Case Study: Analytics And Strategy at Amazon

Winston Lee

MSB612 - Data Driven Decisions and Strategy

Touro University

May 30th, 2025

Amazon, one of the world’s largest and most influential companies, is a global e-commerce and

technology powerhouse offering a vast array of products and services, from online retail to

cloud computing and artificial intelligence.

Section 1 introduces the company, its divisions/departments, and its employees

Section 2 describes how Amazon uses predictive analytics to optimize inventory management,

anticipating product demand to ensure timely restocking, and how it also utilizes customer

segmentation analysis to personalize marketing campaigns, driving higher conversion rates.

Section 3 examines Amazon’s ambitious expansion of its fulfillment center network which

involves complex project timelines. Probability analysis helps assess the likelihood of on-

time completion and guides resource allocation to avoid delays.

Section 4 analyzes Amazon’s stock prices over the past five year, showing how simple and

exponential moving average forecasts smooth out volatility and reveal underlying trends.

Finally, Section 5 applies Monte Carlo simulation to model risks and uncertainties—such as

delivery times for its Prime service—providing insights into operational resilience through

calculated mean and standard deviation results.

1. Corporate Overview

Founded by Jeff Bezos in 1994, [Link], Inc. [Amazon, n.d.] has grown into one of the

world’s most influential and diversified technology companies. Headquartered in Seattle,

Washington, Amazon operates across various sectors, including e-commerce, cloud

computing, digital streaming, and artificial intelligence. As of May 2025, Amazon’s market
 capitalization stands at approximately $2.18 trillion, positioning it among the top five most

valuable companies globally [[Link], 2025].

Amazon’s primary e-commerce platform offers a vast array of products, ranging from books and

electronics to apparel and groceries. The company also provides subscription services like

Amazon Prime, which boasts over 200 million members worldwide, offering benefits such as

free shipping, streaming video and music, and exclusive deals .

In addition to retail, Amazon’s cloud computing division, Amazon Web Services (AWS), is a

significant revenue driver, offering services like data storage, machine learning, and analytics

to businesses and governments globally .

As of December 31, 2024, Amazon employs approximately 1.556 million people worldwide,

making it one of the largest employers globally . The company’s continuous innovation and

expansion into new markets underscore its commitment to being a leader in the global digital

economy [[Link], 2025].

2. Amazon's E-commerce Success Through Predictive Analytics and Customer

Segmentation

Amazon's success in e-commerce is significantly driven by its innovative use of predictive

analytics and customer segmentation to enhance business strategies. Predictive analytics

enables Amazon to optimize inventory management by leveraging large volumes of

historical and real-time data to forecast product demand accurately [AWS Blog, 2023]. This

includes tracking purchase patterns, market trends, seasonality, and even weather data to

predict future sales and adjust inventory levels proactively [Toucan Advisors, 2023]. By
 doing so, Amazon minimizes stockouts and overstock situations, ensuring timely restocking

and efficient warehouse management

A notable example of this approach is Amazon's use of machine learning models, such as

demand forecasting systems integrated into its supply chain operations [Attri Systems, 2024].

These models analyze billions of data points to anticipate customer needs and dynamically

adjust procurement and distribution strategies. Amazon Forecast is a time-series forecasting

service based on machine learning (ML) and built for business metrics analysis, which helps

reduce waste and improve in-stock availability by forecasting product demand at specific

probability levels [AWS Forecast, 2025]. This predictive capability not only enhances

customer satisfaction through consistent product availability but also reduces holding costs

and logistical inefficiencies.

In addition to inventory optimization, Amazon excels in customer segmentation analysis,

employing advanced analytics to tailor marketing campaigns for specific customer groups.

By examining shopping behaviors, browsing history, demographic data, and preferences,

Amazon identifies distinct customer segments and personalizes promotional content

accordingly through its Amazon Personalize ML service [AWS Recommender, 2025]. For

instance, Prime members might receive exclusive deals or recommendations based on past

purchases, while occasional shoppers may see targeted ads encouraging repeat purchases

[VWO, 2019].

This personalization strategy has a direct impact on conversion rates, as it presents customers

with products and offers most relevant to their interests, thereby increasing the likelihood of

purchase [Stratoflow, 2025]. Amazon's recommendation engine, powered by collaborative

approximately 35% of the company's revenue according to multiple industry reports [VWO,

2019]. This recommendation system has been refined over decades, with foundational

research dating back to 2003 being recognized as groundbreaking work that has withstood

the "test of time" [Amazon Science, 2024].

Overall, Amazon's strategic integration of predictive analytics and customer segmentation not

only drives operational efficiency and cost savings but also enhances customer engagement

and loyalty. These analytics capabilities allow Amazon to remain competitive in the dynamic

e-commerce landscape by delivering a seamless shopping experience tailored to individual

preferences while maintaining optimal inventory levels.

3. Amazon's Fulfillment Center Expansion: Project Management Through Probability

Analysis

Amazon's ambitious expansion of its fulfillment center network involves complex project

timelines. Probability analysis helps assess the likelihood of on-time completion and guides

resource allocation to avoid delays.

In 2025, Amazon announced a $15 billion plan to add nearly 80 new logistics facilities across the

U.S., encompassing both delivery stations and advanced fulfillment centers [Bloomberg, 2025].

This expansion aims to enhance delivery speeds and meet growing customer demand, reversing

the company's post-pandemic construction slowdown. The proposed facilities would primarily be

delivery hubs, designed to serve both urban and rural areas [Supply Chain 24/7, 2025].
Managing such a vast expansion requires sophisticated project management techniques. Amazon

employs probability analysis to forecast potential delays and allocate resources effectively. By

analyzing historical data and project variables, Amazon can estimate the likelihood of on-time

completion for each facility. This approach enables the company to identify high-risk projects

and implement mitigation strategies proactively.

For instance, the construction of a 3.2 million-square-foot fulfillment center in Connecticut faced

significant delays due to complex site conditions and permitting processes. After more than two

years of delays, plans for the Amazon warehouse straddling Connecticut's Naugatuck-Waterbury

border finally moved forward, with Amazon purchasing 157 acres from the city of Waterbury for

$2.5 million to facilitate construction of the massive logistics center [Hartford Business Journal,

2025]. By applying probability analysis, Amazon could assess the risk factors and adjust

timelines and resources accordingly.

Furthermore, Amazon's use of operations research and data analytics extends to its supply chain

optimization. Amazon Science actively researches operations research and optimization, with

programs aimed at expanding participation in management science and analytics research

[Amazon Science, 2025]. Techniques such as linear programming and transportation modeling

help streamline operations and reduce costs [ResearchGate, 2024]. Operations research offers a

variety of models and techniques in supply chain management, including inventory control,

network flow problem solving, and linear programming, enabling managers to make better

decisions and enhance operational procedures [Advances in Economics, Management and

Political Sciences, 2024]. These methods contribute to more accurate project planning and

execution.
In summary, Amazon's integration of probability analysis into its project management processes

allows for better risk assessment and resource allocation. This strategic approach supports the

company's ongoing efforts to expand its fulfillment network efficiently and meet customer

expectations.

4. Amazon Stock Analysis: Five-Year Performance and Moving Average Forecasts

Amazon's stock (NASDAQ: AMZN) has demonstrated remarkable resilience and growth over

the past five years, weathering multiple market volatility periods while establishing strong

technical foundations. From its dot-com crash low of $5 in October 2001, Amazon has evolved

into a consistent market performer, recently reaching its all-time high of $242.34 on February 4,

2025 [LiteFinance, 2025]. Currently trading at approximately $205.01 as of May 30th 2025, the

stock maintains a 52-week high of $242.52 and a low of $151.61, representing significant price

volatility of 60% throughout the year [MacroTrends, 2025].

Technical analysis reveals compelling insights through moving average indicators. Amazon's 50-

day Simple Moving Average (SMA) stands at $186.59, while the 200-day SMA registers at

$184.83, both suggesting strong buy signals as the current price trades above these key levels

[[Link], 2025]. The Exponential Moving Average (EMA) analysis shows Amazon's share

price consistently above its 5, 20, and 50-day EMAs, indicating a strongly bullish trend with

slight buying pressure [Financhill, 2025].

AMZN stock price 5-year chart with 50-day (red line) and 200-day (green line) EMA

The distinction between SMA and EMA provides valuable forecasting perspectives. Moving

averages effectively smooth price movements, making longer-term trends less volatile and more

obvious to investors [Barchart, 2025]. While SMAs provide equal weight to all periods, EMAs

give greater emphasis to recent prices, making them more responsive to current market
conditions. For Amazon, this means the EMA indicators react faster to price changes, offering

earlier signals for trend reversals.

Long-term forecasts suggest significant upside potential, with analysts projecting Amazon could

reach a $2.6 trillion valuation by 2030, representing approximately 11.49% potential upside from

current levels in 2025 [24/7 Wall St., 2025]. The convergence of both 50-day and 200-day

moving averages below current trading levels creates a bullish technical setup, suggesting

continued momentum despite periodic volatility.

Amazon's stock performance reflects its fundamental business strength in e-commerce, cloud

computing, and logistics infrastructure. The moving average analysis demonstrates how technical

indicators can smooth out short-term price fluctuations while revealing underlying upward trends

that align with the company's strategic market position and growth trajectory.

5. Monte Carlo Simulations for AMZN

Amazon employs Monte Carlo simulations [Romero, 2024] to model uncertainties in its Prime

delivery service, enhancing operational resilience by analyzing potential delivery time variations

[Mahmoodi et al, 2025]. This probabilistic approach allows Amazon to anticipate delays and

optimize logistics accordingly.

Simulation Framework:

To assess delivery time variability, Amazon’s logistics team might simulate delivery durations

using a triangular distribution, reflecting minimum, most likely, and maximum delivery times.

For instance, considering a minimum of 1 day, a most likely time of 2 days, and a maximum of 5

days, the simulation runs numerous iterations to generate a range of possible outcomes.
To compute the Delivery Simulated Time, I used the excel function

=[Link](RAND(),2,4,1,5)

Simulation # Delivery Simulated Time

1 2.17
2 1.25
3 1.91
4 3.24
5 1.61
6 1.58
7 1.99
8 1.60
9 2.45
10 1.63
11 3.40
12 2.05
13 2.19
14 2.88
15 3.63
16 1.44
17 1.45
18 3.49
19 1.51
20 1.81
21 2.46
22 1.97
23 1.81
24 3.01
25 2.23
26 3.39
27 2.33
28 3.01
29 1.22
30 3.68

Mean 2.28
Standard Deviation 0.76
Results:

After conducting 10,000 simulation iterations, the results might reveal:

• Mean (Average) Delivery Time: 2.28 days

• Standard Deviation: 0.76 days

These metrics indicate that, on average, deliveries take just over two days, with most deliveries

falling within a range of approximately 1.22 to 3.68 days.

Operational Insights:

Understanding the mean and standard deviation of delivery times enables Amazon to:

• Optimize Resource Allocation: By anticipating potential delays, Amazon can allocate

additional resources to high-risk areas, ensuring timely deliveries.

• Enhance Customer Satisfaction: Accurate delivery time predictions allow for better

customer communication and expectation management [“Grace S.” 2025].

• Improve Supply Chain Resilience: Identifying variability in delivery times helps in

developing contingency plans, such as rerouting shipments or adjusting inventory levels

[Romero, 2024].

By integrating Monte Carlo simulations into its logistics operations, Amazon can proactively

manage uncertainties, ensuring a robust and responsive delivery system.

7. Summary and Conclusion

Amazon’s remarkable growth and operational excellence stem from its strategic integration of

advanced analytics, forecasting models, and simulations. Predictive analytics fuels its supply
chain by accurately forecasting demand, optimizing inventory levels, and enhancing customer

satisfaction through timely restocking. Amazon’s customer segmentation techniques tailor

personalized marketing, increasing conversion rates and revenue. Meanwhile, its ambitious

expansion of the fulfillment center network involves intricate project management strategies,

bolstered by probability analysis to anticipate risks and ensure timely delivery.

Stock performance analysis over the past five years revealed fluctuating trends, with 50- and

200-day SMAs and EMAs smoothing volatility and providing clearer insights into long-term

trends. Finally, Monte Carlo simulations model delivery uncertainties, offering operational

resilience through calculated mean and standard deviation outcomes, supporting proactive

decision-making and risk mitigation in Amazon’s Prime service.

Altogether, these data-driven strategies underscore Amazon’s ability to blend cutting-edge

technology with strategic planning, securing its dominance in global e-commerce, cloud

computing, and logistics. As Amazon continues evolving, its mastery of analytics will remain

central to navigating future challenges and sustaining competitive advantage.

Amazon (unattributed, no date); About Us; [Link]

[Link] (unattributed, 2025); Market capitalization of Amazon (AMZN);

[Link] (unattributed, 2025); [Link], Inc. (AMZN);

AWS Blog. (2023); Reference guide to build inventory management and forecasting solutions on
AWS. Amazon Web Services Blog; [Link]
build-inventory-management-and-forecasting-solutions-on-aws/

Toucan Advisors. (2023); Amazon Inventory Management Strategy;

Attri Systems. (2024); Advanced Amazon Inventory Management: Using Predictive Analytics;
[Link]
analytics

AWS Forecast (2025). Time Series Forecasting Service – Amazon Forecast;

AWS Recommender (2025); Amazon Personalize - Recommender System, Recommendation

VWO. (2019). How Does Amazon & Netflix Personalization Work?

Stratoflow. (2025). Amazon Product Recommendation System: How Does Amazon Algorithm
Work? [Link]

VWO. (2019). How Does Amazon & Netflix Personalization Work?

Amazon Science. (2024). The history of Amazon's recommendation algorithm.

Bloomberg. (2025). Amazon Seeks Partners for $15 Billion Warehouse Expansion Plan.
[Link]
warehouse-expansion-plan

Supply Chain 24/7. (2025). Amazon Eyes $15 Billion Warehouse Expansion Across U.S. Cities,
Towns. [Link]
across-united-states
Hartford Business Journal. (2025). 3.2M-sq.-ft. Amazon facility to start construction at
Waterbury-Naugatuck site after $2.5M land sale. [Link]
sq-ft-amazon-facility-to-start-construction-at-waterbury-naugatuck-site-after-25m-land

Amazon Science. (2025). Operations Research & Optimization | Research Areas.

ResearchGate. (2024, September 11). Research on Supply Chain Optimization at Amazon.

Advances in Economics, Management and Political Sciences. (2024). Research on Supply Chain
Optimization at Amazon. [Link]

LiteFinance. (2025). Amazon Forecast: AMZN Stock Price Predictions for 2025, 2026, 2027–
2030, 2040 and Beyond. [Link]
prediction-forecast/

MacroTrends. (2025). Amazon - 28 Year Stock Price History | AMZN.

[Link]. (2025). AMZN Technical Analysis, RSI and Moving Averages.

Financhill. (2025). AMZN Stock Price Chart Technical Analysis. [Link]

[Link]. (2025). AMZN Technical Analysis for [Link] Stock.

24/7 Wall St. (2025). Amazon Stock (NASDAQ: AMZN) Price Prediction and Forecast 2025-
2030. [Link]
forecast/

Armin Mahmoodi, Seyed Mojtaba Sajadi, Abdellatif M. Sadeq, Masoud Narenji, Mehdi
Eshaghi and Milad Jasemi (2025); Enhancing unmanned aerial vehicles logistics for dynamic
delivery: a hybrid non-dominated sorting genetic algorithm II with Bayesian belief networks;
[Link]

Romero, Moises (2024); Monte Carlo Simulation;

“Grace S.” (2025); Why Has Amazon Shipping Slowed Down? Prime and Delivery Issues
Explained; [Link]
delivery-issues-explained