kettle Sensations Camembert & Fig 150g',
'Smith Crinkle Cut Mac N Cheese 150g',
'Kettle Honey Soy Chicken 175g',
'Thins Chips Seasonedchicken 175g',
'Smiths Crinkle Cut Salt & Vinegar 170g',
'Infuzions BBQ Rib Prawn Crackers 110g',
'GrnWves Plus Btroot & Chilli Jam 180g',
'Tyrrells Crisps Lightly Salted 165g',
'Kettle Sweet Chilli And Sour Cream 175g',
'Doritos Salsa Medium 300g', 'Kettle 135g Swt Pot Sea Salt',
'Pringles SourCream Onion 134g',
'Doritos Corn Chips Original 170g',
'Twisties Cheese Burger 250g',
'Old El Paso Salsa Dip Chnky Tom Ht300g',
'Cobs Popd Swt/Chlli &Sr/Cream Chips 110g',
'Woolworths Mild Salsa 300g',
'Natural Chip Co Tmato Hrb&Spce 175g',
'Smiths Crinkle Cut Chips Original 170g',
'Cobs Popd Sea Salt Chips 110g',
'Smiths Crinkle Cut Chips Chs&Onion170g',
'French Fries Potato Chips 175g',
'Old El Paso Salsa Dip Tomato Med 300g',
'Doritos Corn Chips Cheese Supreme 170g',
'Pringles Original Crisps 134g',
'RRD Chilli& Coconut 150g',
'WW Original Corn Chips 200g',
'Thins Potato Chips Hot & Spicy 175g',
'Cobs Popd Sour Crm &Chives Chips 110g',
'Smiths Crnkle Chip Orgnl Big Bag 380g',
'Doritos Corn Chips Nacho Cheese 170g',
'Kettle Sensations BBQ&Maple 150g',
'WW D/Style Chip Sea Salt 200g',
'Pringles Chicken Salt Crips 134g',
'WW Original Stacked Chips 160g',
'Smiths Chip Thinly CutSalt/Vinegr175g', 'Cheezels Cheese 330g',
'Tostitos Lightly Salted 175g',
'Thins Chips Salt & Vinegar 175g',
'Smiths Crinkle Cut Chips Barbecue 170g', 'Cheetos Puffs 165g',
'RRD Sweet Chilli & Sour Cream 165g',
'WW Crinkle Cut Original 175g',
'Tostitos Splash Of Lime 175g', 'Woolworths Medium Salsa 300g',
'Kettle Tortilla ChpsBtroot&Ricotta 150g',
'CCs Tasty Cheese 175g', 'Woolworths Cheese Rings 190g',
'Tostitos Smoked Chipotle 175g', 'Pringles Barbeque 134g',
'WW Supreme Cheese Corn Chips 200g',
'Pringles Mystery Flavour 134g',
'Tyrrells Crisps Ched & Chives 165g',
'Snbts Whlgrn Crisps Cheddr&Mstrd 90g',
'Cheetos Chs & Bacon Balls 190g', 'Pringles Slt Vingar 134g',
'Infuzions SourCream&Herbs Veg Strws 110g',
'Kettle Tortilla ChpsFeta&Garlic 150g',
 'Infuzions Mango Chutny Papadums 70g',
'RRD Steak & Chimuchurri 150g',
'RRD Honey Soy Chicken 165g',
'Sunbites Whlegrn Crisps Frch/Onin 90g',
'RRD Salt & Vinegar 165g', 'Doritos Cheese Supreme 330g',
'Smiths Crinkle Cut Snag&Sauce 150g',
'WW Sour Cream &OnionStacked Chips 160g',
'RRD Lime & Pepper 165g',
'Natural ChipCo Sea Salt & Vinegr 175g',
'Red Rock Deli Chikn&Garlic Aioli 150g',
'RRD SR Slow Rst Pork Belly 150g', 'RRD Pc Sea Salt 165g',
'Smith Crinkle Cut Bolognese 150g', 'Doritos Salsa Mild 300g'],
dtype=object)

In [18]: split_prods = merged_data["PROD_NAME"].str.replace(r'([0-9]+[gG])','').str.replace(r'[^\w]',' ').str.split()

In [19]: word_counts = {}
def count_words(line):
for word in line:
if word not in word_counts:
word_counts[word] = 1
else:
word_counts[word] += 1
split_prods.apply(lambda line: count_words(line))
print(pd.Series(word_counts).sort_values(ascending = False))

Chips 49770
Kettle 41288
Smiths 28860
Salt 27976
Cheese 27890
...
Sunbites 1432
Pc 1431
Garden 1419
NCC 1419
Fries 1418
Length: 198, dtype: int64
In [20]: print("\n ----- Statistical Summary of Merged Data ----- \n")
print(merged_data.describe())
print("\n ----- Basic Information of Merged Data ----- \n")
print(merged_data.info())

----- Statistical Summary of Merged Data -----

LYLTY_CARD_NBR STORE_NBR TXN_ID PROD_NBR \

count 2.648360e+05 264836.00000 2.648360e+05 264836.000000
mean 1.355495e+05 135.08011 1.351583e+05 56.583157
std 8.057998e+04 76.78418 7.813303e+04 32.826638
min 1.000000e+03 1.00000 1.000000e+00 1.000000
25% 7.002100e+04 70.00000 6.760150e+04 28.000000
50% 1.303575e+05 130.00000 1.351375e+05 56.000000
75% 2.030942e+05 203.00000 2.027012e+05 85.000000
max 2.373711e+06 272.00000 2.415841e+06 114.000000

PROD_QTY TOT_SALES
count 264836.000000 264836.000000
mean 1.907309 7.304200
std 0.643654 3.083226
min 1.000000 1.500000
25% 2.000000 5.400000
50% 2.000000 7.400000
75% 2.000000 9.200000
max 200.000000 650.000000

----- Basic Information of Merged Data -----

<class 'pandas.core.frame.DataFrame'>
Int64Index: 264836 entries, 0 to 264835
Data columns (total 10 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 LYLTY_CARD_NBR 264836 non-null int64
1 LIFESTAGE 264836 non-null object
2 PREMIUM_CUSTOMER 264836 non-null object
3 DATE 264836 non-null datetime64[ns]
4 STORE_NBR 264836 non-null int64
5 TXN_ID 264836 non-null int64
6 PROD_NBR 264836 non-null int64
 7 PROD_NAME 264836 non-null object
8 PROD_QTY 264836 non-null int64
9 TOT_SALES 264836 non-null float64
dtypes: datetime64[ns](1), float64(1), int64(5), object(3)
memory usage: 22.2+ MB
None

In [21]: merged_data["PROD_QTY"].value_counts(bins=4).sort_index()

Out[21]: (0.8, 50.75] 264834

(50.75, 100.5] 0
(100.5, 150.25] 0
(150.25, 200.0] 2
Name: PROD_QTY, dtype: int64

♦ From above binning we see that "PROD_QTY" values above 50.75

In [22]: merged_data.sort_values(by="PROD_QTY", ascending=False).head()

Out[22]:
LYLTY_CARD_NBR LIFESTAGE PREMIUM_CUSTOMER DATE STORE_NBR TXN_ID PROD_NBR PROD_NAME PROD_QT

Dorito Corn
2018-
69762 226000 OLDER FAMILIES Premium 226 226201 4 Chp Supreme 20
08-19
380g

Dorito Corn
2019-
69763 226000 OLDER FAMILIES Premium 226 226210 4 Chp Supreme 20
05-20
380g

Pringles
2019-
217237 201060 YOUNG FAMILIES Premium 201 200202 26 Sweet&Spcy
05-18
BBQ 134g

Pringles
YOUNG 2018-
238333 219004 Mainstream 219 218018 25 SourCream
SINGLES/COUPLES 08-14
Onion 134g

Infuzions BBQ
YOUNG 2019- Rib Prawn
238471 261331 Mainstream 261 261111 87
SINGLES/COUPLES 05-19 Crackers
110g
 ♦ Two outliers of value 200 in PROD_QTY will be removed. Both entries are by the same customer and
will be examined by this customer's transactions.

In [23]: merged_data = merged_data[merged_data["PROD_QTY"] < 6]

In [24]: len(merged_data[merged_data["LYLTY_CARD_NBR"]==226000])

Out[24]: 0

In [25]: merged_data["DATE"].describe()

Out[25]: count 264834

unique 364
top 2018-12-24 00:00:00
freq 939
first 2018-07-01 00:00:00
last 2019-06-30 00:00:00
Name: DATE, dtype: object

♦ There are 365 days in a year but in the DATE column there are only 364 unique values so one is missing.

In [26]: pd.date_range(start=merged_data["DATE"].min(),
end=merged_data["DATE"].max()).difference(merged_data["DATE"])

Out[26]: DatetimeIndex(['2018-12-25'], dtype='datetime64[ns]', freq=None)

♦ Using the difference method we see that 2018-12-25 was a missing date

In [27]: check_null_date = pd.merge(pd.Series(pd.date_range(start=merged_data["DATE"].min(),

end = merged_data["DATE"].max()),
name="DATE"), merged_data, on = "DATE", how = "left")
In [28]: trans_by_date = check_null_date["DATE"].value_counts()
dec = trans_by_date[(trans_by_date.index >= pd.datetime(2018,12,1)) & (trans_by_date.index < pd.datetime(2019,1,
dec.index = dec.index.strftime('%d')
ax = dec.plot(figsize=(15,3), color='orange')
ax.set_xticks(np.arange(len(dec)))
ax.set_xticklabels(dec.index)
plt.title("Sales of December 2018", fontsize=20, fontweight='bold', color='brown')
plt.xlabel("Date", fontsize=15, fontweight='bold', color='brown')
plt.ylabel("Number of Sales", fontsize=10, fontweight='bold', color='brown')
plt.savefig("Sales of December 2018.png", bbox_inches="tight")
plt.grid()
plt.legend()
plt.show()
In [29]: check_null_date["DATE"].value_counts().sort_values().head()

Out[29]: 2018-12-25 1
2018-11-25 648
2018-10-18 658
2019-06-13 659
2019-06-24 662
Name: DATE, dtype: int64

The day with no transaction is a Christmas Day (25th December). That is when the store is closed. So there is no anomaly in
this.

Analyzing Packet sizes

In [30]: merged_data["PROD_NAME"] = merged_data["PROD_NAME"].str.replace(r'[0-9]+(G)','g')
pack_sizes = merged_data["PROD_NAME"].str.extract(r'([0-9]+[gG])')[0].str.replace("g","").astype("float")
​
print("\n ----- Statistical Summary ----- \n")
print(pack_sizes.describe())
​
print("\n ----- Value Counts ----- \n")
print(pack_sizes.value_counts())
​
print("\n ----- Histogram of Packet sizes ----- \n")
pack_sizes.plot.hist()
plt.show()

----- Statistical Summary -----

count 258770.000000
mean 182.324276
std 64.955035
min 70.000000
25% 150.000000
50% 170.000000
75% 175.000000
max 380.000000
Name: 0, dtype: float64

----- Value Counts -----

175.0 64929
150.0 41633
134.0 25102
110.0 22387
170.0 19983
165.0 15297
300.0 15166
330.0 12540
380.0 6416
270.0 6285
200.0 4473
135.0 3257
250.0 3169
210.0 3167
90.0 3008
190.0 2995
160.0 2970
220.0 1564
70.0 1507
180.0 1468
125.0 1454
Name: 0, dtype: int64

----- Histogram of Packet sizes -----

In [31]: merged_data["PROD_NAME"].str.split().str[0].value_counts().sort_index()

Out[31]: Burger 1564

CCs 4551
Cheetos 2927
Cheezels 4603
Cobs 9693
Dorito 3183
Doritos 24962
French 1418
Grain 6272
GrnWves 1468
Infuzions 11057
Infzns 3144
Kettle 41288
NCC 1419
Natural 6050
Old 9324
Pringles 25102
RRD 11894
Red 5885
Smith 2963
Smiths 28860
Snbts 1576
Sunbites 1432
Thins 14075
Tostitos 9471
Twisties 9454
Tyrrells 6442
WW 10320
Woolworths 4437
Name: PROD_NAME, dtype: int64

♦ Some product names are written in more than one way. Example : Dorito and Doritos, Grains and GrnWv
es, Infusions and Ifzns, Natural and NCC, Red and RRD, Smith and Smiths and Snbts and Sunbites.
In [32]: merged_data["PROD_NAME"].str.split()[merged_data["PROD_NAME"].str.split().str[0] == "Red"].value_counts()

Out[32]: [Red, Rock, Deli, Sp, Salt, &, Truffle, g] 1498

[Red, Rock, Deli, Thai, Chilli&Lime, 150g] 1495
[Red, Rock, Deli, SR, Salsa, &, Mzzrlla, 150g] 1458
[Red, Rock, Deli, Chikn&Garlic, Aioli, 150g] 1434
Name: PROD_NAME, dtype: int64

In [33]: merged_data["Cleaned_Brand_Names"] = merged_data["PROD_NAME"].str.split().str[0]

In [34]: def clean_brand_names(line):

brand = line["Cleaned_Brand_Names"]
if brand == "Dorito":
return "Doritos"
elif brand == "GrnWves" or brand == "Grain":
return "Grain Waves"
elif brand == "Infzns":
return "Infuzions"
elif brand == "Natural" or brand == "NCC":
return "Natural Chip Co"
elif brand == "Red":
return "RRD"
elif brand == "Smith":
return "Smiths"
elif brand == "Snbts":
return "Sunbites"
elif brand == "WW":
return "Woolworths"
else:
return brand

In [35]: merged_data["Cleaned_Brand_Names"] = merged_data.apply(lambda line: clean_brand_names(line), axis=1)

In [36]: merged_data["Cleaned_Brand_Names"].value_counts(ascending=True).plot.barh(figsize=(15,8), color='darkgoldenrod')
plt.title("Brand Name", fontsize=25, fontweight='bold', color='olive')
plt.grid(color='olive', linestyle='--')
plt.savefig("Brand Names.png", bbox_inches="tight")
plt.show()
In [37]: sns.heatmap(merged_data.isnull())
plt.show()

In [38]: merged_data.isnull().sum()

Out[38]: LYLTY_CARD_NBR 0
LIFESTAGE 0
PREMIUM_CUSTOMER 0
DATE 0
STORE_NBR 0
TXN_ID 0
PROD_NBR 0
PROD_NAME 0
PROD_QTY 0
TOT_SALES 0
Cleaned_Brand_Names 0
dtype: int64
Questions :-

♦ Who spends the most on chips (total sales), describing customers by lifestage and how premium the
ir general purchasing behaviour is ?
♦ How many customers are in each segment ?
♦ How many chips are bought per customer by segment ?
♦ What is the average chip price by customer segment ?
In [39]: grouped_sales = pd.DataFrame(merged_data.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"])["TOT_SALES"].agg(["sum", "me
grouped_sales.sort_values(ascending=False, by="sum")

Out[39]:
sum mean

LIFESTAGE PREMIUM_CUSTOMER

OLDER FAMILIES Budget 168363.25 7.269570

YOUNG SINGLES/COUPLES Mainstream 157621.60 7.558339

RETIREES Mainstream 155677.05 7.252262

YOUNG FAMILIES Budget 139345.85 7.287201

OLDER SINGLES/COUPLES Budget 136769.80 7.430315

Mainstream 133393.80 7.282116

Premium 132263.15 7.449766

RETIREES Budget 113147.80 7.443445

OLDER FAMILIES Mainstream 103445.55 7.262395

RETIREES Premium 97646.05 7.456174

YOUNG FAMILIES Mainstream 92788.75 7.189025

MIDAGE SINGLES/COUPLES Mainstream 90803.85 7.647284

YOUNG FAMILIES Premium 84025.50 7.266756

OLDER FAMILIES Premium 80658.40 7.208079

YOUNG SINGLES/COUPLES Budget 61141.60 6.615624

MIDAGE SINGLES/COUPLES Premium 58432.65 7.112056

YOUNG SINGLES/COUPLES Premium 41642.10 6.629852

MIDAGE SINGLES/COUPLES Budget 35514.80 7.074661

NEW FAMILIES Budget 21928.45 7.297321

Mainstream 17013.90 7.317806

Premium 11491.10 7.231655

In [40]: grouped_sales["sum"].sum()

Out[40]: 1933115.0000000002

In [41]: grouped_sales["sum"].sort_values().plot.barh(figsize=(15,8), color='darkgoldenrod')

plt.grid(color='olive', linestyle='--')
plt.show()
In [42]: # Values of each group
bars1 = grouped_sales[grouped_sales.index.get_level_values("PREMIUM_CUSTOMER") == "Budget"]["sum"]
bars2 = grouped_sales[grouped_sales.index.get_level_values("PREMIUM_CUSTOMER") == "Mainstream"]["sum"]
bars3 = grouped_sales[grouped_sales.index.get_level_values("PREMIUM_CUSTOMER") == "Premium"]["sum"]

bars1_text = (bars1 / sum(grouped_sales["sum"])).apply("{:.1%}".format)

bars2_text = (bars2 / sum(grouped_sales["sum"])).apply("{:.1%}".format)
bars3_text = (bars3 / sum(grouped_sales["sum"])).apply("{:.1%}".format)

# Names of group and bar width

names = grouped_sales.index.get_level_values("LIFESTAGE").unique()
​
# The position of the bars on the x-axis
r = np.arange(len(names))
​
plt.figure(figsize=(13,5))
​
# Create brown bars
budget_bar = plt.barh(r, bars1, edgecolor='grey', height=1, label="Budget")
# Create green bars (middle), on top of the firs ones
mains_bar = plt.barh(r, bars2, left=bars1, edgecolor='grey', height=1, label="Mainstream")
# Create green bars (top)
tmp_bar = np.add(bars1, bars2)
prem_bar = plt.barh(r, bars3, left=bars2, edgecolor='grey', height=1, label="Premium")
​
for i in range(7):
budget_width = budget_bar[i].get_width()
budget_main_width = budget_width + mains_bar[i].get_width()
plt.text(budget_width/2, i, bars1_text[i], va='center', ha='center', size=8)
plt.text(budget_width + mains_bar[i].get_width()/2, i, bars2_text[i], va='center', ha='center', size=8)
plt.text(budget_main_width + prem_bar[i].get_width()/2, i, bars3_text[i], va='center', ha='center', size=8)

# Custom X axis
plt.yticks(r, names)
plt.ylabel("LIFESTAGE")
plt.xlabel("TOTAL SALES")
plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))
​
plt.title("Total Sales per Lifestage")
​
plt.savefig("lifestage_sales.png", bbox_inches="tight")
​
# Show graphic
plt.show()
In [43]: stage_agg_prem = merged_data.groupby("LIFESTAGE")["PREMIUM_CUSTOMER"].agg(pd.Series.mode).sort_values()
print("\n ----- Top contributor per LIFESTAGE by PREMIUM category ----- \n")
print(stage_agg_prem)

----- Top contributor per LIFESTAGE by PREMIUM category -----

LIFESTAGE
NEW FAMILIES Budget
OLDER FAMILIES Budget
OLDER SINGLES/COUPLES Budget
YOUNG FAMILIES Budget
MIDAGE SINGLES/COUPLES Mainstream
RETIREES Mainstream
YOUNG SINGLES/COUPLES Mainstream
Name: PREMIUM_CUSTOMER, dtype: object

The top 3 total sales contributor segment are (in order) :-

1. Older families (Budget) $156,864

2. Young Singles/Couples (Mainstream) $147,582
3. Retirees (Mainstream) $145,169
In [44]: unique_cust = merged_data.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"])["LYLTY_CARD_NBR"].nunique().sort_values(asc
pd.DataFrame(unique_cust)

Out[44]:
LYLTY_CARD_NBR

LIFESTAGE PREMIUM_CUSTOMER

YOUNG SINGLES/COUPLES Mainstream 8088

RETIREES Mainstream 6479

OLDER SINGLES/COUPLES Mainstream 4930

Budget 4929

Premium 4750

OLDER FAMILIES Budget 4675

RETIREES Budget 4454

YOUNG FAMILIES Budget 4017

RETIREES Premium 3872

YOUNG SINGLES/COUPLES Budget 3779

MIDAGE SINGLES/COUPLES Mainstream 3340

OLDER FAMILIES Mainstream 2831

YOUNG FAMILIES Mainstream 2728

YOUNG SINGLES/COUPLES Premium 2574

YOUNG FAMILIES Premium 2433

MIDAGE SINGLES/COUPLES Premium 2431

OLDER FAMILIES Premium 2273

MIDAGE SINGLES/COUPLES Budget 1504

NEW FAMILIES Budget 1112

Mainstream 849

Premium 588
In [45]: unique_cust.sort_values().plot.barh(figsize=(15,8), color='darkgoldenrod')
plt.grid(color='olive', linestyle='--')
plt.show()
In [46]: # Values of each group
ncust_bars1 = unique_cust[unique_cust.index.get_level_values("PREMIUM_CUSTOMER") == "Budget"]
ncust_bars2 = unique_cust[unique_cust.index.get_level_values("PREMIUM_CUSTOMER") == "Mainstream"]
ncust_bars3 = unique_cust[unique_cust.index.get_level_values("PREMIUM_CUSTOMER") == "Premium"]

ncust_bars1_text = (ncust_bars1 / sum(unique_cust)).apply("{:.1%}".format)

ncust_bars2_text = (ncust_bars2 / sum(unique_cust)).apply("{:.1%}".format)
ncust_bars3_text = (ncust_bars3 / sum(unique_cust)).apply("{:.1%}".format)

# # Names of group and bar width

#names = unique_cust.index.get_level_values("LIFESTAGE").unique()
​
# # The position of the bars on the x-axis
#r = np.arange(len(names))
​
plt.figure(figsize=(13,5))
​
# # Create brown bars
budget_bar = plt.barh(r, ncust_bars1, edgecolor='grey', height=1, label="Budget")
# # Create green bars (middle), on top of the firs ones
mains_bar = plt.barh(r, ncust_bars2, left=ncust_bars1, edgecolor='grey', height=1, label="Mainstream")
# # Create green bars (top)
prem_bar = plt.barh(r, ncust_bars3, left=ncust_bars2, edgecolor='grey', height=1, label="Premium")
​
for i in range(7):
budget_width = budget_bar[i].get_width()
budget_main_width = budget_width + mains_bar[i].get_width()
plt.text(budget_width/2, i, ncust_bars1_text[i], va='center', ha='center', size=8)
plt.text(budget_width + mains_bar[i].get_width()/2, i, ncust_bars2_text[i], va='center', ha='center', size=8
plt.text(budget_main_width + prem_bar[i].get_width()/2, i, ncust_bars3_text[i], va='center', ha='center', si

# Custom X axis
plt.yticks(r, names)
plt.ylabel("Lifestage", fontsize=15, fontweight='bold', color='darkgoldenrod')
plt.xlabel("Unique Customers", fontsize=15, fontweight='bold', color='darkgoldenrod')
plt.legend(loc='center left', bbox_to_anchor=(1.0, 0.5))
​
plt.title("Unique Customers per Lifestage", fontsize=20, fontweight='bold', color='darkgoldenrod')
​
plt.savefig("lifestage_customers.png", bbox_inches="tight")
​
# View
plt.show()

The high sales amount by segment "Young Singles/Couples - Mainstream" and "Retirees - Mainstream" are due to their large
number of unique customers, but not for the "Older - Budget" segment. Next we'll analyze if the "Older - Budget" segment
has:

High Frequency of Purchase and Average Sales per Customer compared to the other segment.
In [47]: freq_per_cust = merged_data.groupby(["LYLTY_CARD_NBR", "LIFESTAGE", "PREMIUM_CUSTOMER"]).count()["DATE"]
freq_per_cust.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"]).agg(["mean", "count"]).sort_values(ascending=False, by=

Out[47]:
mean count

LIFESTAGE PREMIUM_CUSTOMER

OLDER FAMILIES Mainstream 5.031438 2831

Budget 4.954011 4675

Premium 4.923009 2273

YOUNG FAMILIES Budget 4.760269 4017

Premium 4.752569 2433

Mainstream 4.731305 2728

OLDER SINGLES/COUPLES Premium 3.737684 4750

Budget 3.734429 4929

Mainstream 3.715619 4930

MIDAGE SINGLES/COUPLES Mainstream 3.555090 3340

RETIREES Budget 3.412887 4454

Premium 3.382231 3872

MIDAGE SINGLES/COUPLES Premium 3.379679 2431

Budget 3.337766 1504

RETIREES Mainstream 3.313166 6479

NEW FAMILIES Mainstream 2.738516 849

Premium 2.702381 588

Budget 2.702338 1112

YOUNG SINGLES/COUPLES Mainstream 2.578388 8088

Budget 2.445621 3779

Premium 2.440171 2574

•♦• The above table describes the "Average frequency of Purchase per segment" and "Unique custom
 er per segment". The top three most frequent purchase is contributed by the "Older Families" lifestage
segment. We can see now that the "Older - Budget" segment contributes to high sales partly because of
the combination of:

High Frequency of Purchase and, Fairly high unique number of customer in the segment
In [48]: grouped_sales.sort_values(ascending=False, by="mean")

Out[48]:
sum mean

LIFESTAGE PREMIUM_CUSTOMER

MIDAGE SINGLES/COUPLES Mainstream 90803.85 7.647284

YOUNG SINGLES/COUPLES Mainstream 157621.60 7.558339

RETIREES Premium 97646.05 7.456174

OLDER SINGLES/COUPLES Premium 132263.15 7.449766

RETIREES Budget 113147.80 7.443445

OLDER SINGLES/COUPLES Budget 136769.80 7.430315

NEW FAMILIES Mainstream 17013.90 7.317806

Budget 21928.45 7.297321

YOUNG FAMILIES Budget 139345.85 7.287201

OLDER SINGLES/COUPLES Mainstream 133393.80 7.282116

OLDER FAMILIES Budget 168363.25 7.269570

YOUNG FAMILIES Premium 84025.50 7.266756

OLDER FAMILIES Mainstream 103445.55 7.262395

RETIREES Mainstream 155677.05 7.252262

NEW FAMILIES Premium 11491.10 7.231655

OLDER FAMILIES Premium 80658.40 7.208079

YOUNG FAMILIES Mainstream 92788.75 7.189025

MIDAGE SINGLES/COUPLES Premium 58432.65 7.112056

Budget 35514.80 7.074661

YOUNG SINGLES/COUPLES Premium 41642.10 6.629852

Budget 61141.60 6.615624

•♦• Highest average spending per purchase are contributed by the Midage and Young "Singles/Couple
s". The difference between their Mainstream and Non-Mainstream group might seem insignificant (7.6 vs
 6.6), but we'll find out by examining if the difference is statistically significant.

In [49]: from scipy.stats import ttest_ind

mainstream = merged_data["PREMIUM_CUSTOMER"] == "Mainstream"
young_midage = (merged_data["LIFESTAGE"] == "MIDAGE SINGLES/COUPLES") | (merged_data["LIFESTAGE"] == "YOUNG SING
​
budget_premium = (merged_data["PREMIUM_CUSTOMER"] == "Budget") | (merged_data["PREMIUM_CUSTOMER"] == "Premium")
​
a = merged_data[young_midage & mainstream]["TOT_SALES"]
b = merged_data[young_midage & budget_premium]["TOT_SALES"]
stat, pval = ttest_ind(a.values, b.values, equal_var=False)
​
print(pval)
pval < 0.0000001

1.8542040107534844e-281

Out[49]: True

•♦• P-Value is close to 0. There is a statistically significant difference to the Total Sales betwee
n the "Mainstream Young Midage" segment to the "Budget and Premium Young Midage" segment.

Next, let's look examine what brand of chips the top 3 segments contributing to Total Sales are buying.
In [50]: merged_data.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"])["Cleaned_Brand_Names"].agg(pd.Series.mode).sort_values()

Out[50]: LIFESTAGE PREMIUM_CUSTOMER

MIDAGE SINGLES/COUPLES Budget Kettle
YOUNG FAMILIES Premium Kettle
Mainstream Kettle
Budget Kettle
RETIREES Premium Kettle
Mainstream Kettle
Budget Kettle
OLDER SINGLES/COUPLES Premium Kettle
YOUNG SINGLES/COUPLES Mainstream Kettle
OLDER SINGLES/COUPLES Mainstream Kettle
OLDER FAMILIES Mainstream Kettle
Budget Kettle
NEW FAMILIES Premium Kettle
Mainstream Kettle
Budget Kettle
MIDAGE SINGLES/COUPLES Premium Kettle
Mainstream Kettle
OLDER SINGLES/COUPLES Budget Kettle
YOUNG SINGLES/COUPLES Premium Kettle
OLDER FAMILIES Premium Smiths
YOUNG SINGLES/COUPLES Budget Smiths
Name: Cleaned_Brand_Names, dtype: object
In [51]: for stage in merged_data["LIFESTAGE"].unique():
for prem in merged_data["PREMIUM_CUSTOMER"].unique():
print("----------",stage, '-', prem,"----------\n")
summary = merged_data[(merged_data["LIFESTAGE"] == stage)
& (merged_data["PREMIUM_CUSTOMER"] == prem)]["Cleaned_Brand_Names"].value_counts()
print(summary)
plt.figure()
summary.plot.barh(figsize=(6,2), color='orangered')
plt.show()

---------- YOUNG SINGLES/COUPLES - Premium ----------

Kettle 838
Smiths 826
Doritos 570
Name: Cleaned_Brand_Names, dtype: int64

---------- YOUNG SINGLES/COUPLES - Budget ----------

Smiths 1245
Kettle 1211
Doritos 899
Name: Cleaned_Brand_Names, dtype: int64

---------- YOUNG SINGLES/COUPLES - Mainstream ----------

Kettle 3844
Doritos 2541
Pringles 2315
Name: Cleaned_Brand_Names, dtype: int64

---------- MIDAGE SINGLES/COUPLES - Premium ----------

Kettle 1206
Smiths 986
Doritos 837
Name: Cleaned_Brand_Names, dtype: int64

---------- MIDAGE SINGLES/COUPLES - Budget ----------

Kettle 713
Smiths 633
Doritos 533
Name: Cleaned_Brand_Names, dtype: int64
---------- MIDAGE SINGLES/COUPLES - Mainstream ----------

Kettle 2136
Smiths 1337
Doritos 1291
Name: Cleaned_Brand_Names, dtype: int64

---------- NEW FAMILIES - Premium ----------

Kettle 247
Doritos 167
Pringles 165
Name: Cleaned_Brand_Names, dtype: int64

---------- NEW FAMILIES - Budget ----------

Kettle 510
Doritos 343
Smiths 341
Name: Cleaned Brand Names, dtype: int64

---------- NEW FAMILIES - Mainstream ----------

Kettle 414
Doritos 274
Smiths 254
Name: Cleaned_Brand_Names, dtype: int64

---------- OLDER FAMILIES - Premium ----------

Smiths 1515
Kettle 1512
Doritos 1065
Name: Cleaned_Brand_Names, dtype: int64
---------- OLDER FAMILIES - Budget ----------

Kettle 3320
Smiths 3093
Doritos 2351
Name: Cleaned_Brand_Names, dtype: int64

---------- OLDER FAMILIES - Mainstream ----------

Kettle 2019
Smiths 1835
Doritos 1449
Name: Cleaned_Brand_Names, dtype: int64

---------- OLDER SINGLES/COUPLES - Premium ----------

Kettle 2947
Smiths 2042
Doritos 1958
Name: Cleaned_Brand_Names, dtype: int64
---------- OLDER SINGLES/COUPLES - Budget ----------

Kettle 3065
Smiths 2098
Doritos 1954
Name: Cleaned_Brand_Names, dtype: int64

---------- OLDER SINGLES/COUPLES - Mainstream ----------

Kettle 2835
Smiths 2180
Doritos 2008
Name: Cleaned_Brand_Names, dtype: int64

---------- RETIREES - Premium ----------

Kettle 2216
Smiths 1458
Doritos 1409
Name: Cleaned_Brand_Names, dtype: int64

---------- RETIREES - Budget ----------

Kettle 2592
Doritos 1742
Smiths 1679
Name: Cleaned_Brand_Names, dtype: int64

---------- RETIREES - Mainstream ----------

Kettle 3386
Smiths 2476
Doritos 2320
Name: Cleaned_Brand_Names, dtype: int64
---------- YOUNG FAMILIES - Premium ----------

Kettle 1745
Smiths 1442
Doritos 1129
Name: Cleaned_Brand_Names, dtype: int64

---------- YOUNG FAMILIES - Budget ----------

Kettle 2743
Smiths 2459
Doritos 1996
Name: Cleaned_Brand_Names, dtype: int64

---------- YOUNG FAMILIES - Mainstream ----------

Kettle 1789
Smiths 1772
Doritos 1309
Name: Cleaned_Brand_Names, dtype: int64
•♦• Every segment had Kettle as the most purchased brand. Every segment except "YOUNG SINGLES/COUPLES M
ainstream" had Smiths as their second most purchased brand. "YOUNG SINGLES/COUPLES Mainstream" had Dori
tos as their second most purchased brand.
In [52]: from mlxtend.frequent_patterns import apriori
from mlxtend.frequent_patterns import association_rules
​
temp = merged_data.reset_index().rename(columns = {"index": "transaction"})
temp["Segment"] = temp["LIFESTAGE"] + ' - ' + temp['PREMIUM_CUSTOMER']
segment_brand_encode = pd.concat([pd.get_dummies(temp["Segment"]), pd.get_dummies(temp["Cleaned_Brand_Names"])],
​
frequent_sets = apriori(segment_brand_encode, min_support=0.01, use_colnames=True)
rules = association_rules(frequent_sets, metric="lift", min_threshold=1)
​
set_temp = temp["Segment"].unique()
rules[rules["antecedents"].apply(lambda x: list(x)).apply(lambda x: x in set_temp)]

C:\Users\Admin\AppData\Local\Programs\Python\Python310\lib\site-packages\mlxtend\frequent_patterns\fpcommon.p
y:111: DeprecationWarning: DataFrames with non-bool types result in worse computationalperformance and their s
upport might be discontinued in the future.Please use a DataFrame with bool type
warnings.warn(

Out[52]:
antecedent consequent
antecedents consequents support confidence lift leverage conviction
support support

1 (OLDER FAMILIES - Budget) (Smiths) 0.087451 0.120162 0.011679 0.133549 1.111409 0.001171 1.015451

(OLDER SINGLES/COUPLES -
3 (Kettle) 0.069504 0.155901 0.011573 0.166513 1.068064 0.000738 1.012731
Budget)

(OLDER SINGLES/COUPLES -
5 (Kettle) 0.067038 0.155901 0.011128 0.165991 1.064716 0.000676 1.012097
Premium)

7 (RETIREES - Mainstream) (Kettle) 0.081055 0.155901 0.012785 0.157738 1.011779 0.000149 1.002180

(YOUNG SINGLES/COUPLES -
8 (Kettle) 0.078744 0.155901 0.014515 0.184329 1.182344 0.002239 1.034852
Mainstream)

•♦• By looking at our a-priori analysis, we can conclude that Kettle is the brand of choice for mos
t segment.

Next, we'll find out the pack size preferences of different segments
In [53]: merged_pack = pd.concat([merged_data, pack_sizes.rename("Pack_Size")], axis=1)
​
for stage in merged_data["LIFESTAGE"].unique():
for prem in merged_data["PREMIUM_CUSTOMER"].unique():
print("----------",stage, '-', prem,"----------\n")
summary = merged_pack[(merged_pack["LIFESTAGE"] == stage)
& (merged_pack["PREMIUM_CUSTOMER"] == prem)]["Pack_Size"].value_counts().head(3).s
print(summary)
plt.figure()
summary.plot.barh(figsize=(6,2), color='olive')
plt.show()

---------- YOUNG SINGLES/COUPLES - Premium ----------

134.0 537
150.0 961
175.0 1587
Name: Pack_Size, dtype: int64

---------- YOUNG SINGLES/COUPLES - Budget ----------

134.0 832
150.0 1439
175.0 2262
Name: Pack_Size, dtype: int64
---------- YOUNG SINGLES/COUPLES - Mainstream ----------

134.0 2315
150.0 3159
175.0 4928
Name: Pack_Size, dtype: int64

---------- MIDAGE SINGLES/COUPLES - Premium ----------

134.0 781
150.0 1285
175.0 2034
Name: Pack_Size, dtype: int64
---------- MIDAGE SINGLES/COUPLES - Budget ----------

134.0 449
150.0 821
175.0 1256
Name: Pack_Size, dtype: int64

---------- MIDAGE SINGLES/COUPLES - Mainstream ----------

134.0 1159
150.0 1819
175.0 2912
Name: Pack_Size, dtype: int64

---------- NEW FAMILIES - Premium ----------

134.0 165
150.0 245
175.0 371
Name: Pack_Size, dtype: int64

---------- NEW FAMILIES - Budget ----------

134.0 309
150.0 448
175.0 763
Name: Pack_Size, dtype: int64

---------- NEW FAMILIES - Mainstream ----------

134.0 224
150.0 384
175.0 579
Name: Pack_Size, dtype: int64
---------- OLDER FAMILIES - Premium ----------

134.0 1014
150.0 1750
175.0 2747
Name: Pack_Size, dtype: int64

---------- OLDER FAMILIES - Budget ----------

134.0 1996
150.0 3708
175.0 5662
Name: Pack_Size, dtype: int64
---------- OLDER FAMILIES - Mainstream ----------

134.0 1234
150.0 2261
175.0 3489
Name: Pack_Size, dtype: int64

---------- OLDER SINGLES/COUPLES - Premium ----------

134.0 1744
150.0 2854
175.0 4382
Name: Pack_Size, dtype: int64
---------- OLDER SINGLES/COUPLES - Budget ----------

134.0 1843
150.0 2899
175.0 4535
Name: Pack_Size, dtype: int64

---------- OLDER SINGLES/COUPLES - Mainstream ----------

134.0 1720
150.0 2875
175.0 4422
Name: Pack_Size, dtype: int64

---------- RETIREES - Premium ----------

134.0 1331
150.0 2015
175.0 3232
Name: Pack_Size, dtype: int64

---------- RETIREES - Budget ----------

134.0 1517
150.0 2381
175.0 3768
Name: Pack_Size, dtype: int64

---------- RETIREES - Mainstream ----------

134.0 2103
150.0 3415
175.0 5187
Name: Pack_Size, dtype: int64
---------- YOUNG FAMILIES - Premium ----------

134.0 1007
150.0 1832
175.0 2926
Name: Pack_Size, dtype: int64

---------- YOUNG FAMILIES - Budget ----------

134.0 1674
150.0 2981
175.0 4800
Name: Pack_Size, dtype: int64
---------- YOUNG FAMILIES - Mainstream ----------

134.0 1148
150.0 2101
175.0 3087
Name: Pack_Size, dtype: int64
In [54]: (temp.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"])["PROD_QTY"].sum()
/ temp.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"])["LYLTY_CARD_NBR"].nunique()).sort_values(ascending=False)

Out[54]: LIFESTAGE PREMIUM_CUSTOMER

OLDER FAMILIES Mainstream 9.804309
Budget 9.639572
Premium 9.578091
YOUNG FAMILIES Budget 9.238486
Premium 9.209207
Mainstream 9.180352
OLDER SINGLES/COUPLES Premium 7.154947
Budget 7.145466
Mainstream 7.098783
MIDAGE SINGLES/COUPLES Mainstream 6.796108
RETIREES Budget 6.458015
Premium 6.426653
MIDAGE SINGLES/COUPLES Premium 6.386672
Budget 6.313830
RETIREES Mainstream 6.253743
NEW FAMILIES Mainstream 5.087161
Premium 5.028912
Budget 5.009892
YOUNG SINGLES/COUPLES Mainstream 4.776459
Budget 4.411485
Premium 4.402098
dtype: float64
In [55]: (temp.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"])["PROD_QTY"].sum()
/ temp.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"])["LYLTY_CARD_NBR"].nunique()).unstack().plot.bar(figsize=(15,4
​
plt.title("Average purchase quantity per segment", fontsize=20, fontweight='bold', color='darkorange')
plt.xlabel("Lifestage", fontsize=15, fontweight='bold', color='darkorange')
plt.legend(loc="center left", bbox_to_anchor=(1.0, 0.5))
plt.savefig("Average purchase quantity per segment.png", bbox_inches="tight")
plt.show()
In [56]: #Average chips price per transaction by segments
​
print("\n ----- Average chips price per transaction by segments ----- \n")
temp["Unit_Price"] = temp["TOT_SALES"] / temp["PROD_QTY"]
temp.groupby(["Segment"]).mean()["Unit_Price"].sort_values(ascending=False)

----- Average chips price per transaction by segments -----

Out[56]: Segment
YOUNG SINGLES/COUPLES - Mainstream 4.071485
MIDAGE SINGLES/COUPLES - Mainstream 4.000101
RETIREES - Budget 3.924883
RETIREES - Premium 3.921323
NEW FAMILIES - Budget 3.919251
NEW FAMILIES - Mainstream 3.916581
OLDER SINGLES/COUPLES - Premium 3.887220
OLDER SINGLES/COUPLES - Budget 3.877022
NEW FAMILIES - Premium 3.871743
RETIREES - Mainstream 3.833343
OLDER SINGLES/COUPLES - Mainstream 3.803800
YOUNG FAMILIES - Budget 3.753659
MIDAGE SINGLES/COUPLES - Premium 3.752915
YOUNG FAMILIES - Premium 3.752402
OLDER FAMILIES - Budget 3.733344
MIDAGE SINGLES/COUPLES - Budget 3.728496
OLDER FAMILIES - Mainstream 3.727383
YOUNG FAMILIES - Mainstream 3.707097
OLDER FAMILIES - Premium 3.704625
YOUNG SINGLES/COUPLES - Premium 3.645518
YOUNG SINGLES/COUPLES - Budget 3.637681
Name: Unit_Price, dtype: float64
In [57]: temp.groupby(["LIFESTAGE", "PREMIUM_CUSTOMER"]).mean()["Unit_Price"].unstack().plot.bar(figsize=(15,4), rot=0)
​
plt.xlabel("Lifestage", fontsize=15, fontweight='bold', color='darkorange')
plt.legend(loc="center left", bbox_to_anchor=(1,0.5))
plt.show()
In [58]: z = temp.groupby(["Segment", "Cleaned_Brand_Names"]).sum()["TOT_SALES"].sort_values(ascending=False).reset_index
z[z["Segment"] == "YOUNG SINGLES/COUPLES - Mainstream"]

Out[58]:
Segment Cleaned_Brand_Names TOT_SALES

0 YOUNG SINGLES/COUPLES - Mainstream Kettle 35423.6

8 YOUNG SINGLES/COUPLES - Mainstream Doritos 21705.9

23 YOUNG SINGLES/COUPLES - Mainstream Pringles 16006.2

24 YOUNG SINGLES/COUPLES - Mainstream Smiths 15265.7

55 YOUNG SINGLES/COUPLES - Mainstream Infuzions 8749.4

59 YOUNG SINGLES/COUPLES - Mainstream Old 8180.4

65 YOUNG SINGLES/COUPLES - Mainstream Twisties 7539.8

73 YOUNG SINGLES/COUPLES - Mainstream Tostitos 7238.0

74 YOUNG SINGLES/COUPLES - Mainstream Thins 7217.1

92 YOUNG SINGLES/COUPLES - Mainstream Cobs 6144.6

124 YOUNG SINGLES/COUPLES - Mainstream RRD 4958.1

129 YOUNG SINGLES/COUPLES - Mainstream Tyrrells 4800.6

148 YOUNG SINGLES/COUPLES - Mainstream Grain Waves 4201.0

189 YOUNG SINGLES/COUPLES - Mainstream Cheezels 3318.3

246 YOUNG SINGLES/COUPLES - Mainstream Natural Chip Co 2130.0

258 YOUNG SINGLES/COUPLES - Mainstream Woolworths 1929.8

318 YOUNG SINGLES/COUPLES - Mainstream Cheetos 898.8

327 YOUNG SINGLES/COUPLES - Mainstream CCs 850.5

383 YOUNG SINGLES/COUPLES - Mainstream French 429.0

393 YOUNG SINGLES/COUPLES - Mainstream Sunbites 391.0

415 YOUNG SINGLES/COUPLES - Mainstream Burger 243.8

Insights from Data :-

 Top 3 total sales contributor segment are :-

i. Older families (Budget) $156,864

ii. Young Singles/Couples (Mainstream) $147,582
iii. Retirees (Mainstream) $145,169

•♦• Young Singles/Couples (Mainstream) has the highest population, followed by Retirees (Mainstream). W
hich explains their high total sales.

•♦• Despite Older Families not having the highest population, they have the highest frequency of purcha
se, which contributes to their high total sales.

•♦• Older Families followed by Young Families has the highest average quantity of chips bought per purc
hase.

•♦• The Mainstream category of the "Young and Midage Singles/Couples" have the highest spending of chip
s per purchase. And the difference to the non-Mainstream "Young and Midage Singles/Couples" are statist
ically significant.

•♦• Chips brand Kettle is dominating every segment as the most purchased brand.

•♦• Observing the 2nd most purchased brand, "Young and Midage Singles/Couples" is the only segment with
a different preference (Doritos) as compared to others' (Smiths).

•♦• Most frequent chip size purchased is 175gr followed by the 150gr chip size for all segments.

Future Recommendations :-
 •♦• Older Families: Focus on the Budget segment. Strength: Frequent purchase. We can give promotion
s that encourages more frequency of purchase. Strength: High quantity of chips purchased per visit. We
can give promotions that encourage them to buy more quantity of chips per purchase.

•♦• Young Singles/Couples: Focus on the Mainstream segment. This segment is the only segment that h
ad Doritos as their 2nd most purchased brand (after Kettle). To specifically target this segment it mig
ht be a good idea to collaborate with Doritos merchant to do some branding promotion catered to "Young
Singles/Couples - Mainstream" segment. Strength: Population quantity. We can spend more effort on maki
ng sure our promotions reach them, and it reaches them frequently.

•♦• Retirees: Focus on the Mainstream segment. Strength: Population quantity. Again, since their po
pulation quantity is the contributor to the high total sales, we should spend more effort on making sur
e our promotions reaches as many of them as possible and frequent.

•♦• General: All segments has Kettle as the most frequently purchased brand and 175gr (regardless o
f brand) followed by 150gr as the preferred chip size. When promoting chips in general to all segments
it is good to take advantage of these two points.

•♦•♦•♦•