Jigsaw Academy tied up with a data science leader in launching a project for our students through the Great Cafe Contest. The students got a chance to work on industry dataset and a real life project. The final presentation is the answer to the business problem highlighting the analytical skills of the winners.
The overall objective of the project was to demonstrate the use of analytical tools for drawing insights from the available dataset. The learning objective was to demonstrate the use of tools like SAS/R to extract, manipulate, analyze and visualize data.
I lead a team in the intense 2015 Target Case Competition. The objective of the Target Case was to find a way to enhance Target Corporation’s grocery offerings using technology. Upon further research, we were able to identify that Target’s offerings were not perceived as fresh by their consumers. The company was interested in enticing millennials with focused product assortments but they had yet to push these offerings in a way that was compelling for their desired target market. Knowing that Target often creates special experiences within each section of their stores, it is important they carry this within their grocery sections, too. Therefore, we recommend that Target Corp. install interactive digital signage within its grocery sections. We offered them uses of the digital screens and methods to calculate ROI. We also took it a step further to calculate that the expected cost of installing such technology would be roughly $2000 per store. We projected that implementing our recommendation would highlight the freshness and sustainability of Target’s grocery offerings and its value chain. We won the first place grand prize, here is our presentation.
Created an Integrated Marketing Communications Plan with five-step funnel for the client to increase restaurant revenue and frequency based on 20K transaction and customer demographic database, using SPSS, Excel, and Tableau
• According to the surveys of urban and suburban customers, we concluded which features and the feature combination could make Dominick's become a strong player in Fresh Produce.
• Used perceptual mapping to find Dominick's market place and which factors that customers are concerning when purchasing. We found Dominick's should place itself between Trader Joe's and Farmer's Markets.
• Then we utilized SPSS to make conjoint analysis, and we found more information could help us make good feature combination.
• Combined the results of two tools, we made a feature combination that can help Dominick's get bigger market share from 32% to 46%. We provided effective recommendations.
This report covers those establishments where coffee is the primary sales item or the main selling point to attract consumers. They are based on the European and North American coffee shop models, offering a wide variety of coffee drinks, eg cappuccino, latte, mocha, etc. Other items are usually on sale, such as pastries, pasta, tea, coffee beans etc. However, the food offer may be restricted
I lead a team in the intense 2015 Target Case Competition. The objective of the Target Case was to find a way to enhance Target Corporation’s grocery offerings using technology. Upon further research, we were able to identify that Target’s offerings were not perceived as fresh by their consumers. The company was interested in enticing millennials with focused product assortments but they had yet to push these offerings in a way that was compelling for their desired target market. Knowing that Target often creates special experiences within each section of their stores, it is important they carry this within their grocery sections, too. Therefore, we recommend that Target Corp. install interactive digital signage within its grocery sections. We offered them uses of the digital screens and methods to calculate ROI. We also took it a step further to calculate that the expected cost of installing such technology would be roughly $2000 per store. We projected that implementing our recommendation would highlight the freshness and sustainability of Target’s grocery offerings and its value chain. We won the first place grand prize, here is our presentation.
Created an Integrated Marketing Communications Plan with five-step funnel for the client to increase restaurant revenue and frequency based on 20K transaction and customer demographic database, using SPSS, Excel, and Tableau
• According to the surveys of urban and suburban customers, we concluded which features and the feature combination could make Dominick's become a strong player in Fresh Produce.
• Used perceptual mapping to find Dominick's market place and which factors that customers are concerning when purchasing. We found Dominick's should place itself between Trader Joe's and Farmer's Markets.
• Then we utilized SPSS to make conjoint analysis, and we found more information could help us make good feature combination.
• Combined the results of two tools, we made a feature combination that can help Dominick's get bigger market share from 32% to 46%. We provided effective recommendations.
This report covers those establishments where coffee is the primary sales item or the main selling point to attract consumers. They are based on the European and North American coffee shop models, offering a wide variety of coffee drinks, eg cappuccino, latte, mocha, etc. Other items are usually on sale, such as pastries, pasta, tea, coffee beans etc. However, the food offer may be restricted
Creating Omnichannel Experiences with Loblaw Digital - eCommerce Toronto MeetupDemac Media
For May's eCommerce Toronto Meetup we were joined by the team at Loblaw Digital to explore how they create omnichannel experiences through their eCommerce properties.
Coffee and Ready-to-Drink Coffee in the U.S.: Retail & Food Service, 8th Edit...MarketResearch.com
Make no mistake, coffee is big business: Packaged Facts forecasts that retail and foodservice sales of coffee will top $48 billion in 2014. Of this amount, we expect $11.2 billion (or 23%) to come from retail sales and $37 billion (or 77%) to come from sales at foodservice establishments.
Mafias in training Team - Kudwi Dataset AnalysisApoorv Parmar
Team Mafias In Training consisting of Apoorv Parmar and Kushaang Deswal Presented their analysis on the Kudwi Dataset provided by the IIT Kanpur Analytics Club at their Annual Management fest Prabandhan 2017
The iKeg™ system is easy for retailers to set up and use. Simply log-in to the iOS or Android app to manage personal settings and oversee your entire inventory ordering process. With the iKeg system, the ordering process is collapsed to literally minutes compared to the hours it may have taken in the past. The iKeg app also allows you to send customizable automatic social alerts to Facebook and Twitter every time a new beer goes on tap. It also provides a number of fully-integrated tools to help promote your establishment, special event or a unique beer that you might want to highlight.
For distributors, The iKeg system requires light ERP integration using published APIs. SteadyServ’s® SaaS (Software as a Service) can integrate with your enterprise’s ERP system, IMS, or WMS systems, or stand alone as a separate system. Information is delivered in real-time to the handset. Reports can be generated on-demand or based on stated schedules.
The iKeg system is secure, reliable and highly scalable. Designed to be ‘Apple-easy’, set up and systems management is simple. The iKeg system also uses its own secure wireless backbone, which incorporates the highest level of security and ensures that distributor and retailer orders and information systems won’t be compromised. And, the iKeg system uses state-of-the-art M2M (machine-to-machine) communication technologies to ensure the highest level of inventory and order management accuracy.
As part of our Global Strategic Management (GSM) module, we were required to read through a Royco case study analyse the issues that the company was facing and perform our own analysis on the company and the industry.
From this analysis we were required to come up with recommendations to help Royco grow their business and resolve problems within the company
Creating Omnichannel Experiences with Loblaw Digital - eCommerce Toronto MeetupDemac Media
For May's eCommerce Toronto Meetup we were joined by the team at Loblaw Digital to explore how they create omnichannel experiences through their eCommerce properties.
Coffee and Ready-to-Drink Coffee in the U.S.: Retail & Food Service, 8th Edit...MarketResearch.com
Make no mistake, coffee is big business: Packaged Facts forecasts that retail and foodservice sales of coffee will top $48 billion in 2014. Of this amount, we expect $11.2 billion (or 23%) to come from retail sales and $37 billion (or 77%) to come from sales at foodservice establishments.
Mafias in training Team - Kudwi Dataset AnalysisApoorv Parmar
Team Mafias In Training consisting of Apoorv Parmar and Kushaang Deswal Presented their analysis on the Kudwi Dataset provided by the IIT Kanpur Analytics Club at their Annual Management fest Prabandhan 2017
The iKeg™ system is easy for retailers to set up and use. Simply log-in to the iOS or Android app to manage personal settings and oversee your entire inventory ordering process. With the iKeg system, the ordering process is collapsed to literally minutes compared to the hours it may have taken in the past. The iKeg app also allows you to send customizable automatic social alerts to Facebook and Twitter every time a new beer goes on tap. It also provides a number of fully-integrated tools to help promote your establishment, special event or a unique beer that you might want to highlight.
For distributors, The iKeg system requires light ERP integration using published APIs. SteadyServ’s® SaaS (Software as a Service) can integrate with your enterprise’s ERP system, IMS, or WMS systems, or stand alone as a separate system. Information is delivered in real-time to the handset. Reports can be generated on-demand or based on stated schedules.
The iKeg system is secure, reliable and highly scalable. Designed to be ‘Apple-easy’, set up and systems management is simple. The iKeg system also uses its own secure wireless backbone, which incorporates the highest level of security and ensures that distributor and retailer orders and information systems won’t be compromised. And, the iKeg system uses state-of-the-art M2M (machine-to-machine) communication technologies to ensure the highest level of inventory and order management accuracy.
As part of our Global Strategic Management (GSM) module, we were required to read through a Royco case study analyse the issues that the company was facing and perform our own analysis on the company and the industry.
From this analysis we were required to come up with recommendations to help Royco grow their business and resolve problems within the company
Applebee’s Neighborhood Grill and Bar has been in a prolonged sales slump for five consecutive quarters, which must have hit the casual dining giant hard, considering its previous five consecutive quarters of gains that ended abruptly in mid-2015.
Jigsaw Academy conducted a case study contest for its students in collaboration with Pexitics Analytics. These are the submissions by different student groups
A brief introduction to Web Analytics
For a course in Web Analytics - See more at: http://www.jigsawacademy.com/web-analytics-training#sthash.LsGc9ZIT.dpuf.
This course is for professionals interested in assessing online business performances and students interested in e-commerce marketing & sales careers.
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...pchutichetpong
M Capital Group (“MCG”) expects to see demand and the changing evolution of supply, facilitated through institutional investment rotation out of offices and into work from home (“WFH”), while the ever-expanding need for data storage as global internet usage expands, with experts predicting 5.3 billion users by 2023. These market factors will be underpinned by technological changes, such as progressing cloud services and edge sites, allowing the industry to see strong expected annual growth of 13% over the next 4 years.
Whilst competitive headwinds remain, represented through the recent second bankruptcy filing of Sungard, which blames “COVID-19 and other macroeconomic trends including delayed customer spending decisions, insourcing and reductions in IT spending, energy inflation and reduction in demand for certain services”, the industry has seen key adjustments, where MCG believes that engineering cost management and technological innovation will be paramount to success.
MCG reports that the more favorable market conditions expected over the next few years, helped by the winding down of pandemic restrictions and a hybrid working environment will be driving market momentum forward. The continuous injection of capital by alternative investment firms, as well as the growing infrastructural investment from cloud service providers and social media companies, whose revenues are expected to grow over 3.6x larger by value in 2026, will likely help propel center provision and innovation. These factors paint a promising picture for the industry players that offset rising input costs and adapt to new technologies.
According to M Capital Group: “Specifically, the long-term cost-saving opportunities available from the rise of remote managing will likely aid value growth for the industry. Through margin optimization and further availability of capital for reinvestment, strong players will maintain their competitive foothold, while weaker players exit the market to balance supply and demand.”
As Europe's leading economic powerhouse and the fourth-largest hashtag#economy globally, Germany stands at the forefront of innovation and industrial might. Renowned for its precision engineering and high-tech sectors, Germany's economic structure is heavily supported by a robust service industry, accounting for approximately 68% of its GDP. This economic clout and strategic geopolitical stance position Germany as a focal point in the global cyber threat landscape.
In the face of escalating global tensions, particularly those emanating from geopolitical disputes with nations like hashtag#Russia and hashtag#China, hashtag#Germany has witnessed a significant uptick in targeted cyber operations. Our analysis indicates a marked increase in hashtag#cyberattack sophistication aimed at critical infrastructure and key industrial sectors. These attacks range from ransomware campaigns to hashtag#AdvancedPersistentThreats (hashtag#APTs), threatening national security and business integrity.
🔑 Key findings include:
🔍 Increased frequency and complexity of cyber threats.
🔍 Escalation of state-sponsored and criminally motivated cyber operations.
🔍 Active dark web exchanges of malicious tools and tactics.
Our comprehensive report delves into these challenges, using a blend of open-source and proprietary data collection techniques. By monitoring activity on critical networks and analyzing attack patterns, our team provides a detailed overview of the threats facing German entities.
This report aims to equip stakeholders across public and private sectors with the knowledge to enhance their defensive strategies, reduce exposure to cyber risks, and reinforce Germany's resilience against cyber threats.
Techniques to optimize the pagerank algorithm usually fall in two categories. One is to try reducing the work per iteration, and the other is to try reducing the number of iterations. These goals are often at odds with one another. Skipping computation on vertices which have already converged has the potential to save iteration time. Skipping in-identical vertices, with the same in-links, helps reduce duplicate computations and thus could help reduce iteration time. Road networks often have chains which can be short-circuited before pagerank computation to improve performance. Final ranks of chain nodes can be easily calculated. This could reduce both the iteration time, and the number of iterations. If a graph has no dangling nodes, pagerank of each strongly connected component can be computed in topological order. This could help reduce the iteration time, no. of iterations, and also enable multi-iteration concurrency in pagerank computation. The combination of all of the above methods is the STICD algorithm. [sticd] For dynamic graphs, unchanged components whose ranks are unaffected can be skipped altogether.
Show drafts
volume_up
Empowering the Data Analytics Ecosystem: A Laser Focus on Value
The data analytics ecosystem thrives when every component functions at its peak, unlocking the true potential of data. Here's a laser focus on key areas for an empowered ecosystem:
1. Democratize Access, Not Data:
Granular Access Controls: Provide users with self-service tools tailored to their specific needs, preventing data overload and misuse.
Data Catalogs: Implement robust data catalogs for easy discovery and understanding of available data sources.
2. Foster Collaboration with Clear Roles:
Data Mesh Architecture: Break down data silos by creating a distributed data ownership model with clear ownership and responsibilities.
Collaborative Workspaces: Utilize interactive platforms where data scientists, analysts, and domain experts can work seamlessly together.
3. Leverage Advanced Analytics Strategically:
AI-powered Automation: Automate repetitive tasks like data cleaning and feature engineering, freeing up data talent for higher-level analysis.
Right-Tool Selection: Strategically choose the most effective advanced analytics techniques (e.g., AI, ML) based on specific business problems.
4. Prioritize Data Quality with Automation:
Automated Data Validation: Implement automated data quality checks to identify and rectify errors at the source, minimizing downstream issues.
Data Lineage Tracking: Track the flow of data throughout the ecosystem, ensuring transparency and facilitating root cause analysis for errors.
5. Cultivate a Data-Driven Mindset:
Metrics-Driven Performance Management: Align KPIs and performance metrics with data-driven insights to ensure actionable decision making.
Data Storytelling Workshops: Equip stakeholders with the skills to translate complex data findings into compelling narratives that drive action.
Benefits of a Precise Ecosystem:
Sharpened Focus: Precise access and clear roles ensure everyone works with the most relevant data, maximizing efficiency.
Actionable Insights: Strategic analytics and automated quality checks lead to more reliable and actionable data insights.
Continuous Improvement: Data-driven performance management fosters a culture of learning and continuous improvement.
Sustainable Growth: Empowered by data, organizations can make informed decisions to drive sustainable growth and innovation.
By focusing on these precise actions, organizations can create an empowered data analytics ecosystem that delivers real value by driving data-driven decisions and maximizing the return on their data investment.
Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...John Andrews
SlideShare Description for "Chatty Kathy - UNC Bootcamp Final Project Presentation"
Title: Chatty Kathy: Enhancing Physical Activity Among Older Adults
Description:
Discover how Chatty Kathy, an innovative project developed at the UNC Bootcamp, aims to tackle the challenge of low physical activity among older adults. Our AI-driven solution uses peer interaction to boost and sustain exercise levels, significantly improving health outcomes. This presentation covers our problem statement, the rationale behind Chatty Kathy, synthetic data and persona creation, model performance metrics, a visual demonstration of the project, and potential future developments. Join us for an insightful Q&A session to explore the potential of this groundbreaking project.
Project Team: Jay Requarth, Jana Avery, John Andrews, Dr. Dick Davis II, Nee Buntoum, Nam Yeongjin & Mat Nicholas
2. Café Great Contest
Quick service restaurant (cafe) sells a usual mix of Coffee, pastas, beer, and
sandwiches. Despite being in a prime location in Bangalore and having an
incredible spread of menu items, the Café' is still struggling to achieve the
sales targets since last 2 Quarters
Business Background
Objectives
• To understand the sale pattern across different categories with respect
to different time frame.
•Conduct market basket analysis to find out the potential of cross selling
and up selling
3. Food and beverages items sells more but Tobacco has higher contribution towards the total sales
Café Great Contest- Data Exploration
4. Above are top ten selling items of each category which contributes towards total sales
Café Great Contest- Top Ten items in each category
5. Top ten items across categories contributes towards total sales for each quarter
Café Great Contest- Top ten items in each quarter
6. Total sales for each weekday for each category, last three days of the week are pre-dominant in most
categories
Café Great Contest- Sales of each category on each weekday
7. Weekend sales are more as compared weekdays sales on an average which goes the Bangalore
trend.
• Weekend - Fri, Sat, Sun
• Weekday – Mon, Tue, Wed, Thu
Café Great Contest- Average sales of each category
8. • Weekday – Mon, Tue, Wed, Thu
• Weekend - Fri, Sat, Sun
Café Great Contest- Sales of each category by hour
Sales of items is increasing as day progress, Food and beverage items get decline after 8 pm
and sales of liquor product increase in the night. But there are hardly any sale in morning time.
9. • Weekday – Mon, Tue, Wed, Thu
• Weekend - Fri, Sat, Sun
Café Great Contest- Data Exploration
Sales of items is increasing as day progress, merchandise and beverage items get decline after 8
pm and sales of liquor product increase in the night.
10. CAPPUCCINO
MOCHA SHAKE
CARLSBERG
NIRVANA HOOKAH SINGLE
JR.CHL AVALANCHE
QUA MINERAL WATER(1000ML)
MINT FLAVOUR SINGLE
TUBORG
POUTINE WITH FRIES
NIRVANA HOOKAH SINGLE
CAPPUCCINO
MINT FLAVOUR SINGLE
MOCHA SHAKE
QUA MINERAL WATER(1000ML)
SAMBUCA
CAFFE LATTE
B.M.T. PANINI
POUTINE WITH FRIES
PHILLYCREAM CHEESE &CHILLY PAN
MASALA CHAI CUTTING
LEMON ICED TEA
SILVER APPLE SINGLE
JR.CHL AVALANCHE
GREEN APPLE FLAVOUR SINGLE
Café Great Contest- Top selling items in different time
Top selling items across categories during different time of the days as the day progress people consume
more food and beverages items but they also consume tobacco products.,
11. Café Great Contest- Data Exploration
NIRVANA HOOKAH SINGLE
CAPPUCCINO
MOCHA SHAKE
MINT FLAVOUR SINGLE
SAMBUCA
POUTINE WITH FRIES
MASALA CHAI CUTTING
JR.CHL AVALANCHE
QUA MINERAL WATER(1000ML)
B.M.T. PANINI
LEMON ICED TEA
N R G HOOKAH
CAFFE LATTE
CARLSBERG
OREO COOKIE SHAKE
MOROCCAN MINT TEA
BERRY BLAST
CALCUTTA MINT
NIRVANA HOOKAH SINGLE
CAPPUCCINO
MOCHA SHAKE
MINT FLAVOUR SINGLE
POUTINE WITH FRIES
CARLSBERG
SAMBUCA
TUBORG
JR.CHL AVALANCHE
B.M.T. PANINI
MASALA CHAI CUTTING
QUA MINERAL WATER(1000ML)
MOROCCAN MINT TEA
GREEN APPLE FLAVOUR SINGLE
LEMON ICED TEA
BERRY BLAST
CAFFE LATTE
NIRVANA HOOKAH SINGLE
MINT FLAVOUR SINGLE
SAMBUCA
CAPPUCCINO
MOCHA SHAKE
CARLSBERG
QUA MINERAL WATER(1000ML)
JR.CHL AVALANCHE
MOROCCAN MINT TEA
TUBORG
POUTINE WITH FRIES
GREEN APPLE FLAVOUR SINGLE
B.M.T. PANINI
Top selling items across categories during evening
time people consume more beverages items and
tobacco products. During night times people are
having along with food items, liquor and tobacco
products.
12. Café Great Contest – Market Basket Analysis Result
Most customer prefer café latte with
vanilla, Irish cream, caramel,
cinnamon and hazelnut flavor with
more 20 time than any other
customer with more than 60% of
confidence
Most customer prefer Saigon
noodles with chicken with more
950 time than any other customer
with more than 40% of confidence
Most customer prefer Kheema
ghotala with toast butter with
more 115 time than any other
customer with more than 50% of
confidence
13. Café Great Contest – Market Basket Analysis Result
Most customer prefer cappuccino
with Hazelnut or Whipped Cream
with more 5 time than any other
customer with more than 40% of
confidence
Most customer prefer Red bull with
N.R.G Hookah with more 25 time
than any other customer with more
than 80% of confidence
Most customer prefer Red bull,
Maggie noodles arrabiata with
Sambuca with more 8 time than any
other customer with more than 50%
of confidence
14. Café Great Contest – Market Basket Analysis Result
Most customer prefer fries, masala
chai with B.M.T. Panini with more 11
time than any other customer with
more than 40% of confidence
Most customer prefer B.M.T. Panini,
Red bull with Sambuca with more 7
time than any other customer with
more than 40% of confidence
Most customer prefer Vanilla ice
cream with mocha shake with more
5 time than any other customer with
more than 40% of confidence
15. Most customer prefer qua water,
fries, sambuca with B.M.T. Panini
with more 11 time than any other
customer with more than 40% of
confidence
Café Great Contest – Market Basket Analysis Result
Most customer prefer Red bull ,
mocha shake, B.M.T. Panini with
sambuca with more 10 time than any
other customer with more than 65%
of confidence
Most customer prefer qua water,
B.M.T. Panini, mocha shake with
sambuca with more 7 time than any
other customer with more than 40%
of confidence
16. Café Great Contest – Conclusion
•Tobacco category contributes highest to total sales.
• For most of categories weekend sales are high as compared to weekday sales.
• Sales are high during night time where there is hardly any sale during morning time.
• Customer buying tobacco products with foods products across the day.
• Liquor sales are high during night time but hardly any sales in first half of day.
• Food items demand are high across the day as compared to other categories.
•Merchandise sales are high during day time as well as night times.
17. Café Great Contest – Recommendation
• Give Happy hours offer in liquor category to increase the sales of the liquors during day time.
• Combine the food products with tobacco product with discounted prices.
• Combine different food products and sales them at discounted prices.
• Operate the store during morning time to increase the sales and include breakfast items.
• During weekday sales the products at discount prices.
• Increase the sales of merchandise by providing them at discount price with different food
combination or introducing buy 2 and get 1 scheme.
• Combine Hookah with food or beverage or liquor items to increase the sales
20. CONTENTS:
Products in the café
Total Sales and Billing Time
The Contribution of a Category in Quantity Sold
The Contribution of a Category in Revenue Generated
Conclusion From Monthly Pattern in Quantity sold and Revenue
Generated
Total Number of Products Sold And Revenue Generated on Each
week Day
Number of Products of specific category Sold And Revenue
Generated on Each weekday
Why Customers Visited cafe
Categorizing Bill amount
Spending Pattern in Beverage,Food,Liquor,tobacco,wine & its
combination
Products and Price range
Rate of products Vs. Number of Units sold for different products
Best Selling products & Grading products Quantity sold wise.
Best Selling products & Grading products Revenue Earned wise
Best seller and Worst seller Product wise
For Increasing Sale of Different Product
◉
◉
21. EXECUTIVE SUMMARY
This report summarizes the statistical modeling and analysis results for Bangalore based Café.
The purpose of this document is to present finding from the café’s dataset to measures various
trends in the café sales and use them to enhance the sales in coming future. Observational data is
the Café’s sales data collected for around 12 months. Dataset contains more than one lac of
billing record for 5 quarters (from Jan 2010 to Mar 2011).
Data has been analyzed to detect the different trends of sales and to get the best solution for
increasing sales. SAS, MS excel has been used to analyze the trend in the 5 quarter of data with
the help of graphs and charts. This report says there are various factors like peak hours, menu
optimization and peak months in a year through which café sales can be increased.
Bivariate Statistics used to summarize the association between total café sales with different
variables like Billing Time, Date of billing, Item Desc, Billing ID and Product category. Also data
has been analyzed to understand the menu preference from the customers.
22. The Report has been generated to analyze the Café data and come up with the best practice that can be followed to
increase the Sales in future. Despite a lot of business potential like incredible spread menu items and prime location,
Cafe is facing some issues in sales from last few quarters. So main purpose of this project was to analyze the café’s
data and to come up with best solution to improve the probability by either increasing sales or brought down cost by
improving operational efficiency.
Dataset contains 145830 observations and 10 variables stating information like bill number, date and time of billing,
total, item quantity, Item category, rate, total tax, discount value and total amount for that particular bill. Also data
contains product and category wise sold item information.
In this report, analysis has been performed in different sections and each section contains analysis on different
variables with total sales of café.
Section – 1:- Analysis has been performed to establish the relationship between time variable and total sales.
Section – 2:- Analysis has been performed to establish the relationship between category and total sales.
Section – 3:- Analysis has been performed to establish the relationship between Day of the week and total sales.
Section – 4:- Analysis has been performed to establish the relationship between Spending pattern and total sales
Section – 5:- Analysis has been performed to establish the relationship between date (month wise) and total sales.
Part 1st says there are some time slots which have more sales than others like 8 PM to 12 PM and few time slots
have almost null sales like 6AM to 8 AM. Part 2 says few category like tobacco and food are more in demand than
others. Third part tells us Saturday is the most productive day while Tuesday is the least. Fourth part explains the
spending pattern of the Customer spending and in 5th part, variable that is used against total sales is date and
according to it month like Dec, Jan are the most productive month with high sales while months like April, May and
June are with low sales. Seasonality also helped in predictive target customers are from colleges. Products analysis
has been done in last part for menu optimization.
INTRODUCTION
23. PRODUCTS IN THE CAFE
Cafe has a total of 574 number of products
As expected, the most number of products come under the category of food (35%) and least
number of products come under the combined category of Liquor & Tobacco (1%).
Category
Number of
products
Food 209 (35%)
Beverage 120 (20%)
Merchandise 87 (15%)
Tobacco 52 (9%)
Miscellaneous 46 (8%)
Wines 45 (7%)
Liquor 32 (5%)
Liquor &
Tobacco
4 (1%)
24. TOTAL SALES AND BILLING TIME
Is there any relation between café sales and time of the day?
To find the relation between time of the day and café sale, we performed the analysis on two variables from the given dataset:-
Billing Time and Total Sales
Purpose of this analysis was to understand the selling pattern at different time of the day (at what time of the day café usually has the
highest sales and at what time sales is quite slow).
For this, firstly we divided time variable into 2 hours slots (Fig 1.1) and then applied various procedures in SAS like Proc format, Proc
mean, Proc freq, Proc gplot & Proc gchart.
Division of Time Interval
1 00:00:00 to 2:00:00 12 AM to 2 AM
2 2:00:01 to 4:00:00 2 AM to 4 AM
3 4:00:01 to 6:00:00 4 AM to 6 AM
4 6:00:01 to 8:00:00 6 AM to 8 AM
5 8:00:01 to 10:00:00 8 AM to 10 AM
6 10:00:01 to 12:00:00 10 AM to 12 PM
7 12:00:01 to 14:00:00 12 PM to 2 PM
8 14:00:01 to 16:00:00 2 PM to 4 PM
9 16:00:01 to 18:00:00 4 PM to 6 PM
10 18:00:01 to 20:00:00 6 PM to 8 PM
11 20:00:01 to 22:00:00 8 PM to 10 PM
12 22:00:01 to 00:00:00 10 PM to 12 AM
Fig 1.1
25. Time Interval Number of billings Percent (%)
12 AM to 2 AM (1) 15857 10.94
2 AM to 4 AM (2) 137 0.09
4 AM to 6 AM(3) 65 0.04
6 AM to 8 AM(4) 1 0.00
8 AM to 10 AM(5) 10 0.01
10 AM to 12 PM(6) 3358 2.32
12 PM to 2 PM (7) 10974 7.57
2 PM to 4 PM (8) 16412 11.32
4 PM to 6 PM (9) 22129 15.26
6 PM to 8 PM (10) 25674 17.71
8 PM to 10 PM (11) 25215 17.39
10 PM to 12 AM(12) 25146 17.34
Simple Statistics
Variable N Mean Std Dev Sum Minimum Maximum Label
Time
Interva
l
14497
8
8.826 3.176
127966
0
1 12
Total
14574
3
224.912 164.928 3.3E+07 0.01 14231
Tota
l
Pearson Correlation Coefficients, N =
144919 Prob > |r| under H0: Rho=0
Total
Time interval -0.0134
<.0001
Bivariate Statistics
To understand the relation of time period with sales, we performed analysis on two variables, time interval (independent variable) and Total Sales
(dependent Variable).
Pearson correlation has been used to understand the relation of two variables (time interval has used as quantitative variable with value of 1 to
12).
Univariate Statistics
Here we have analyzed frequency of occurrence of each point of time in the divided 2 hours of time slot.
Table (Fig 1.2 ) and Chart (Fig1.3) are showing the number of billing in its corresponding time interval. Time interval 6 PM to 8 PM has
maximum number of billing across the day while 6 AM to 8 AM has the almost no billing (no item sold). It shows no customer shows up at this
time of period.
Fig 1.2 Fig 1.3
26. Correlation coefficient value is negative shows the relation is not linear as time interval in actual is a time variable and there is no
such thing like sales is increasing when time interval is increasing. However p value (<0.0001) is highly significant (more than
0.05) shows that both variables are dependent on each other (99.99% confidence).
Graph (Fig 1.6) below is to show the total sales for each 2 hours slot of time
According to data 10 PM to 12 AM is the peak time for café sale. Time between 6 PM to 12 PM is the good time for selling, 12 AM
to 2 AM has average sale but time interval 10 AM to 12 PM, 12 PM to 2 PM has less sales. From 2 AM to 10 AM sales are very
less.
Fig
1.6
27. Conclusion and Suggestions (Based on Time Variable)
Happy hours scheme can be applied for less
productive hours
• For less productive time slots like 10AM to 2PM and 12 PM to
2 PM, happy hours scheme (discount on items or buy one get
one free) can be applied to attract the customers.
Staff management according to the productivity of the
day.
• Highly efficient staff should be applied on peak hours so that
customers can be easily handled, get satisfied and order
fulfillment
• New employees or average efficient staff on less productive
hours.
• Number of staff at a time (day and night) should also be
applied on the basis of less and high sales time slots as idle
workers are cost waste of money.
28. ◉Is there any relation between category and total sales?
Purpose : To understand the selling pattern category wise.
TOTAL SALES AND PRODUCT CATEGORY
31%
38%
6%
0%
1%
23%
1% 0%
Quantity Sold
19%
30%
4%0%
4%
42%
1% 0%
Revenue Generated
CATEGORY QUANTITY SOLD REVENUE GENERATED
BEVERAGE 31.02% 20.49%
FOOD 38.05% 31.54%
LIQUOR 6.27% 4.42%
MERCHANDISE 0.33% 0.42%
MISCELLANEOUS 0.85% 3.86%
TOBACCO 22.84% 45.09%
WINES 0.60% 0.88%
LIQUOR &TOBACCO 0.04% 0.12%
• Food beverages are the highest quantity sold categories and despite
of it Tobacco has maximum selling across all categories more than
40 %
• Second highest selling category is Food share of more than 30%.
Food and tobacco together have more 70% of the café sales.
• Wine, MISC, Merchandise are the least selling categories with share
of around 2% overall.
29. To understand the further role of category in sales, we performed the analysis, based on category sales in each Time slot.
Table of Time Interval by Category
Timeinterval Category
TotalBEVERAGE FOOD LIQUOR LIQUOR & TOBACCO
MERCHA
NDISE MISC TOBACCO WINES
10 AM to 12 PM 153004.32 204222.18 7361.79 4258.66 1596.37 240817.8 220.5 611481.62
10 PM to 12 AM 867937.76 1798479.57 523904.63 20432.75 27800.2 40290 2646887.16 104965.26 6030697.3
12 AM to 2 AM 605752.32 1064520.61 386913.27 1976.25 13576.2 27469.1 1712787.06 81466.19 3894460.98
12 PM to 2 PM 418214.12 693030.32 60682.09 984.38 11669.1 11493.6 1093746.82 3622.5 2293442.95
2 AM to 4 AM 6069.96 10593.1 4068.75 18326.3 3064.8 4536 46658.88
2 PM to 4 PM 621608.91 1149455.16 116553.26 2636.25 9105.69 14849 1605000.6 14017.76 3533226.62
4 AM to 6 AM 1707.76 3755.85 1312.5 12313.9 1022.4 20112.41
4 PM to 6 PM 875943.05 1418866.55 208461.65 2964.38 11553.2 20756 2437504.2 23121.76 4999170.78
6 AM to 8 AM 105 105
6 PM to 8 PM 1035311.84 1730495.11 312019.33 6975.01 19362.7 35670.9 2439138.9 33707.52 5612681.31
8 AM to 10 AM 1237.53 136.13 1373.66
8 PM to 10 PM 897525.54 1820323.42 460761.71 6592.5 21604.2 59653.4 2271361.2 68480.13 5606302.05
Total 5484313.11 9893878 2082038.98 42561.52 118930 242419 14451435.94 334137.62 32649713.56
• Tobacco is the highest selling category for each time slot and
food is the second highest. It means there is no time
dependency on category sale.
• Buyer for each category are visiting in all the slots except 6AM
to 8 AM when there is no buyer and 8 AM to 10 AM when there
is no buyer for liquor and tobacco related categories.
30. MOST PREFERRED CATEGORY BY CUSTOMERS
Category of
Products
Purchased
No. of
Customers
Percentage of
Customers
BEVERAGE 3901150 16.62%
FOOD 7400515.55 31.52%
LIQUOR 1039137 4.43%
LIQUOR
&TOBACCO
27800 0.12%
MERCHAND 97776.5 0.42%
MISC 134333 0.57%
TOBACCO 10672792.2 45.46%
WINES 205721 0.88%
On the basis of Analysis of bills, we observed that :-
Maximum Customers (45%) visit cafe for purchasing Tobacco products.
Food and Beverage products are purchased secondly by customers.
Only 4% of customers purchase Liquor products.
Very less customers buy Merchandise.
To understand the customer preferred category, we analyzed the bill ID Variable with Total Sales.
31. Inventory Management on the basis of category demand
• After knowing the category in demand inventory can be easily
maintained according to demand. Items should be stocked as per the
demand (more inventory for category that is high in demand) like
tobacco and Food.
• Most preferred category by customers Tobacco should have good
inventory.
Large Menu for highly demand categories
• Adding more items in Categories like tobacco and Food will increase
more options for category of his choice. Also helps in attracting more
customers.
Conclusion and Suggestions (Based on Category)
32. • Is any day of the week has more sales than other?
Purpose: To understand the sales pattern on each day of the week.
Total quantity sold and total sales on each day of week .
• Saturday is the highest quantity of product sold and revenue generated day of the Week.
• Least productive day is Tuesday with lowest number of products sold and lowest revenue generated.
TOTAL SALES ON WEEKLY BASIS
33. As per the data shown in graphs, conclusion says Product of all category has same pattern for sale and quantity sold
like total café sales for each day of week (Sunday to Saturday).
• Highest number of products sold and Higest revenue generated on Saturday.
• Lowest number of Products sold and Lowest revenue Generated on Tuesday.
To understand further , we analyzed the total sales data category wise for each day of the week
Total sales and Product sold for each day of week on category basis.
34. • Deferent Pattern is observed in Merchandise products.
• Number of Product sold is maximum on Tuesdays , Minimum on Thursday.
• Revenue generated is maximum on Monday and Minimum on Thursday.
Since Minimum sale of Merchandise observed on Thursday it is Recommended to use strategy or to
give discount or combo etc. On merchandise on Thursday to increase sale on that day
35. Discounts and others for less productive days.
• Many offers like happy hours and discounts can be offered on the less
productive days like Tuesday to attract more customers.
Staff Management on the basis more and less busy days
• More and efficient staff on busy days like Saturday, so that heavy
customer crowd can be easily managed
• Operational cost can be reduced by using new and less number of staff
on less productive days like Tuesdays
Conclusion and Suggestions (Based on Day of the
Week)
36. BILLING INFO AND TOTAL SALES
Categor
y
Bill Amount
Luxury More than 2000
High 800 to 2000
Medium 300 to 800
Low 100 to 300
Very low Less than 100
Spend Frequency Percent
High 7911 11.37
Low 20144 28.94
Luxury 394 0.57
Medium 38269 54.98
Very low 2887 4.15
• Total 69605 Bills were generated during April 2010 – March 2011
• Customers spent Rs. 469 on an average. Maximum Bill Amount was Rs.14231.25, Minimum was
Rs.18.56.
• Maximum customer spends between 300 to 800. There are only 0.57% ie. 394 customers who
spent more than 2000.
• Total sales generated from each bill.
Purpose: To understand the billing pattern for total sales.
37. • 30626 customers purchased Beverage in the period.
• Average amount spent on Beverage is Rs. 127.
• Maximum Bill of beverage is Rs. 1095 Minimum 20.
• Maximum customers Buy total Beverages of cost between 50 to
Rs.120.
CATEGORY WISE SPENDING DISTRIBUTION
To understand the billing pattern for total sales more, we bifurcated the billing sales data category wise.
• 32306 customers purchased food in the last 12 months.
• Average Bill on the Food is Rs. 229.
• Maximum Rs.2095 was spent on food.
• Maximum customer buy food worth Rs. 100 to 230.
• Very less customers spent more than Rs.700 on food.
38. • 33537 customers bought tobacco products.
• Average Bill of Tobacco product is Rs.318.23.
• Maximum Bill Rs.1950 and Minimum Rs.100.
• Maximum customers buy Tobacco products worth between
Rs.225 to Rs.320.
• Very less customers spent More than Rs.750 of Tobacco.
• 5590 customers had Liquor last year.
• Average Bill of Liquor was Rs.186.
• Maximum Bill of Liquor was Rs.1950 and Minimum Rs.100.
•
• Maximum customer spend between Rs. 100 to 180 on Liquor.
• There are only 55 Customers who spent more than 650 Rs on
Liqiour.
39. • 671 customers had Wine last year.
• Average Bill of Wine was Rs.306.
• Maximum customer spend between Rs. 300 to 900 on Wine.
• There are only 9.6% Customers who spent more than Rs. 900 on
Liqiour.
• 366 customers bought Merchandise.
• Avarage spend on merchandise was Rs.267.
• Maximum customer spend between 300 to 900 Rs on
Merchandise
• There are only 9.6% Customers who spent more than 900 Rs on
Merchandise.
• Very less customers spend less than Rs. 60 on Merchandise.
40. SPENDING PATTERN ON COMBINATION
Observations:-
◉After looking at statistics of the Bills produced we found that:-
◉136 customers bought products from all Food Beverage Merchandise and Tobacco catagory.
◉4826 Customers bought Products from all Food, Beverage and Tobacco.
◉15737 Customers spent on both Food and Beverages .
◉11665 Customers spent on both Food and Tobacco.
◉223 on Liquor and Wine.
41. Loyalty programs for the customers.
• Customer who spend good can be given points on there bill amount or
coupons which can be redeem in future to make sure such customers
visit the cafe more often. Also such customers can be targeted for
Loyalty programs.
• Customer spending upper limit of any category like beverages, tobacco
etc. can be given points that can be redeemed in future. It will help in
retaining good customers.
Combo packs of Products from different categories as per the customer
choices.• We can combine products from Food, Beverage, Merchandise and Tobacco to form a
Combo pack And also Liquor and Wine Products in Combo pack.
Conclusion and Suggestions (Based on Bill Information)
42. ◉Relation between Menu Items (products) and Total sales
In this section, analysis has been performed to find the top selling items of café and menu optimization.
Purpose of this analysis is to understand the products selling pattern and what best can be done with menu to
maximize the sales and profit.
Excel pivot table and chart has been used to analyze the product selling data.
Top 10 products on the basis of their total sales
Top 10 most selling products
POUTINE WITH FRIES
SILVER APPLE SINGLE
JR.CHL AVALANCHE
GREEN APPLE FLAVOUR SINGLE
MOCHA SHAKE
N R G HOOKAH
CALCUTTA MINT
MINT FLAVOUR SINGLE
SAMBUCA
NIRVANA HOOKAH DOUBLE
PRODUCTS AND TOTAL SALES
43. Top 10 products (fig 4.1 and 4.2) are café’s best seller products. Overall share of these
products in total sales is more than 40%. All these products are from Tobacco and Food
categories.
Top 5 Products from each categoryTOBACCO FOOD BEVERAGES LIQUOR WINES Merchandise MISC
Products
% of category
sales Products
% of category
sales Products
% of category
sales Products
% of
category
sales Products
% of category
sales Products
% of
categor
y sales
Product
s
% of
categor
y sales
N R G HOOKAH 8.24%B.M.T. PANINI 3.93%BERRY BLAST 4.65%
KF DRAUGHT
(1/2LTR) 11.73%
VLN CHENIN
BLANC (GLS) 6.64%CH TINS 4.89%
PLAIN
JANE
(STRAW
BERRY) 6.27%
CALCUTTA MINT 11.35%
OREO COOKIE
SHAKE 4.08%LEMON ICED TEA 5.01%
KF DRAUGHT
(1LTR) 15.32%
RED SANGRIA
(GLS) 7.23%
CH
COFFEE
MUGS 5.06%
ROSE
FLAVOR
S
SINGLE 6.62%
MINT FLAVOUR
SINGLE 11.81%
POUTINE WITH
FRIES 5.83%
RED BULL
ENERGY DRINK 6.72%
KF DRAUGHT
PITCHER (2LTR) 15.37%
SULA BLUSH
ZINFANDEL(GL
S) 7.97%
FLAVOU
R 500
GMS 5.33%
PARTY
CHARGE
S @
500/- 7.21%
SAMBUCA 15.85%
JR.CHL
AVALANCHE 7.17%RED BULL 2+1 6.82%TUBORG 17.64%
RED SANGRIA
(CARAFE) 8.52%SANDASS 7.82%
PLAIN
JANE
(CHOCO
LATE) 11.41%
NIRVANA HOOKAH
DOUBLE 20.42%MOCHA SHAKE 8.47%CAPPUCCINO 9.83%CARLSBERG 25.04%
VLN CAB SAUV
(GLS) 13.94%
MOCHA
T-SHIRTS 11.17%
RED
BULL
SHEESH
A 30.99%
Rest of products 32.33%Rest of products 70.53%Rest of products 66.97%Rest of products 14.90%Rest of products 55.69%
Rest of
products 65.74%
Rest of
product
s 37.51%
44. PRODUCTS AND PRICE RANGE
Price Range
Number OF
products
Product %
150 to 230 202 24.57
230 to 370 113 13.75
370 to 600 46 5.60
60 to 150 337 41.00
600 to 1200 33 4.01
less 60 83 10.10
more 1200 8 0.97
• Price of product sold in cafe varies from 20 to 2100.
• Average price of product is 211 Rs.
• Maximum products (total 337, 41%) are of price range 60 -150.
• Cafe have less number of Expensive products having price range more than 2100.
• Above charts shows Number and percentage of products of various price range.
45. RATE OF PRODUCTS VS NUMBER OF UNITS SOLD OF TOBACCO PRODUCTS & FOOD PRODUCTS
• Rate of Tobacco products varies from 73 to 500
• Avarage rate of tobacco products is Rs. 247.
• Maximum Mumber of customers purchase
tobacco product near to mean price.
• Rate of Food products varies widely from Rs. 15 to
1700.
• Maximum Mumber of customers purchase Food
product near to mean price.
46. RATE OF PRODUCTS VS NUMBER OF UNITS SOLD OF BEVERAGE PRODUCTS & FOOD PRODUCTS
• Rate of Bevarage products varies from 20 to 450
• Avarage rate of tobacco products is 135 Rs.
• Maximum Mumber of customers purchase tobacco
product of lower price range.
• Rate of Liquor products varies widely from 100 to 1300
Rs.
• Avarage rate of Food products is 463.52 Rs.
• Maximum Mumber of customers purchase Liquor product
of lower price range.
47. RATE OF PRODUCTS VS NUMBER OF UNITS SOLD OF WINE & MERCHANDISE PRODUCTS
• Rate of Tobacco products varies from 125 to 2100
• Avarage rate of Wine products is 833 Rs.
• Maximum Mumber of customers purchase Wine
product of lower price range.
• Rate of Merchandise products varies widely from
12 to 1470 Rs.
• Avarage rate of Food products is 254 Rs.
• Customer purchase more merchandise of lower
and middle price range
48. Table and graphs are showing the top 5 products from each category with percent of their
share in category they belongs.
Fig 4.4
49. Products with less sales but high unit price
These are the products which have less Sales but high unit price(Rate) in each category.
WINES
2 OCEAN PINOTAGE (BTL)
GOSSIPS CHARD AUS (BTL)
MATEUS ROSE PORTUGAL(BTL)
MAISON PIERRE SAUV MARSAN
B1G1 4SEASON CLAS
SAUV(BTL)
LIQUOR
STELLA 1LTR 2+1
BROOKLYN BUCKET - 4
SCHNEIDER BUCKET - 6
UNLIMITED BEER
WHITE SANGRIA (CARAFE)
FOOD
SCHNEIDER 2+1
VEGETABLE PASTA
3COURSE NON-VEG MEAL
J.PCHENET SPARKLING ROSE
(BTL)
ANDALUSIAN SAUSAGES
MERCHANDISE
BODHI PLANTER CUM CANDLE
STAND
FLAVOUR 500 GMS
FLAVOR 1000 GMS
FLAVOR 250 GMS
AVALANCHE BOWL
TOBACCO
SPICE SHEESHA
BLUE LAGOON SHEESHA
LATE HARVEST SULA CHENIN
(BTL)
VALENTINE SPECIAL SHEESHA
BEVERAGES
WHAT A MELON
N R G HOOKAH
MISC
HOEGAARDEN GLS (2+1)
HOEGAARDEN LTR MUGS (2+1)
50. Products with less sales and less unit price
These are the products which have less Sales and less unit price(Rate) in each category.
WINES
MANDALA VALLEY CHENIN
BLANC(GL
B1G1 4SEASON CLAS SAUV(GLS)
B1G1 4SEASON CLAS
SYRAH(GLS)
MANDALA VALLEY RED
ZINFANDEL(G
4 SEASONS CLAS SAUV(GLS
LIQUOR
WHISKEY (SM)
STELLA ARTOIS
SCHNEIDER WEISSE
WHITE RUM (SM)
BROOKLYN
FOOD
ADD BUTTERED TOAST
MUSHROOM & CORN
CHICKEN HAM
SUNNY SIDEUP + BEVERAGE
TOAST CIABATA
MERCHANDISE
CUTTING GLASS
DIP BOWL
MUGS - PLAIN COLOUR
CH WRAPPING PAPER
MOCHA MUG SINGLE
TOBACCO
GOLD FLAKE ULTRA
LIGHTS(20)
CLASSIC REGULAR
INDIA KINGS OCEAN BLUE
CLASSIC MENTHOL RUSH
GOLD FLAKE LIGHTS-BIG
BEVERAGES
MOCAFE HOT CHOCOLATE(SF)
DECAFFINATE COFFEE FRAPPE
BOTTLED WATER (1LITRE)
NEW ORLEANS BLUE (REG)
2 AXE TWIST
MISC
ADD GROUND MEAT
ADD CHICKEN BACON
51. Products with high sales and high unit price
These are the products which have high Sales and high unit price(Rate) in each category.
WINES
RED SANGRIA (CARAFE)
SULA BRUT (BTL)
SULA BLUSH ZINFANDEL (BTL)
SANGRIA ROSE (CARAFE)
VLN CHENIN BLANC (BTL)
LIQUOR
HOEGAARDEN MUG (1 LITRE)
BEER TANK 3.5 LITRE
STELLA ARTOIS MUG (1 LTR)
BEER HOOKAH
BEER TANK 3.5 LITRE
FOOD
CHEESE FONDUE
LINDT CHOCOLATE SHAKE
CHOCOLATE FONDUE
SR.CHL AVALANCHE
JR.CHL AVALANCHE
MERCHANDISE
SANDASS
MOCHA T-SHIRTS
TOBACCO
N R G HOOKAH
CALCUTTA MINT
SAMBUCA
ARABIAN MIST
NIRVANA HOOKAH DOUBLE
BEVERAGES
RED BULL 3+2
GRENADINE
3 RED BULL
RED BULL 2+1
MISC
RED BULL SHEESHA
PARTY CHARGES @ 500/-
52. Products with high sales and less unit price
These are the products which have high Sales but less unit price(Rate) in each category.
WINES
VLN CAB SAUV (GLS)
VLN SAUV BLANC (GLS)
VLN CHENIN BLANC (GLS)
RED SANGRIA (GLS)
SULA BLUSH ZINFANDEL(GLS)
LIQUOR
TUBORG
CARLSBERG
KF DRAUGHT (1/2LTR)
BUDWEISER
1+1 KF 1/2 LITER
FOOD
GOOEY CHOCOLATE FUDGE
BUN MASKA & CHAI
SCRAMBLED EGGS
CLASSIC BELGIAN WAFFLE
MERCHANDISE
MUGS - PATTERN
CH NOTE BOOKS
TOBACCO
AL SIKANDARI HOOKAH
DOUBLE
SPICE SHEESHA
BLUE LAGOON SHEESHA
LATE HARVEST SULA CHENIN
(BTL)
AL SIKANDARI HOOKAH
DOUBLE
BEVERAGES
CAPPUCCINO
QUA MINERAL
WATER(1000ML)
MASALA CHAI CUTTING
CAFFE LATTE
MOROCCAN MINT TEA
MISC
ADD ON S
ADD HERB ROAST CHICKEN
53. Grading products Quantity sold wise.
Grade(Quantity
sold wise)
Quantity sold
best seller More than 624
units
more than avg 284 To 625 units
need to improve(
Medium)
174 To 285 units
very less sold 11 To41 units
Rarely sold Less than 10 unit
• We grade products according to No of units sold in Different grades as given in the table.
• All the grades are provided in document best_sell.xlsx for reference.
• We have 59 products having no of units sold more than 624 graded as ‘best seller’.
• This 59 products have mejority (79% ) share in total quantity of all products sold.
54. Grading products revenue Earned wise.
Grade (Revenue
wise)
Revenue Earned
(Rs)
best More than 176310
Vgood (very good) 79150 To 176311
Good 40975 to 79151
NeedToImprove 19550 To 40976
less 4950 To 19551
bad 360 to 1051
Barely earned Leess than 360
• We grade products according to No of units sold in Different grades as given in the table.
• All the grades are provided in document best_sell.xlsx for reference.
• There are 30 products earning revenue more than 176320 Rs individually and having huge ( 61
%) share in the revenue earned.
55. FOR INCREASING SALE OF DIFFERENT PRODUCTQuantity Sold
Grade
Revenue Earned
Grade
Strategy to be used
Best Seller Best • Don’t let supply to fall short of
Best Seller
Vgood,
Good,
Need to Improve
• See if the products can be sale for slightly higher cost( as quantity sold is
good but lesser revenue earned)
More than Avg.
Best ,
Vgood
• Suggest customers this products when they don’t ask for to up sell(since
quantity sold is lesser compared to revenue)
More than Avg. Need to improve
• See if this products can be included in happy hrs of other discounts
• Since are less recognize products suggest customer to buy them to upsell
• Spend on advertising them as customer can recognize them more.
Less sold,
very less sold,
need to improve
Less • Combo with better selling products, so more quality can be sold.
Less sold,
very less sold
Vless,
Bad
• Try to upsell with recommending
• Include in combos
• Try to create recognition
Rarely sold Barely earned • Think if can be removed from shelf to make place better for selling
products.
56. Marketing using TOP seller products
• Top 10 seller products are the star products of the café and that can
be used for café marketing.
• Top 5 seller products category wise can be used for campaigning in
the time of high demand of their category to attract customers
Menu Optimization can be done using product rate and popularity
• Products with less sales and high rate should be promoted by
employers and feedback should be collected on them as they are most
profitable products.
• Products with less sales and less rate can be omitted from the menu
as they are less popular and less profitable also.
• Products with high sales and high rate are the star of the menu. They
always be kept on top of the menu as they give more profit and also in
demand and always be used for upselling.
• Products with high sales and less rate can be offered in the
combination with other products to make them more profitable. As
they are in demand, can be useful to sell other non demanded
products
Conclusion and Suggestions (Based on Products
information)
57. ◉Is there any relation between café sales and month of the year?
In this section we have analyzed the months and total sales.
Purpose: To find is there any seasonality in the sales data. Is there any specific time of the year when café has more sales
than other months.
Table (Fig 3.1) and chart (Fig 3.2) used to represent the sales pattern for each of the 12 months in the given dataset (months
with high sales, average sales and very less sales).
There is a pattern in sales where sales is going down while approaching high summers. December (end of the year) has
high sales than other months while months like April, May and June have low sales.
Month Month Sales
Jan 3160894.61
Feb 2839305.01
Mar 2801709.79
Apl 2402157.76
May 2360411.46
Jun 2354925.93
Jul 2734469.03
Aug 2833205.87
Sep 2498613.01
Oct 2704746.52
Nov 2654877.1
Dec 3473934.35
Fig 3.1
TOTAL SALES AND MONTH OF THE YEAR
Fig 3.2
58. MONTH BEVERAGE FOOD LIQUOR LIQUOR & TOBACCO MERCHANDISE MISC TOBACCO WINES
Jan 557069.69 955174.1 175810.3 9374.2 15771.96 1422998 24696
Feb 472730.14 830656.9 179317 8268.77 16047.52 1308124 24160.5
Mar 468938.01 799857.2 214613.4 34292.75 16470.52 2625 1234956 29956.5
Apr 380017.83 765268.1 183952.2 10581.66 23066.89 1006291 32980.5
May 368945.81 739285 144074.4 10677.84 10721.76 1057238 29468.25
Jun 357034.95 773830.4 141781.5 11288.65 13347.9 1031435 26208
Jul 430847.82 907434.8 158181.3 7240.33 15799.68 1198491 16474.5
Aug 461703.5 926697.5 152906.3 9801.7 15049.15 1230823 36225
Sep 391027 774512.7 142480.9 9732.19 10372.69 1146639 23848.97
Oct 477216.82 794462.1 163360.1 5364.96 11783.2 1231897 20661.9
Nov 512416.21 720907.2 170846.1 6692.27 6845.57 1208473 28696.5
Dec 631022.85 964722.1 274573.6 6802.18 117034.8 1423520 56259
To understand the sales pattern more we have further bifurcated data into category sales.
• There is a decline of each category during April, May, June
which could be because of summer vacation and increase in
sales after that in month of July and August. Also tobacco
can only be used by 18+ age, so both the points shows our
majority of daily consumers are from colleges.
• Tobacco, Liquor and MISC major sale is in month of December
and that could be because of the Christmas and New Year time.
59. Dataset contains sales from Jan 2010 to Mar 2011, so to understand the trend we performed the time series
analysis on excel using pivot table and chart.
Table and graph are describing the time series pattern for the sales data through out the 5 quarters (Jan 2010 to
Mar 2011). Fig 3.5 table is showing the sales value for each months since Jan 2010 to Mar 2011 and Fig 3.6
graph is to show the sales pattern of this time.
Year Quarter Month(t) Sales
2010
First 1 50152.28
Second
4 2402157.76
5 2360411.46
6 2354925.93
Third
7 2734469.03
8 2833205.87
9 2498613.01
Fourth
10 2704746.52
11 2654877.1
12 3473934.35
2011 First
1 3110742.33
2 2839305.01
3 2801709.79
60. Inventory Management based on Seasonal demand
• More stock for the high sales months for fulfillment of the customer
demand while less inventory for the low sales months to minimize the
inventory cost. Category based Inventory management (more product
for the category more in demand) for seasonal demand will helps in
improving sales. E.g more liquor for month of December
Focusing on target customer and Improvising Menu Items for Categories
based on Seasonality
• After analyzing seasonal trend and category demand, it seems our
target customers are from College. So menu should be customized
considering target customer like variety of menu items, variety of
flavors. Various marketing strategies like discount coupons for non
productive hours can be applied. Social media marketing also works
wonder to attract these customers.
• Increasing Menu size with top selling items in market for categories
more in demand in peak sales months will attract more customers by
providing more and best choices.
Conclusion and Suggestions (Based on Month of the
Year)
61. MODEL – LINEAR REGRESSION FOR RESPONSE
VARIABLE (TOTAL)
◉To increase Profit of the cafe we need to increase Amount (Bill) payed by customer to the cafe ie. Total.
◉Hence we need to know How diffrent factors could possibly affect the amount in the bill ie. Total.
◉Hence Linear regression model built with Response verible total.
◉ Response variable : - Total
◉Independent variable:-
◉Quantity( Number of units of product bught by customer)
◉Rate( cost of the product bought)
◉Tax
◉Category of product ( Dummy verible of each catagory )
◉Aim :- Is to see how above all parameters affect the Total and hence the Bill amount and hence profit
62. ◉Variables used:-
1.Total ( DV):- amount to pay by customer afeter purchase of perticular product.
1.Quantity(Iv):- Units of that products purchased by the customer.
2.Rate (Iv) :- Rate of the Product.
3.Tax (Iv):- Total tax to pay on purchase of that product.
Some veriables was generated.( Dummy for diffrent catagory of products)
1.bev_dum :- If product belong to Bevorage catagory (if yes 1 else 0)
2.liq_dum :- If product belong to Liquor catagory (if yes 1 else 0)
3.food_dum :-If product belong to Food catagory (if yes 1 else 0)
4.mer_dum :-If product belong to Merchandise catagory (if yes 1 else 0)
5.misc_dum :- If product belong to Miscellaneous catagory (if yes 1 else 0)
6.Tob_dum :-If product belong to Tobacco catagory (if yes 1 else 0)
7.Wine_dum :- If product belong to Wine catagory (if yes 1 else 0)
63. ResultsProducedand interpretation:-
Root MSE 25.295814 R-Square 0.976755
Dependent Mean 225.107612 Adj R-Sq 0.976754
Coeff Var 11.237210
Parameter Estimates
Variable Label DF Parameter
Estimate
Standard
Error
t
Value
Pr >
|t|
Variance
Inflation
Intercept Intercept 1 -78.66695 3.34141 -23.54 <.0001 0
Quantity Quantity 1 17.54039 0.22898 76.60 <.0001 2.70999
Rate Rate 1 0.12306 0.00199 61.92 <.0001 9.35889
Tax Tax 1 4.12631 0.00541 762.83 <.0001 10.76114
bev_dum 1 72.36280 3.32912 21.74 <.0001 527.42990
liq_dum 1 35.28956 3.32903 10.60 <.0001 102.00447
food_dum 1 79.85293 3.32391 24.02 <.0001 597.27431
mer_dum 1 186.84067 3.52500 53.00 <.0001 9.30948
misc_dum 1 91.12961 3.40220 26.79 <.0001 21.29648
Tob_dum 1 24.60968 3.30555 7.44 <.0001 464.97571
Wine_dum 1 86.37968 3.43713 25.13 <.0001 13.30034
All variables are significant.
Rate:- Rate of the product is directly praportional to
the Total.
As rate of the product increse by 1 unit(keeing all
other values constant) the Total increases by 0.123
units.
Hence by sell of Higher range products chance of
erning higher revenue are increased.
Tax:- Is directly praportional to the total
As tax increase by 1 unit Total increases by 4.12
units( keepeng all other values constant)
64. Quantity :- Quantity of the product purchased is directly praportional to bill generated.
If Quantity of product bought increases by 1 unit (other constant) total increases by 17.54 units .
Hence cafe should build startegy to sell more units of the product
Bev dum:- If customer purchases product from Beverage catagory Total increases by 72.36
Food_dum:- If customer purchases product from Food catagory Total increases by 79.85
Wine_dum:- If customer purchases product from Wine catagory Total increases by 86.37
Mer_dum:- If customer purchases product from Merchandise catagory Total increases by 186.86 .
Misc _dum:- If customer purchases product from Miscellaneous catagory Total increases by 91.12 .
Liq_dum:-If customer purchases product from Liquor catagory Total increases by 35.28.
Tob_dum:- If customer purchases product from Tobacoo catagory Total increases by 24.60
Hence we should sale more products of on Merchendise,Wine,Beverage and Food category
Merchandise Products will lead to the maximum revenue increase.
65. Our model guesses the value of total corect most of the times with good R2 of 97%
Hence Total can be predicted as
Predicted Total
=(17.54*Quantity)+(0.12*Rate)+(4.12*Tax)+(75.36*Bev_dum)+(35.28*liq_Dum)+(79.85*food_dum)+(189.84*
mer_dum)+(91.12*Misc_dum)+(24.60*Tobb_Dum)+(86.37*winedum)-78.66
66. Conclusion
Café dataset has been used to analysis sales pattern from different aspects to give best suggestion for future sales
enhancement. We analyzed different variables like time, product category, date, product rate to understand the total sale
variable dependency and factor influencing it. So, after the completion of this analysis report we drawn the conclusion that
Total Sales variable has dependent upon Time duration in a day, Category of the Product, Seasonality (summer/winter)
and Specific product brand.
Recommendations for sales improvement from the report are:
Inventory Management can be done on the basis of seasonality (More in peak season time like December and less in
low sales time like April, May and June) and Product category (like more for tobacco that has more sales than others).
More customers are expected on Saturdays and less on Tuesdays hence to handle more customers on Saturday more staff
can be assigned, On Tuesdays discount can be given to attract more customers.
Since Minimum sale of Merchandise observed on Thursday it is Recommended to use strategy or to give discount or
combo etc. On merchandise on Thursday to increase sale on that day
Bigger Menu for category more in demand. Menu optimization can be performed with the help of high and less sales/
rate products.
Happy hours scheme can be applied on the less productive time interval ie .4 Am- 8 Am to attract customers and also
staff can be management by the knowing the high and low sales time interval(more and efficient staff on peak hours while
new or average and less staff during low sales time periods.
67. Conclusioncontinued
Marketing strategy like coupons for non productive hours and loyalty program can be run to retain the target
customers Social media is a good medium to attract our young target crowd.
Top selling products can be used as star products to promote café sales
Products not recognized by customers can be suggested for upsell.
Products not recognized ie. bought less by customers can be suggested for upsell.
Since Tobacco is most favorite category of customer can be used to creating brand value for café. As well as
marketing.
Customers spending High amount ( Bill amount ) can be given coupons to redeemed in the next visit
applicable for limited time to insure that such customers visit more often .
More Customer spend less than average on the Liquor hence we Should see Customer buy liquor more
frequently by combining them with food or Tobacco or frequent selling products.
Is observed that customer tend to buy Liquor and Wine of lower price we should recommend customer to
buy wine and liquor of higher price range to get more revenue.
We need to see supply of best selling product is optimum so no shortage occurred on the pick hr.
70. POS Data
●POS data refers to the sales data of a particular store
which includes information about product, transaction
date, time, quantity sold, amount ,discount and tax.
●The POS data of any store can give knowledge of the
sales of that particular store and insights to what more
can be done for increasing the sales.
●In this project we have been given a case scenario of a
cafe where based on the data we have to come up with
recommendations for the cafe that will be useful for its
sales.
71. Data available for Cafe
●We have three data sets for the cafe based on which we
have to analyze the shopping trend for that cafe and give
insights to how we can increase the revenue for the cafe. The
data sets which we have are:
1) POS data having the date, time, product, category, quantity,
tax, discount and sales attributes.
2)Category wise data having the aggregated quantity and
sales of each category defined on daily basis for the entire
year.
3)Product wise data having the quantity and sales of each
product throughout the year.
72. Data Overview
◉By looking at the datasets we have the following important
attributes:
●Category
●Product
●Days
●Hours
●Amount
●Quantity
73. Data Understanding
●From the data we could figure out below mentioned 7 major
different categories across which all the items are distributed:
●Beverage (127 items)
●Food (220 items)
●Liquor (33 items)
●Tobacco (60 items)
●Merchandise (87 items)
●Wines (46 items)
●Misc (52 items)
74. ●Based on each category we can calculate the total sales for
that particular category on hourly,daily, weekly, monthly and
yearly basis.
●We can find out which category is generating more revenue.
●Since we have information of time, we can calculate what are
the rush hours for the cafe.
●Product wise data set can give us insight to which products
are generating maximum revenue and falls under which
category.
●By looking at the invoice details we can find out the shopping
trend by consumers.
75. Data Pre-processing
●Since to begin any analysis, the data needs to be pre-
processed and cleansed so that the insights given are more
accurate and realistic.
●We too had to perform a lot of data cleaning and pre-
processing to make the data ready for actual analysis.
●Some of the data preprocessing steps involved during the
analysis are:
●The POS data required a lot of cleansing since we could find
a lot of special characters (like #,%,^,.,$,+) in Item.Description
as well as in numerical columns like Rate,Tax,Quantity.
76. ●The Category column had many rows with names like: 'TOBACCCOOOO',
'BEVERAGEEEE', 'MERCHANDISE <<<' and also few numbers, which required
cleansing and had to be given their proper name according to the respective items .
–Note: Few transactions had 'LIQUOR & TPBACCO' as their category, which were allotted to
the Tobacco category, since most of the items were contained in the Tobacco category
●Some transactions had no value in their category column, so the major or more selling
products with those scenarios were allocated to the category based on their previous
records or other transactions.
●Few transactions had no products and quantity which were removed from the data set
to bring accurate results in Market Basket analysis.
●In many cases, the Quantity / Rate/ total columns contained some very large values or
random number. Those rows had to be individually corrected based on other valid
column and also verified from the Product Sales data for their actual rate.
●NOTE: All the analysis were done on the cleaned and pre-processed
data and all the value and figures mentioned in this presentation are
actual values derived from the code (attached along with this ppt).
77. Approaches
●After cleaning the data the following approaches were
implemented or analysis done to provide insights to the
business:
●1. Basic Analysis
●2. Average Sales Information
●3. Category Overview Table
●4. Pareto 80/20 Analysis
●5. Market Basket Analysis
78. 1. Basic Analysis
●Based on the descriptive analysis of the data we have come
up with the following basic information:
●1. Most revenue generating category : Tobacco
●2. Least revenue generating category : Merchandise
●3. Average No. of bills generated per day: 190.2
●4. Average basket size per bill: 2 (~2.3)
●5. Average No. of unique items per basket: 2
●6. Average sale per bill: Rs. 467
79. 2. Average Sales Information
◉We have calculated the average sales of the cafe across
categories using descriptive analysis on the POS data and the
category-wise data
➢Quarterly
➢Monthly
➢Daily
➢Hourly
80. Quarterly Graph Analysis
●In the Quarterly Sales graph we
can see a good increase in
sales of the cafe from Q1 to Q3
of the fiscal year.
●The sales is gradually
decreasing after Q3.
●No much information can be
obtained in the quarterly sales,
so lets drill down the quarterly
data and take a look at the
monthly level sales graph.
81. Monthly Graph Analysis
◉As we can see from the graph, the sales
of December month is higher as
compared to other months. This is due to
heavy discounts or schemes for
Christmas and New Year. And also
because December is a festive month so
more customers visit the cafe with their
family and friends.
◉The sale after December is gradually
falling till summer. So, to retain the sales
of the cafe we can give weekly schemes
or offers every month after December
month to retain its customers.
◉In July and August the sales are fairly
high due to increase in sales in the
beverage and liquor category.
82. Daily Sales Graph Analysis
●On calculating the average sales on
daily basis we observed that on Saturday
the sales are the highest as compared to
and other days,so we can increase the
staffs in the cafe and reduce the staffs
on weekdays to lower the operational
cost.
●Since sales are low on weekdays:
Monday-Thursday, the cafe may plan to
give special offers on the items or give
discount on each categories on a regular
basis on the weekdays.
●For example, Monday 20% discount on
Beverage, Tuesday 20% discount on
Food, Wednesday 30% on Tobacco,etc.
83. Hourly Graph Analysis
●We can see from the graph that the sales
increases as the day ends, with maximum
customers/sales during the night time from
8pm-1am.
●So the cafe can allot more staffs for the
evening shift than morning shift which will
help in efficient use of the staffs.
●Since sales are low during the morning or
noon times,the cafe can offer happy hours
in the morning or noon to attract customers.
●Also, the cafe can start providing breakfast
combos or brunch combos to increase the
sales in the first half of the day.
84. 2. Category Overview Table
●The above Category Overview table has been
designed in R, which gives us information on few
metrics like: Avg. Sales / day, Avg.Qty / day,
Penetration and % Share of Business in terms of
sales and quantity across each category.
85. Category Overview Analysis
●As we can see in the Overview table,Tobacco is the
highest share of sales contributor and is present in one
of every two baskets (~50% penetration)
●Since quantity sold in Merchandise category is very low
, it can be used for schemes, gifting options and as free
on shopping above 2000 for its publicity,so that its
visibility is increased among the customers.
●Penetration of beverage is very high but its sale is
less,which means beverage is bought frequently but has
low margin on profit. So the cafe needs to Upsell his
beverage items to his customers, by providing premium
beverages or items which has high margin.
●Since Food category also has high penetration, Wines
or Misc items can be clubbed together at a discounted
price so as to increase the sales in the low selling
categories. Or may be provide combo packs of some
items of Food with some Wine to push the sales of Wine
category.
86. Pareto 80/20 Analysis
●According to Pareto's 80/20 principle, 80% of the effects come from 20% of
the causes.
●The same principle can be applied to the sales across the categories for
their respective products. 80% of the sales of a category come from 20% of
its top-selling products.
●For example, in Beverage category, total sales = 5461340, so its 80% sale
(4369072) comes from only 27 top selling products out of 127 items of
beverage category. So the ratio is: 27/127 = 0.2125 (~20%)
●This analysis can help the marketing head of the Cafe to promote or
advertise only top selling 20% products in each category to increase the
revenue of the cafe. This analysis help to efficiently utilize the marketing
budget by not promoting less useful or low selling items
●It also gives us an idea to club the top selling items with the low selling items
in that category so that the over all sales is increased and hence average
basket sale size is increased.
87. Top Selling items in a Category
●Beverage:
–Cappuccino, Red Bull, Lemon Iced Tea, Berry Blast, Caffe Latte,Qua Mineral Water,etc
●Food:
–Mocha Shake, Jr .Chl Avalanche, Poutine with fries, Oreo Cookie Shake, B.M.T.Panini, Kit
Kat shake,etc.
●Tobacco:
–Nirvana hookah single, Sambuca, Mint flavor single, NRG hookah, Calcutta Mint, Green
Apple flavor single,etc.
●Liquor:
–Carlsberg,Tuborg, KF Draught, Hoegaarden, Budweiser, Stella Artois,etc.
●Wines:
–VLN Cab Sauv,Red Sangria,Sula Blush Zinfandel, VLN Chenin Blanc,etc.
88. Market Basket Analysis
●Market Basket analysis is done on the point of sales data which
gives us the information like which pair of items are bought together
very frequently,or which items are likely to be bought together.
●This analysis helps us in Cross-selling the items, as we can make
combo packs for the items which are likely to be bought together as
sell them at a small discount to increase the purchase of the combos.
●It also gives us the information that which are the items which are
bought very frequently and are found in most of the baskets of the
customers. This will help in marketing those products, Upselling
those products with a premium one,or even club those frequently
bought items with the less bought ones.
89. MBA analysis
●The above image shows few association rules generated
by the Apriori algorithm for Market Basket Analysis in R.
90. MBA analysis (contd.)
●Using the Association rules, we have found few above rules which have a very high lift and are
bought very frequently together:
–Lemon Infused Char Grilled Veg & Add herb Roast Chicken
–Caffe latte & Add Hazelnut flavour
–French Fries & B.M.T. Panini
–Black Currant Iced Tea & Lemon Iced Tea
–Bun Maska and Chai & Masala Chai Cutting
–Cool California & Miami Melons
●Thus we see in the above few examples, that most customers try to add-on to their main items.
So we can provide combo packs/meals with add-ons at a discount so as to increase the sales.
●Other observation is that, customers buy similar items from a same category (like Black Currant
Iced Tea & Lemon Iced Tea, Bun Maska & Masala Chai, Cool California & Miami Melons). So, for
these customers we can provide offers like Buy One-get One, or Buy Two-Get One or 30%
Discount on buying two or more items from the same category.
91. ●Another information available from the Market basket analysis is the above top 20 items
which are very frequently bought .
●The value represents the fraction of transactions that each item appears in. So we see
that, Nirvana Hookah Single appears in more than 10% of the transactions,next 5 items
appears in 5-7%,while other top items appears in 2-4% of the transactions.
●This information can be used to Cross-sell frequently bought items with the ones which
are less bought.
92. More Approaches
●Since the data provided to us had only limited
information, If we had other information available along
with the POS data (like margin for each product, target
information, inventory information, customer information),
we could have provided more insights to increase
revenue based on few other analysis methods:
–Category Scorecard
–Trip Segmentation
●However, we will be using hypothetical data to work on
the above mentioned analysis.
93. Category Scorecard
●If we have target data and margin % for each product in a
category and also the inventory information of the cafe, we can
design a scorecard for every category with the following
metrics and adding a weight to every metric which will help us
calculate the overall score of the category:
–Contribution to Store
–Share of Sales
–Profit
–Inventory
●This final score of the category informs us the status of the
category with respect to the targeted value(100%).
94. Scorecard for Category: Beverage
●The above table shows the scorecard for the Beverage
category, which shows its % achievement with respect to the
target allocated for various metrics and sub-metrics.
95. ●Note: The values shown in the scorecard in red colour and the weights across
every metric are hypothetical values and needs to be changed according to the
business.
●If suppose our Inventory value =24000 and Margin% =23% for Beverage,then:
–Total Days on Hand = Inv. Value/Sales Value per Day
–Total Turns = 365 / Total Days on hand
–Return on Inventory(ROI) = Margin% * Total turns
●In the scorecard the Score of Beverage is 97.5%, which is close to the average 100%
which is at par with the target value. Hence we can say that we are almost at par with our
Target values.
●This score is very helpful to know which categories are performing according to our target,
and focus can be on the categories which have a score much below 100%.
●Also the scorecard can be used to compare between two or more categories and how
well a category is performing just by changing few targets and weights across metrics.
●According to the business needs, other metrics can also be included in the scorecard and
weights can be divided among the metrics.
96. Trip Segmentation
●If we had the customer information making the purchase along with
the transaction data, we can segment customers based on three
attributes:
–Recency : Number of days elapsed since the customer last purchased
something
–Frequency : Number of invoices with purchases during the year
–Monetary : Amount that the customer spend during the year
●Then we can create clusters based on the values of the three
attributes and try to target the cluster having majority of the
customers or also offer discounts based on the customer patterns
belonging to the same cluster.
97. Final Recommendations to increase
Revenue
●Based on the methods and analysis we have done, we can increase the
revenue by:
●Gift less selling products or merchandise with a minimum amount of bill to
increase its publicity and later increase its sales
●In monthly analysis, we can provide provide 'Sale' weekly or for 2-3 days
every month to retain the customers
●We have identified busy days and busy hours of the day, so to increase the
revenue of the cafe, we can give more schemes or offers during less busy
hours and days, and also initiate happy hours during the less peak hours on
the weekdays to improve the weekday sales.
●In category overview table, we see that the merchandise items are the least
sold products, so we can start gifting those items or give for free with a
minimum bill.
98. ●The beverage category has high penetration but low share of sale, so we
can stock premium items or items with higher margin, to Upsell items from
this category.
●Also since Food category has a very high penetration, we can provide
combo meals by including items from Wine category, so as to push sales for
Wines category and also increase its penetration.
●Based on Market Basket Analysis we saw that very similar items from same
category are bought together frequently, so provide buy one get one or other
offers on those items to increase the average basket sale.
●Also using MB Analysis we can cross sell items based on their associations
to increase the overall sales
●Club less revenue items with more revenue items for their advertisement
●Keep category dominance items which are less readily available in the
market to attract customers
99. Other Recommendations
●Pareto's 80/20 analysis can be used to promote only very few items from a
category rather than marketing useless items and reduce marketing costs
●Marketing new items or items which have high potential in the near future
along with the frequently sold items
●During non-peak hours decrease the number of staffs at work to reduce the
operational cost
●Based on the rush hours we can increase the POS counters so that
customers have better shopping experience
●Using the cluster analysis of customers, provide very frequently customers
or customer with high monetary value with loyalty programs / loyalty cards
for offers or discounts so that they become loyal customers to the cafe.
●Also using cluster analysis, market items to limited set of customers based
on their pattern and behavior, and email or message them based on their
visiting times to offer discounts on the products.
102. We can see the sales of beverages,food,liquor spike up at years
end(Nov,Dec,Jan)
There is another spike in Aug,Sep
We can see that the demand for products increases by 50% of
mean on peak seasons
Thus we shall have to maintain a 50% higher(mean demand)
inventory on peak demand season
103. Beverag
e
Liquor Food Tobacco
We can observe that the overall consumption of all the categories follow a similar trend
We can observe that consumption of Food and beverages peaks between 6 pm to 8 pm
The consumption of Liquor and tobacco peak between 8pm to 11 pm
Promotional offers can be given on off peak hours, thus giving discounts on products brought before 3pm
Happy hours can be a given on liquor before 2pm-6pm
Qty = 200918 Qty = 61760 Qty = 185062 Qty = 84723
Amt = 6122814
Amt = 2344684 Amt =
38111363
Amt = 19148180
104. • We can observe that the major contributor towards sales
amount is Food followed by tobacco
• Also Beverage account for major quantity of sales while
contribution to sales amount is just 9%
• Thus our focus should be on Food and tobacco as these products
account for major sales amount even though the sales quantity
Is lower
• We can do bundling of highest selling beverages with food items
or tobacco,thus doing cross selling
• Tobaco products can be bundled with liquor,thus with sales of tobacc
Liquor sales can be increased
• Promotonal offers can be given for the sales of food with beverages
• Since Food Accounts for only 35% of the sales amount but accounts
~60% of the sales, our focus should be on Food items for inventory
management and business analysis(Pareto principle)
qty Amt q.pct a.pct
Beverge 200918 6122814 0.373081 0.092645
Liquor 61760 2344684 0.114681 0.035478
Food 185062 38111363 0.343638 0.576669
Tobacco 84723 19148180 0.157321 0.289734
misc 6074 361740 0.011279 0.005474
538537 66088781 1 1
105. Market basket
Analysis
We can see that no matter what item purchased in lhs
• Beverage is consumed with tobacco with 5% confidence
• Food is consumed with tobacco
• Beverage is consumed with food
106. Segmentation of the
products
We can observe that, Cluster 1 comprise of all the item that are in high demand during festival time and night
hours
Cluster 2 belongs to the items that are in high demand in non peak season(seasonal items)
Cluster 3 is dominated by ever green items that are in demand throughout the year(Majourly food items)
Cluster 4 belongs to the items that are in high demand in but do not contribute majorly to the sales
All the items in a particular cluster belonging to various commodities can be grouped together in order to
optimize the sales
Items belonging to a cluster with high sales should be bundles with items bringing high sales
• Bundle Beverage with tobacco
• Food with tobacco
• Beverage with food
At various time of the day
107. Price as per the season and time of the
day
Mean price of the category
Change in demand with change in price
Maintain a price so that demand does not fall below a level
Lower the price when demand is lowest
• As peak time for a product can be assumed that the dedicated customers of the product are purchasing at that time
• We can give offers to those customers who have come to buy a particular commodity only
• As more high value tobacco is sold between 3pm-6pm,we can think of bundling it with liquor at that time, by giving happy hour
Below offers can be given at different time of the day
• As more high value liquor is sold between 12pm-5 am ,it can be bundled with tobacco items at that time
• As more high value food items are sold between 10am -2pm and 10pm to 2pm,it may be bundled with liqour
• As more high value beverages are sold between 5am-10 am and 8pm-5am,bundle it with food items
108. Analyse Top 20% of the items in demand(Pareto
Principle)
Top20% Total
qty product 306347 327038 0.936732
amountof sales 44259028 47185414 0.937981
Top 20% of the items in demand account for 93% of the
Total quantity sold and 93% of the sales amount
• We need to manage the inventory of these 20%(115 produ
thus managing the bottom line
We can see that the maximum amount of sales is
brought in by Tobacco
We can give bundled offers of misc items like add on s with Tobacco
To increase sales of miscellaneous items
109. Few of the top 20% of the
items
We need to maintain the inventory for these items as per the month and hour of the day