Marketing Final Project

Name

BUS 309 SP 2025

Instructor

Date
The beverage industry has experienced a remarkable transformation over the past decade,

particularly within the health and wellness segment. As consumers become increasingly mindful

of their dietary habits and overall well-being, demand for functional beverages has surged. This

shift is not just a fleeting trend but a reflection of deeper lifestyle changes that prioritize

convenience without compromising on nutrition. For companies operating in this space, the

ability to anticipate and respond to these evolving preferences is essential for sustained success.

In this final project, we explore the dynamic landscape of the health-oriented beverage industry,

focusing specifically on consumer attitudes toward drink products and how those attitudes

compare to real-world purchasing behaviors. Our research leverages two primary sources of

data: first, a comprehensive consumer survey designed to gauge preferences, priorities, and

perceptions surrounding healthy beverage options; and second, historical sales data from Energy

Drinks, a business engaged in the distribution and sale of such products. Together, these datasets

offer a unique opportunity to bridge the gap between what consumers say they want and what

they actually buy.

The purpose of this report is not simply to summarize survey results or to chart out sales figures,

but rather to analyze these two perspectives in tandem. We seek to uncover whether consumer

self-reported behaviors align with actual market activity, and if not, to understand the reasons

behind such discrepancies. The survey component of this study collected a wide range of

insights from participants across different demographics, including age, gender, employment

status, and physical activity levels. Respondents were asked about their drink consumption

habits, preferred ingredients and flavors, importance of factors like price and brand trust, and

their openness to trying new products or paying a premium for organic or natural options. These
responses allow us to build a rich profile of the target customer and identify key motivators that

influence buying decisions.

Complementing this, the sales history data from Energy Drinks provides a factual account of

product performance in the market. This dataset includes detailed records of transaction volumes,

revenue trends, and customer behavior over time. By analyzing these figures, we can pinpoint

which products and features are driving revenue growth, which are underperforming, and how

these trends have evolved. Furthermore, when overlaid with survey data, this enables us to detect

alignment or misalignment between perception and performance, insights that are crucial for

business decision-making. From a strategic standpoint, this dual-layered approach empowers

businesses to move beyond surface-level assumptions and into a realm of evidence-based

planning.
 1. Market Overview and the Importance of Forecast Accuracy

The focus of this analysis is the health-oriented beverage sector, specifically high-protein,

functional, and wellness drinks. This segment has experienced substantial growth recently as

consumer preferences shift towards healthier and more convenient dietary choices. These

beverages attract a diverse audience, including fitness enthusiasts, busy professionals, and

health-conscious consumers seeking nutritious alternatives to traditional soft drinks and snacks.

As competition escalates with the emergence of new brands, accurately predicting consumer

demand becomes essential for success.

In this project, precise forecasting is crucial for enhancing business efficiency, ensuring customer

satisfaction, and maintaining financial health. We are concentrating on predicting sales for

beverages like high-protein drinks, low-sugar options, and nutrient-enriched functional

beverages. Many of these products, especially those with natural ingredients and fewer

preservatives, have short shelf lives, making overestimating demand particularly risky due to

potential spoilage and increased costs. Conversely, underestimating demand risks stoke outs, lost

sales opportunities, and dissatisfied customers, which can negatively affect brand loyalty and

reputation. Our forecasting strategy must balance caution to minimize waste with the need for

responsive product availability, using survey data to gauge customer preferences alongside

historical sales records for reliable consumer behavior insights. By merging these data sources,

we aim to create flexible, data-driven demand forecasts that optimize supply chain decisions and

improve inventory management to better meet market needs.

The line chart depicting monthly can sales for Energy Drinks reveals a steady upward trend,

indicating growing consumer demand. This positive trajectory suggests that the company is

enhancing its market presence, likely due to increased brand awareness, expanded distribution,

or shifting consumer preferences toward health-oriented beverages. Notably, the data highlights

periodic sales spikes during identified Peak Season months, likely influenced by seasonal factors.

This underscores the importance of aligning marketing strategies and inventory management

with seasonal demand fluctuations.

temporary saturation or a decrease in promotional efforts. Such variability highlights the

necessity for accurate forecasting and demand planning. Incorporating analysis of promotional

strategies, such as influencer marketing, could provide deeper insights into the reasons behind

sudden sales increases. Overall, the sales history serves as a crucial basis for forecasting future

demand, refining production schedules, and reducing inventory waste. By understanding these

cyclical sales patterns, Energy Drinks can strategically prepare for high-demand periods while

ensuring cost-efficiency during slower months.

3. Regression Analysis and Comparison

Simple linear regression

It modeled cans sold over time (Month Ordinal) and yielded the following equation;

Cans Sold = –823,252,658.55 + 1,116.52 × Month Ordinal

This model indicates a steady, positive trend in sales over time, with approximately 1,117 more

cans sold each day on average, demonstrating growth. However, the model does not account for

external influences like marketing or seasonality.

Multiple linear regression

This integrated additional variables. Whether an influencer promotion was present and whether

the month was in a peak season. This model produced the equation:
Cans Sold = –834,227,569.39 + 1,131.28 × Month Ordinal + 251,653.11 × Influencer

Promotion + 82,145.54 × Peak Season

The multiple regression model provides a more nuanced and accurate reflection of sales

dynamics. It shows that influencer promotions lead to a substantial increase of over 250,000 cans

in monthly sales, while peak seasons contribute an additional 82,000+ cans. The coefficient for

time remains positive and slightly stronger than in the simple model, reaffirming the upward

sales trend over time.

4. Peak

Season

Analysis

Based on our historical sales data, there is evidence to suggest the presence of a peak season in

Energy Drinks' performance. Months labeled as Peak show noticeably higher sales volumes
compared to non-peak months. Several of the highest-selling months, such as those in mid-year

or holiday-adjacent periods, coincide with the “Peak” season flag in the dataset.

I did run multiple linear regression using Peak Season as one of the predictors, the model

estimates an average increase of approximately 82,146 cans during peak months. This

statistically significant contribution highlights that the Peak Season variable plays a meaningful

role in predicting sales, reinforcing the existence of cyclical, seasonally influenced spikes in

demand.

5. Impact of Promotions on Energy Drinks Sales

The historical data strongly suggests that promotional events, particularly influencer campaigns,

have a significant and measurable impact on Energy Drinks’ sales performance. Specifically the

multiple linear regression analysis, we found that the presence of an influencer promotion is

associated with an average increase of approximately 251,653 cans in monthly sales. This is a

substantial uplift, far exceeding the baseline effect of time-based trends or even seasonal

changes. The regression model treats influencer activity as a binary variable (0 = no promotion, 1

= promotion), and the large positive coefficient indicates that months featuring such promotions

consistently outperform others.

6. Forecasting Limitations and Reasonable Time Horizons

Given the nature of the data, monthly sales figures alongside promotional and seasonal

indicators, it is feasible to generate forecasts with moderate to high accuracy for the short to mid-

term. The data reveals distinct patterns influenced by time, seasonal peaks, and promotional

events, enhancing the effectiveness of regression models. However, the dataset's limited size and
granularity, consisting of monthly data over a relatively short timeframe, restrict the depth of

trend analysis and complicates the capture of more complex market dynamics, as well as external

factors like competitor actions or economic fluctuations.

Therefore, forecasts extending beyond a 6 to 12-month horizon should be approached with

caution. While the historical data provides a solid foundation for short-term planning and

seasonal stocking strategies, longer-term predictions would greatly benefit from additional data

points. Integrating external market or consumer trend data could significantly improve reliability

and resilience, ensuring that forecasts remain relevant in the face of evolving market conditions.
Target Customers

1.

Household Income vs. Age

There is a moderate positive correlation between age and household income up to a

certain point. Income generally increases with age until it stabilizes or becomes more

varied around retirement age. This relationship aligns with typical career and earnings

trajectories, where individuals build wealth over time, often peaking in their 50s or early

60s.
 Household Income vs. Region

Household income levels vary significantly by region, with some areas showing broader income

diversity and others clustering at lower ranges. This suggests regional economic disparity that

could be influenced by factors such as employment opportunities, cost of living, or demographic

composition.
Household Income vs. Single or Married

The scatter plot does not effectively convey the relationship between marital status and

household income, it is designed to visualize the relationship between two numerical variables,

with each point representing a pair of values (x, y).

2. Pivot table
 This pivot table layout was chosen to layer the three variables logically, starting with

broad demographic region, followed by marital status, and breaking down income trends

across age groups. This nested structure makes it easy to identify income differences not

just between regions, but also between singles and married individuals within the same

age bracket and location, enabling more actionable insights for targeted marketing or

resource allocation.

4.

The survey results provide valuable insights into consumer preferences and behaviors within the

health-oriented beverage market. Respondents identified taste, nutritional value, and price as the

primary factors influencing their drink choices, with brand trust also playing a significant,

though secondary, role. The preference for protein sources was notably balanced between plant-

based and whey-based options, reflecting a growing inclination toward alternative nutrition.

Furthermore, many consumers expressed a willingness to pay a premium for organic or natural

drinks, highlighting a shift towards clean-label expectations. A strong preference for ready-to-

drink formats emerged, indicating a clear demand for convenience in product design.

Consumer behavior varied by age, activity level, and purchase location. Active individuals

tended to prioritize nutritional content, while younger consumers were more receptive to trying

new brands. Although online purchasing is on the rise, traditional retail channels remain

dominant. Peak consumption periods aligned with fitness goals and seasonal health trends.

Through conjoint analysis, nutritional value was identified as the leading decision-making factor,

closely followed by taste and price. These insights suggest that companies looking to compete in
this market must focus on delivering health benefits, innovative flavors, and transparent value,

while prioritizing convenience in their offerings.

Additional Requirements

Multifactor ANOVA

Does Age, Region, and Marital Status significantly affect the Caffeine per ounce preference?

Since p < 0.05 → the Age and Region significantly affects Caffeine preference. For marital

status, p > 0.005 clearly indicating that it does not affect caffeine preference in any way.

Chi Square Test

The Chi-Square Test checked for an association between Influencer Promotion and Peak Season.

A p-value < 0.05 was denoted upon analysis meaning there is a statistically significant

relationship, more like , promotions are more or less likely to occur during peak season
 References

NielsenIQ. (2023). The rise of health-conscious consumers: Understanding the shift in beverage

preferences.

American Marketing Association. (2023). 2023 consumer trends report: Wellness and functional

beverage consumption. https://www.ama.org

Grunert, K. G. (2010). Consumer behavior with regard to food innovations: Quality perception

and acceptance. Food Quality and Preference, 21(6), 777–783.

Steenkamp, J.-B. E. M. (2019). Dynamics in consumer behavior with respect to agricultural and

food products. In Wierenga, B. (Ed.), Agricultural marketing and consumer behavior in a

changing world (pp. 143–188). Springer.

Schindler, P. S. (2019). Business research methods (13th ed.). McGraw-Hill Education.