Big Data Capabilities for McDonald's: Approaches and Strategies

This research paper outlines the possible approaches for McDonald’s with regards to big data. The aim is to recognize areas of the business structure of McDonalds that require the application of big data. After the identification, big data approaches are then suggested for McDonalds.

Management Proposal
Big Data Capabilities
[Your Name]
[Date]
Executive Summary
Page 1 of 18

This research paper outlines the possible approaches for McDonald’s with regards to big
data. The aim is to recognize areas of the business structure of McDonalds that require the
application of big data. After the identification, big data approaches are then suggested for
McDonalds.
General Content Analysis and Sentimental Analysis look at describing the views of the
customers at McDonald’s outlets on the quality of the products they have purchased from
the fast food chain as well as the nature of the service they have received. The Cluster
Analysis builds on the outcomes of both the General Content Analysis and Sentimental
Analysis to categorise the outlets and regions where McDonald’s has outlets.
Consumer experience and consumer perception forms an integral part of the business
strategy of McDonald’s since it’s a market leader in the fast food industry. The General
Content Analysis, Sentimental Analysis and Cluster Analysis would enable McDonald’s to
better understand the views and needs of their customers.
Page 2 of 18

Contents
About McDonald’s 4
Key Business priority 4
Big data approach 5
Information and sources 8
Big data technologies 10
Big data visualisation examples 13
Big data adoption challenges and governance 13
Page 3 of 18

About McDonald’s
McDonald’s is a global chain of fast food restaurant started by the McDonald brothers in San
Bernardino in the state of California in the USA (McDonald 2018). The fast foods chain was
established in the year 1940 and mainly dealt with the sale of hamburgers.
In the year 1955, Ray Kroc bought McDonalds from the McDonalds brothers and structures the
brand to become the market leader it is presently. The headquarters for McDonald’s was previously
located in the city of Oak Brook in the state of Illinois in the USA. However, in the year 2018, the fast
food chain moved its headquarters to Chicago (McDonald 2018).
McDonald’s currently offer a range of fast food products apart from the hamburgers for which they
are known for. These products are: soft drinks, salads, milkshakes, chicken, coffee, French fries,
desserts and breakfast wraps (McDonald 2018). The products are among other products offered by
the food chain that are region specific. The region specific menus consist of fast food products that
are indigenous to the given part of the world.
Despite being established in the United States of America, McDonald’s has successfully managed to
use globalization to enter the fast food markets in other regions of the world. This has been achieved
by both having outlets that are owned by the company as well as through franchising. The
franchising has helped the company manoeuvre the strict market entry laws set by countries that
are not free markets. The collaboration with local businesses involved in franchising encourages
governments to ease the restriction for McDonald’s.
The extensive scaling and globalization of McDonald’s has seen the food chain open outlets in over
102 companies around the world. This scaling and globalization has also made the food chain the
second largest employer in the private sector.
Key Business priority
The McDonald’s food chain is a market leader in the fast food industry. This implies that McDonald’s
is less likely to be affected by market forces or market factors that might affect other smaller food
chains. The company size, diversity of products and markets allows McDonalds to enjoy some level
of protection that would otherwise adversely affect their sales and subsequently profits.
The most important aspects of business strategies for companies that are market leaders are image
and brand (Kuehlwein & Schaefer 2017). The image of the company is built by the consumers’
Page 4 of 18

perception of the service and products offered by a company. At times the strength and influence of
a brand may be large to the extent that the company can solely depend on this strength and
influence ((Petty 2016; Twede 2016). The aggregate result of this over reliance on the brand
influence may be neglect of other competences such as product quality, service quality and
marketing (Sirajuddin & Ibrahim 2017; Sheth 2017). The overall effect would be a significant
reduction in the brand influence, sales and subsequently profits.
This therefore makes the key business priority for McDonald’s fast food chain the quality of products
and services. Ensuring that the customers are happy and satisfied with the food and services offered
in the McDonald’s outlets will play a big role in maintaining the brand image of the fast food chain.
The present age of internet and social media has meant that brand influence can be drastically
affected by a single post or review on the service and products offered by the company in question
(French 2017; O'Malley & Lichrou 2016). Hence it is important for McDonald’s to guarantee their
customers satisfactory levels of services and product quality.
Big data approach
Customer satisfaction can be determined and inferences made through the use of analysis of big
data. Three main big data approaches can be used to address the concerns of customer satisfaction:
General Content Analysis, Sentimental Analysis and Cluster Analysis.
1. General Content Analysis
Content Analysis is a statistical technique in qualitative research that conducts a language diagnostic
of views, reviews, comments or responses on a subject of interest (Pashakhanlou 2017).
Content analysis technique applies the use of another data mining tool referred to as text
mining. Text mining is a big data collection method that gathers information that are
generally in the form of texts and processes it into vector, matrices or data frames that can
then be analysed to generate inferences (Badal & Kundrotas 2015).
Application of general content analysis as a big data approach for McDonalds will aim at
summarising the views, reviews, comments or responses of customers using single words.
This will be achieved by trying to singles out the most mentioned significant words from the
entire text.
These frequently mentioned word then provide McDonald’s with information on what aspects of
their products and services they need to focus on. This analysis can be focused on the fast
food chain outlets and therefore provide data on the product and service concerns of the
customers that purchase in the specific outlet of interest.
Page 5 of 18

Content Analysis can also be applied when McDonald’s is interested in determining the product and
service concerns in whole regions. This is especially for cases where the region in question is
exposed to similar product and service related factors. These factors can include sources of
farm produce or supplies for the products and the recruitment agencies used for hiring the
staff.
2. Sentimental Analysis
Sentimental Analysis is a form of a qualitative big data analysis technique. Sentimental Analysis
approach to big data involves the analysis of the contents of views, reviews, comments or
responses of individuals with the purpose of determining the underlying opinions of the
individuals (Mozetic & Smailovic 2016).
Sentimental Analysis measures the attitude of an individual’s view towards the topic or subject of
interest.
Similar to the general content analysis, sentimental analysis applied the use of text mining for
collecting and arranging the views, reviews, comments or responses of individuals. This
analysis also applies the Natural Language Processing (NLP) and computational linguistics for
the generation of the inferences on the attitude (Van Le & Montgomery 2018).
The Sentimental Analysis approach for McDonald’s will involve the collection of consumer views,
reviews, comments or responses through text mining. The data would then undergo
sentimental analysis where the individual view, review, comment or response will be termed
as either positive or negative depending on the number of positive or negative words used
respectively. This specific form of Sentimental Analysis is the Polarity Test.
Polarity Test mainly involves the classification of sentiments into two main groups, positive
sentiments and negative sentiments (Korkontzelos & Nikfarjam 2016).
Once the sentimental analysis has been done, the aggregate attitude score for individual McDonald’s
outlets can then be developed. This aggregate attitude score will give information on
whether the quality of products and services in the specific outlets meets the expectations
of the customers that purchase fast foods from the particular outlet.
The aggregate attitude score can also be generated for whole regions, especially for cases where the
region in question is exposed to similar product and service related factors. These factors
can include sources of farm produce or supplies for the products and the recruitment
agencies used for hiring the staff. This will give information on whether the quality of
products and services in the specific region meets the expectations of the customers that
purchase fast foods from the outlets particular region.
Page 6 of 18

3. Cluster Analysis
Cluster Analysis is an EDA (Exploratory Data Analysis) technique that groups objects with respect to a
standard or condition that is of interest (Kirk 2016; Renganathan 2017).
Cluster analysis puts together items that exhibit similarity in the condition of interest, hence allowing
the identification of items that don’t exhibit similar characteristics (Malki & Rizk 2016; Ying
2016).
Cluster analysis allows businesses to make informed decisions as regards strategies for the different
categories that are produced by the clustering. This presents an efficient and cost effective
means of business strategy development for business entities (Liu & Denxiao 2015).
The cluster analysis big data approach for McDonald will comprise clustering depending on both the
general content analysis and the sentimental analysis. These two clustering analysis will also
be considered in terms of outlets and whole regions.
a. Cluster Analysis of Outlets with respect to General Content Analysis
From the general content analysis, the most frequently mentioned words can be determined for
the different McDonald’s outlets. The results of these analysis can then be used in the
cluster analysis where the most frequently mentioned words for the outlets are
observed. The cluster analysis will then group together the outlets that have the same
most frequently mentioned words.
Closely observing the outlets in the same categories will provide a larger sample for determining
why customers are having a certain opinion about the product or services of
McDonald’s.
b. Cluster Analysis of Whole Regions with respect to General Content Analysis
From the general content analysis, the most frequently mentioned words can be determined for
the different regions (in terms of countries, regional arears or cities) where McDonald’s
has outlets. The results of these analysis can then be used in the cluster analysis where
the most frequently mentioned words for the regions are observed. The cluster analysis
will then group together the regions that have the same most frequently mentioned
words.
Closely observing the regions in the same categories will provide a larger sample for determining
why customers are having a certain opinion about the product or services of
McDonald’s.
Page 7 of 18

c. Cluster Analysis of Outlets with respect to Sentimental Analysis
From the sentimental analysis, the number of positive or negative sentiments can be determined
for the different McDonald’s outlets. The results of these analysis can then be used in
the cluster analysis where the number of positive and negative sentiments for the
outlets are observed. The cluster analysis will group together the outlets that have the
same range in number of positive or negative sentiments.
Closely observing the outlets in the same categories will provide a larger sample for determining
why customers are having a certain sentiment about the product or services of
McDonald’s.
The categories of significantly large positive responses provide McDonalds with a strategy
template for their products and services.
d. Cluster Analysis of Whole Regions with respect to Sentimental Analysis
From the sentimental analysis, the number of positive or negative sentiments can be determined
for the different regions (in terms of countries, regional arears or cities) where
McDonald’s has outlets. The results of these analysis can then be used in the cluster
analysis where the number of positive and negative sentiments for the regions are
observed. The cluster analysis will group together the regions that have the same range
in number of positive or negative sentiments.
Closely observing the regions in the same categories will provide a larger sample for determining
why customers are having a certain sentiment about the product or services of
McDonald’s.
The categories of significantly large positive responses provide McDonalds with a strategy
template for their products and services.
Information and sources
The required data for the big data approach for McDonald’s would need to be collected from the
client in the form of views, reviews, comments or responses. The information sources for this data
will be:
a. Online Surveys: through the use of online questionnaires, McDonald’s can get views that
their customers have about their products and services in different outlets and regions
Page 8 of 18

globally. The filled questionnaires will have to be sorted into outlets and regions for easier
analysis.
b. Social Media Comments: the comments that are posted by customers on the posts
published by the numerous McDonald’s social media sites present an opportunity for data
mining.
In instances where regions have their own McDonald’s social media site, for instance McDonald’s
Australia or McDonald’s Sydney, the collected data becomes more specific and reliable in
drawing inference on the quality of the products and services in McDonald’s outlet in the
region.
The collection of data can be done from any social media platform that McDonald’s has an account
in. This could include Facebook, Instagram and Twitter.
c. Product and Service Reviews: by setting up a site where customers can post reviews of the
quality of products and services at specific outlets, McDonald’s creates a source of
information that can provide useful inference once analysed. The site will provide customers
with the opportunity of giving an instantaneous review of products and service in the fast
food chains.
From this reviews, inferences can be made about the quality of both products and services in specific
McDonald’s outlets.
In the table below we have the big data approaches for McDonalds and description of the nature of
the variables that will be considered for each approach.
Big Data Approach Source of Data Variable type
1. General Content
Analysis
views, reviews, comments or
responses
Categorical data variable
measured on nominal scale.
2. Sentimental Analysis views, reviews, comments or
responses
Categorical data variable
measured on ordinal scale.
3. Cluster Analysis Output of General Content Categorical data variables that
Page 9 of 18

Analysis and Sentimental
Analysis.
are measured on either
nominal or ordinal scales
depending on the source of
the data.
Big data technologies
In the big data approach for McDonald’s the following big data technologies will be applied:
a. Technology for streaming data processing
b. Technology for storage of data
c. Technology for analysis of data
1. Technology for streaming data processing
Data from the views, reviews, comments and responses represent real time data that requires to be
continuously processed once received. The filled questionnaires from the online surveys will
be streaming once filled by the participants in the surveys. The comments on the
McDonald’s posts on social media will also represent streaming data since the posts will be
continuous and the comments are also expected to be made continuously. The site for the
reviews will also be a source of streaming data.
The Apache sparks technology is a reliable big data technology for the processing of streaming data,
which makes this technology appropriate in the big data approach for McDonald’s.
2. Technology for storage of data
The output data from Apache Sparks processing of the streaming data needs to be stored as an
intermediate stage between the processing of data and the analysis stage for the data.
In order to prevent loss of data quality and guarantee the reliability of the inference drawn from the
analysis of data, the storage of the data should be efficient and structured (Karolin &
Schrape 2018).
The NoSQL Database is a big data technology that provides for structured storage of data. This
implies that the data can be easily accessed and organized in a way that makes the analysis
process efficient. Hence NoSQL is a reliable big data approach to data storage for
McDonald’s.
Page 10 of 18

3. Technology for analysis of data
R Studio is an R language environment that provides big data analysis capabilities (Galit, et
al. 2018). R Studio is capable of running all the big data approaches (Text mining, General
Content Analysis, Sentimental Analysis and Cluster Analysis) for McDonald’s.
Using a single data analysis technology gives the inferences a degree of consistency. This
thus means that there are lower chances of producing conflicting inferences (results with
high variability).
R Studio will mine the data from the processed streaming data in Apache Sparks first, then
send the data for storage in the NoSQL Database. From the database, General Content
Analysis, Sentimental Analysis and Cluster Analysis will be conducted on the structured data
using R to produce the final inference.
The diagram below represents the description of the big data technologies application for
McDonald’s.
Page 11 of 18

Page 12 of 18
Comments from Social Media Site
(Facebook, Twitter, Instagram)
INFORMATION
SOURCES
Reviews from Outlets Responses from online
surveys
Streaming Data Processing APACHE SPARKS
R STUDIOText Mining
Data Storage NoSQL DATABASE
Data
Analysis
General Content
Analysis
Sentimental
Analysis
Cluster Analysis
R STUDIO
Results

Big data visualisation examples
Figure 1
Figure 1 above represents the cluster analysis results for the big data. The plot shows the grouping
of items into clusters of different colours. The items falling within the circles with similar pattern and
colours are considered to be in the same category with respect to the condition of interest.
Figure 2
Figure 2 above shows a sample of the results from the general content analysis of the reviews given
by customers at a McDonald’s outlet. From the analysis we can observe that the most frequently
mentioned word is chicken. This implies that chicken is the most sold product for the specific
McDonald’s outlet.
Big data adoption challenges and governance
The table below provides a summary of the challenges that are faced in the adoption of big data
approaches and the recommendations for ways of overcoming such challenges.
Page 13 of 18

Issue Recommendations
Challenges 1. Data Privacy.
Restriction of the
access and usage of
data to individuals
with proper clearance
is key for any
organization (Vicenc
2017).
2. Data Security. Data
needs to be secured so
as to prevent the
leakage and
subsequent usage of
the data by
unauthorised
individuals or
personnel ((Igor &
Lindsey 2016).
Apache Rangers
Governance 1. Data Source. The
source for any data
has to have a level of
integrity attached to
the process of the
collection of data to
ensure high quality
inference from the
analysis (De Jong-Chen
2015).
2. Data Reliability. The
data, even at face
1. IBM
2. SAS
Page 14 of 18

value, has to be
indicative of the actual
data on the ground
without exaggerations
or deductions (Baack
2015).
3. Data Usability. The
usability of data
mainly concerns the
existence of
permission for analysis
and inferencing of
data from the owners
of the data (Lenca &
Ferretti 2018).
4. Data Appropriateness.
The data used for an
analysis has to have
the parameters and
variables of interest
for it to provide useful
inferences (Galit, et al.
2018).
Page 15 of 18

References
Baack, S 2015, Datafication and Empowerment: How the Open Data Movement Re-articulates
Notions of Democracy, Participation and Journalism.
Big Data and Society, vol. 2, no. 2, pp. 1-11.
Available at: https://journals.sagepub.com/doi/full/10.117
[Accessed 5 November 2018].
Badal, VD & Kundrotas, PJ 2015, Text Mining for Protein Docking.
PLoS Computational Biology , vol.
11, no. 12, pp. 1-5.
Available at: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1004630
[Accessed 1 November 2018].
De Jong-Chen, J 2015, Data Sovereignity, Cybersecurity, and Challenges for Globalization.Georgetown Journal of International Affairs, vol. 5, no. 3, pp. 112-122.
Available at: https://www.georgetownjournalsofinternationalaffairs.org/online-edition/data-
sovereignty-cybersecurity-and-challenges-for-globalization
[Accessed 1 November 2018].
French, J 2017, Asset Pricing With Investor Sentiment: On The Use of Investor Group Behaviour to
Forecast ASEAN Markets.
Research in International Business and Finance, vol. 5, no. 1, pp. 124-148.
Available at: https://www.sciencedirect.com/science/article/pii/S027553191630128
[Accessed 2 November 2018].
Galit, S. et al. 2018,
Data Mining for Business Analytics. 1st ed. New Delhi: John Wiley & Sons, Inc..
Available at: https://www.wiley.com/en-us/Data+Mining+for+Business+Analytics%3A+Concepts
%2C+Techiques%2C+and+Applications+in+R-p-9781118879368
[Accessed 4 November 2018].
Igor, G & Lindsey, M 2016, Alternative Models of Ethical Governance: The 2016 New Brunswick-
Otago Declaration on Research Ethics.
New Zealand Sociology, vol. 31, no. 4, pp. 1-7.
Available at:
https://www.researchgate.net/publication/308796133_Alternative_Models_of_Ethical_Governance
_The_2016_New_Brunswick-Otago_Declaration_on_Research_Ethics
[Accessed 7 November 2018].
Karolin, K & Schrape, J 2018, Societal Implications of Big Data.
Kunstliche Intelligenz, vol. 32, no. 1,
pp. 12.
Available at: https://link.springer.com/article/10.1007/s13218-017-0520-x
[Accessed 3 November 2018].
Kirk, A 2016,
Data Visualization: A Handbook for Data Driven Design. 2nd ed. Thousand Oaks, CA:
Sage Publications, Ltd.
Available at: https://uk.sagepub.com/en-gb/eur/data-visualisation/book244865
[Accessed 1 November 2018].
Korkontzelos, I & Nikfarjam, A 2016, Analysis of the Effect of Sentimental Analysis on Extracting
Adverse Drug Reactions from Tweets and Forum Posts.
Journal of Biomedical Informatics , vol. 62,
no. 1, pp. 148-158.
Available at: https://www.ncbi.nlm.nih.gov/m/pubmed/27363901/
[Accessed 5 November 2018].
Page 16 of 18

Kuehlwein, J & Schaefer, W 2017, Ueber-Branding: How Modern Prestige Brands Create Meaning
Through Mission and Myth.
Journal of Brand Strategy, vol. 5, no. 4, pp. 395-409.
Available at: www.academia.edu/37656252/Ueber-
Branding_How_modern_prestige_brands_create_meaning_through_mission_and_myth_Part_1
[Accessed 3 November 2018].
Lenca, M & Ferretti, A 2018. Considerations for Ethics Review of Big Data Health Research: A Scoping
Review.
PLoS ONE, vol. 13, no. 10, pp. 23-25.
Available at: https://journals.plos.org/plosone/article?id=1
[Accessed 5 November 2018].
Liu, Q & Denxiao, R 2015, Research on The Structure of Public Fiscal Expenditures Based on the
Cluster Analysis Methods.
Modern Economy, vol. 6, no. 6, pp. 1-10.
Available at: file.scirp.org/Html/6-7201064_57167.htm
[Accessed 1 November 2018].
Malki, AA & Rizk, MA 2016, Hybrid Genetic Algorithm with K-Means for Clustering Problems.
Open
Journal of Optimization, vol. 5, no. 2, pp. 1-4.
Available at:
https://www.researchgate.net/publication/304189240_Hybrid_Genetic_Algorithm_with_K-
Means_for_Clustering_Problems
[Accessed 2 November 2018].
McDonald's 2018.
About. [Online]
Available at: www.mcdonalds.com/about
[Accessed 1 November 2018].
Mozetic, I & Smailovic, J 2016, Multilingual Twitter Sentiment Classification.
PLOS ONE, vol. 11, no. 5,
pp. 1-7.
O'Malley, L & Lichrou, M 2016, Marketing Theory. In: M. J. Baker & S. Hart, eds.
The Marketing Book.
Routledge: Oxon, pp. 37-52.
Pashakhanlou, A H 2017, Fully Integrated Content Analysis in International Relations.
International
Relations, vol. 31, no. 4, pp. 447-465.
Petty, RD 2016, A History of Brand Identity Protection and Brand Marketing . In: B. Jones & M.
Tadajewski, eds.
The Routledge Companion to Marketing History. Routledge: Oxon, pp. 97-114.
Renganathan, V 2017, Text Mining in Biomedical Domain with Emphasis on Document Clustering.Healthcare Informatics Research, vol. 23, no. 3, pp. 141.
Sheth, JN 2017,
Legends in Marketing. 6th ed. New York: Sage Publications.
Sirajuddin, O & Ibrahim, AH 2017, Five Competitive Forces Model and the Implementation of
Porter's Generic Strategies to Gain Firm Performances.
Science Journal of Business and
Management, vol. 10, no. 16, pp. 1-5.
Twede, D 2016, Commercial Amphoras: The Earliest Consumer Packages?.
Journal of
Micromarketing , vol. 22, no. 1, pp. 98.
Van Le, D & Montgomery, J 2018, Risk Prediction using Natural Language Processing of Electronic
Medical Health Records in an Inpatient Forensic Psychiatric Setting.
Journal of Biomedical
Informatics, vol. 86, no. 2, pp. 49-58.
Page 17 of 18

Vicenc, T 2017,
Studies in Big Data. 1st ed. Chicago: Springer International Publishing .
Ying, M 2016, An Analysis on the Relative Efficiency of Furniture Enterprises in Guangdong Province
Based on DEA-BCC and Clustering Method.
Open Journal of Business and Management , vol. 4, no. 2,
pp. 13.
Page 18 of 18