Super Bowl Ads Analysis Report

type

status

date

slug

summary

Abstract

As the most watched sporting event of the year in the United States and as one of the most lucrative advertising programs in the world, Super Bowl ad traffic has unimaginable reach and commercial value for large companies, with ads featuring a variety of themes such as: national, political, animal, sexual, etc. According to the Super Bowl advertising website, it contains 249 ads from the 10 brands that advertised the most in the Super Bowl from 2000 to 2021, with 16 fields which include links to each brand as well as information on length, estimated cost, YouTube statistics, TV viewers and seven each standard marked as "yes "Yes" or "No" for each criterion. With the help of visualisation tools, relevant data analysis and visualisation between the features and variables in this study, we can get a clearer picture of what audiences want, what they like and what kind of ads they prefer, and to a greater extent, create commercial value.

Keywords: View; Advertising; Super Bowl; Tableau; Brand; Visualisation

Abstract Contents Introduction Motivation, Aims and Objective Literature Review Methodology and Results Methodology Results Conclusion and Future Work Conclusion Future Work References Appendix A. Code

Introduction

Motivation, Aims and Objective

The purpose of this study is to present information on each variable of each Super Bowl ad and to conduct a multivariate visual analysis. Identify the characteristics of the popular ads and analyse their strengths. The data is analysed by quantifying the relationship between values and features. There are outrageously expensive ads here, some comical, some disturbing, thought-provoking and strange. In this study, it will be understood how some of the elements of this programme interact with each other. The study includes the following points.

How do the different features of the commercials change over time?

Which brands have the most Super Bowl commercials? Do they have a unique style?

How have ratings, TV viewership and advertising costs evolved over time?

As the most watched sporting event of the year in the United States and as one of the most profitable advertising programs in the world, Super Bowl ad traffic has unimaginable impact and commercial value for large companies, with many themes featured in the ads, such as: national, political, animal, sexual, etc. It is necessary to examine the relationship between the characteristics of these adverts and the various variables of video viewing. By understanding the relevant data, performing relevant data analysis and visualisation between the features and variables, we can gain a clearer understanding of the audience's needs, what viewers like and what kind of advertising they prefer, and create commercial value to a greater extent. Preliminary findings show that over the 21 years of the Super Bowl, the cost of investment in video advertising increased, and increased with the length of the ad. 10 brands had the highest number of ads in the beverage category. 30 and 60 second ads had high engagement rates, with 30 second ads being more cost effective. Audience engagement was significant for video ads featuring animals, fun and quick presentation of the advertised product.

Literature Review

Super Bowl advertising has always been a hot topic in commercial advertising and marketing research. Many scholars have studied the characteristics, effects and influencing factors of Super Bowl advertising. This paper systematically reviews the existing literature on Super Bowl advertising, in order to provide references for future research.

Research shows that the input-output ratio of Super Bowl advertising is increasing, and 30-second advertisements are the most cost-effective (., 2006)[4]. Website traffic data shows that ads with animals and humor themes get higher attention and interaction online. Mittal (1994) found that humorous ads are more easily remembered and spread in consumers' minds[1]. Du et al. (2015) experimental study found that humorous ads can significantly improve advertising claims and brand recall[3].

The average length, production budget and number of Super Bowl ads are rising, reflecting the increasing importance of companies to Super Bowl ads (DeLorme et al., 1999)[7]. The content and form of Super Bowl ads are also changing, with more product categories and creative techniques tending to be more dramatic, emotional and humorous (Chang et al., 1999)[2]. 30-second ads have become the mainstream of Super Bowl ads, achieving higher cost-effectiveness (DeLorme et al., 1999; .et al., 2006)[7].

Many studies have explored the relationship between Super Bowl advertising and consumer brand cognition and attitude. Studies have found that Super Bowl advertising can significantly increase consumer brand awareness and consumption orientation (Yi, 1990)[5], create brand impressions and influence consumer attitudes (Baker et al., 2002)[8]. Super Bowl advertising with social emotional appeals, such as humor and warmth, can better attract audience attention and generate brand goodwill (Pickett et al., 2012)[6].

The application of interactive media has expanded the influence of Super Bowl advertising. Studies have found that Super Bowl advertising combined with online marketing can produce higher interaction and consumer participation. Word-of-mouth communication from social media can also expand the influence of Super Bowl advertising (Cunningham et al., 2007)[9].

In summary, scholars have studied Super Bowl advertising from the perspectives of characteristics, relationship with consumer cognition and attitude, and new media marketing effect, providing a wealth of theoretical basis and experience. This paper reviews and analyzes these research results to provide references for follow-up Super Bowl advertising research.

Methodology and Results

Methodology

Data visualisation is a key element of modern company impact, and is present in every tool and workflow. It is an important part of the job not only for data engineers, data scientists and data analysts, but also for those who do not have "data" in their job title. Data visualisation appears in product demos, in ad hoc communications on Slack, in leadership presentations to shareholders, and even in marketing materials. We wanted to change the tool-centric and role-centric design approaches one often sees in data visualisation, which force users to jump back and forth between tools or walls of permissions for different roles.

This experiment was preceded by using python to do data processing and simple analysis of graphs. Several Python dependency packages, pandas and plotly, were used to provide information on each variable of each ad and to perform multivariate visual analysis. The features of popular advertisements were identified and analysed for their strengths. The data is analysed by quantifying the relationship between values and features. The project plans to use the D3 visualisation tool to visualise quantitative values and audience preferences. Observe how different variables change over time and look at the relationships between multiple variables for visual analysis.

Results

Data preparation

According to the Super Bowl Ads website, it contains 249 ads from the 10 brands with the most ads in the Super Bowl from 2000 to 2021, with 16 fields which include links to each brand as well as information on length, estimated cost, YouTube statistics, TV viewers and 7 criteria marked as "yes" or "no" for each " or "no" for each criterion. The dataset was evaluated and the graph below shows how the dataset has no missing values for the fields except for the YouTube Views and Youtube Likes fields which have 12 and 18 missing data respectively.

Advertising analysis

According to the dataset, apparently the Top 10 advertising brands spent a total of $1284M on Super Bowl 21, had 25K viewers on TV, 372M views on YouTube and had 1.18M likes, and the average length of the ad video was 44 seconds.

Time is money. What does the analysis data look like for video ad costs, ad hours, TV and viewership on YouTube? We used a line graph to analyse the ad data over a 21 year period from 2000-2021, ad costs increased from $21M in 2000 to a high of $145.6M in 2019, obviously ad costs decreased in 2020 and 2021, but the overall trend is incremental, ad duration is getting longer, and brands are increasing their Super Bowl YouTube ad budgets, as shown in the chart below, with a declining trend in viewership on TV and an incremental increase in viewing and click likes for ads on new media social platforms such as YouTube.

Line plot for the Estlmated Cost the years

Line plot for the Youtube Views the years

Line plot for the Youtube Likes the years

There are several reasons for this situation:

1. The rise of new media platforms. The emergence of platforms such as YouTube and social media has provided new advertising channels for brands. Brands can place ads on these platforms to attract more audiences, driving up the cost of video advertising.

2. The expanding audience scale of new media platforms. As YouTube and social media users continue to increase, brands have a larger potential audience base. Brands are also willing to spend more on advertising on these platforms to reach more users.

3. Increase in ad duration. With the rise of new media platform advertising, ad forms have become more diversified, no longer limited to short videos. Brands have also started to place long video ads on these platforms, which inevitably pushes up advertising costs.

In summary, this situation is brought about by the new advertising platforms from technological development and the interaction of users and brands on these new platforms, driving up video advertising costs and changes in advertising strategies. But at the same time, it also brings new opportunities for brands to reach a wider audience at a lower cost.

Engagement rates are high for 30 and 60 second ads, with 30 second ads costing approximately $2.1M to $5.6M and 60 second generation ads costing approximately $4.2M to $11.20M, with 30 second ads being more cost effective.

Advert Length Year(Large circle=Higher engagement rate)

There are several reasons for this result:

1. 30-second and 60-second ads can provide more information, attract higher attention and interest of audiences, thus generating higher engagement and interaction. This drives up the advertising costs of these two types of ads.

2. Longer ads require more production costs, need longer and more sophisticated shooting, post-editing and other work, which also directly pushes up costs.

3. In TV advertising time slots, 30-second and 60-second ads occupy more time length, so advertising fees are higher. TV advertising fees are usually calculated based on ad length, the longer the time, the higher the cost.

In summary, the rising cost of 30-second and 60-second ads is because advertisers, TV stations and other media all attach more importance to the effects that these two longer ad forms can achieve, and are willing to spend higher costs. At the same time, these two ad forms themselves also require higher production and investment costs. However, compared with 60-second ads, the increase in the cost of 30-second ads is slightly smaller, due to their smaller investment and easier to achieve the ideal effect. Therefore, the room for price increases is relatively small and the cost-effectiveness is higher.

Each ad is divided according to the click-through rate on YouTube, and the engagement of each ad is calculated and sorted, and clustered according to "top5", "6-10" and "10+ " are clustered and the 7 characteristics of these three categories are calculated as a percentage of the ads. top5 ads with significant engagement with funny and animal characteristics will be clicked on by the public. The highest rated TV ad (Budweiser, Puppy Love, 2014) had an engagement rate of 9.83%, which was for animal and patriotic ads.

Youtube advertising click categories heat density map

The above results occurred for the following reasons:

1. Advertising features such as humor and animals are more likely to attract audience interest and resonance, so these types of ads typically have higher click-through rates and engagement. Audiences tend to click on and engage with ads that are relevant to their interests.

2. Animal ads trigger emotional resonance in audiences through the cute pet effect, making it easy for audiences to like and be interested, and then engage in interaction. This is also why animal ads often get high click-through rates and engagement.

3. Patriotic ads achieve high engagement by arousing audiences' patriotic feelings and enthusiasm. Audiences are more willing to interact and comment below these types of ads.

In summary, factors such as ad features, ad forms, and brand effects all influence the click-through rate and engagement of YouTube ads. Choosing attractive and interactive ad features and forms can improve the overall effectiveness and communication effect of the ad. Well-known brands are also more likely to get higher ad click-through rates and engagement on the platform.

Advertisers need to comprehensively consider all factors and design video ads that are most likely to trigger audience resonance and interaction. Only in this way can they achieve the ideal advertising effects on platforms such as YouTube.

The chart below shows a pie chart of the number of advertisements for the Top 10 brands over a 21 year period. 4 of the Top 10 brands, Bud Light, Budweiser, Doritos and Pepsi, accounted for over 50% of the advertisements for the Top 10 brands, with Bud Light in first place, producing a total of 62 video advertisements over a 21 year period.

The above results occurred for the following reasons:

1. The influence of brand advertising frequency. Brands like Bud Light with the largest number of ads have adopted a high-frequency advertising strategy, publishing a large number of video ads on YouTube, which increases brand exposure and awareness, and thus has the greatest influence on audiences.

2. The influence of brand marketing budget. Brands with more ample marketing budgets can invest more resources in video ad production and placement, so the number of their YouTube ads is also greater. These brands can occupy more resources on platforms like YouTube and have a competitive advantage.

3. The influence of brand promotional products. Brands with more types of promotional products also need to produce more diverse video ads to cover all their products, which indirectly increases the number of their YouTube ads. Brands like Bud Light and Budweiser have many products and need many ads to promote them.

In summary, factors such as brand advertising strategies, marketing budgets, product categories, and platform mechanisms all affect the number of YouTube ads. To gain greater exposure and influence on YouTube, brands need to adopt proactive advertising strategies and invest sufficient marketing resources.

To be more specific, Brand was categorized into 10 brands, mainly beverage, automotive and Other, with Bud Light, Budweiser, Pepsi and Coca-Cola categorized as Drink, Hyundai, KIA and Toyota categorized as Automobile, and Doritos, E-trade and NFL categorized as Other. The total cost of the Beverage category was $664M, or 52%, and the cost of the Automotive category was $340M, or 26%. The majority of ads in the Beverage category were for interesting and quick to see advertised items.

Bud Light, Budweiser and Coca-Cola in beverages, Hyundai in cars, and the NFL in Other genres dominated the top 5 in terms of video ad spend in the Super Bowl, costing 50% of all cost spend for these 10 brands.

Analysing brands by year for advertising costs, it is clear that Budweiser is spending more on Super Bowl advertising than other brands from 2018 to 2021, and one other brand is the NFL also spending significantly more on Super Bowl advertising from 2017 to 2020, and that beverage, automotive and Other type brands have also increased their spending on Super Bowl advertising in recent years. So will the investment in advertising pay off? To explore this question, we looked at audience engagement with brand advertising.

The above situation occurred for the following reasons:

1. The Super Bowl has extremely high media exposure, and advertising can generate huge brand influence. Brands hope to gain widespread attention through Super Bowl ads, quickly increase brand awareness and reputation, so they increase investment.

2. The Super Bowl audience is huge, reaching hundreds of millions, allowing ads to reach the maximum range of audiences. These audiences are also high-value target groups for brands, making Super Bowl ads more cost-effective.

3. In recent years, Super Bowl ads have become a major cultural event, and the ads themselves have become a focus of audience attention. Creative Super Bowl ads can generate high social impact, become a hot topic of discussion, which also drives brands to increase investment to maximize exposure.

In summary, the huge potential value of Super Bowl ads drives brands to increase investment, but the uncertainty of investment output also enhances the blindness of advertising investment. To achieve a perfect balance between Super Bowl advertising investment and output, brands must systematically analyze the cognitive and reactive differences of audiences to different ads, minimize the blindness and limitations of advertising investment, and optimize the use of advertising budgets.

The analysis shows that three brands, Budweiser, KIA and Toyota, have the highest engagement with the ad among the 10 brands, as the limited ad data does not allow for an analysis of economic benefits. Analysing the brand types by year, Budweiser had the highest engagement pair in 2014 with 98%, followed by E-trade with 33.9% engagement in 2021.

The above results occurred for the following reasons:

1. The influence of brand ad content and form. Ads with content and forms that match the Super Bowl culture and audience can generate higher audience interaction rates, such as ads for Budweiser, KIA and Toyota. The creativity of these three brands' ads may match the theme and audience preferences of the game better, resonate more easily, and have the highest interaction rates.

2. The influence of brand ad themes. Ad themes that audiences are more interested in, such as humor and inspiration, will have higher interaction rates. Different brands' ad themes have different appeal, so interaction rates vary. For example, E-trade's ad theme in 2021 may be more compelling than other years, so the interaction rate is higher.

3. The influence of ad creation. Creative and well-produced ads will have higher interaction rates. In different years, brands' ad creativity, conception and production will differ, so interaction rates will also differ. Budweiser's ad creativity in 2014 may have exceeded 2021, so the former has better interaction effects.

In summary, factors such as the content and form of brand Super Bowl ads, the attractiveness of themes, the level of creativity, the impact of Super Bowl events, and ad placement strategies can have a significant impact on the interaction rates generated by ads. To increase the interaction rate of Super Bowl ads, brands must comprehensively consider these various factors, adjust advertising strategies and content, continuously optimize work quality and placement, match audiences and event themes closely, create the most impactful ads, and enhance audience interaction.

A quantitative analysis of advertising by year for commodity types in the Super Bowl shows that it is clear that advertising investment and output for beverages has been high every Super Bowl, followed by advertising investment and output for cars, and that investment gradually began to increase from 2009 to 2021.

The above results occurred for the following reasons:

1. The influence of product categories. Products such as beverages and automobiles are more suitable for advertising through sports events, so their Super Bowl ad investment and output effects are the highest. This is the reason why the two types of brands invest the most.

2. Brand awareness of the value of Super Bowl ads. In recent years, the influence and value of Super Bowl ads have continued to increase, and brands' awareness of their advertising effects has gradually deepened. This drives brands to continue increasing advertising investment to maximize effectiveness.

In summary, factors such as product category attributes, brand marketing goals, awareness of advertising value, historical data analysis, fit with events, and advertising budget will affect the extent and continuity of brands' investment in Super Bowl advertising. To develop the best Super Bowl advertising investment strategy, brands must comprehensively consider all these factors, understand the match between themselves and Super Bowl ads, achieve the best balance between investment and output, and optimize the use of advertising budgets.

In 21 years, Budweiser's TV Views and YouTube Views of advertising videos are the highest, and Budweiser has the largest number of advertising videos.

30s & 60s ads have high engagement rate. 30s ads cost around $2.95M to $5.6M while 60s cost around $5.9M to $11.2M. 30s ads is more cost efficient.

The ads have gotten longer and the brands increase their budget on Super Bowl YouTube ads.

The above results occurred for the following reasons:

1. The brand's emphasis on Super Bowl advertising. Brands like Budweiser have long paid close attention to Super Bowl advertising, observing and analyzing audience data, and continuously optimizing advertising strategies. Therefore, their advertising investment and effectiveness are the best. Other brands that want to achieve similar success also need to pay close attention to this advertising platform.

2. The influence of ad duration and budget. 60-second ads are longer and more expensive, so the interaction effect is better, but from a cost-effectiveness perspective, 30-second ads are more practical. Brands need to find a balance between the two and develop the best advertising investment strategy.

3. The influence of advertising platforms. Relying solely on television advertising is no longer applicable. Brands must expand to new platforms such as YouTube. Advertising on different platforms has different characteristics. Brands need to understand the advertising mechanism and audience of each platform and invest appropriate resources to expand exposure.

In summary, factors such as the brand's emphasis on Super Bowl advertising, the balance of advertising budget and duration, the use of advertising platforms, data analysis and advertising optimization, and the fit between products and culture jointly determine the success or failure of the brand's Super Bowl advertising. To achieve the same excellent results, other brands must learn from Budweiser, continue to deepen their understanding and increase investment in all these aspects.

Total views over the years ,costs and roi

Due to the viewership and influence of the superbowl game, advertisers tend to be willing to pay a premium for ad spots. but are the high costs justified by equally higher roi?

In this analysis, we will look at super bowl commercials' trendsover the years, and try to answer the question - what should the advertisers do to get the most out of their spend.

There is no clear trend in views through out the years. The vast majority of views come from TV (as opposed to YT).

Views per dollar on the x-axis measures the number of views per dollar spent (used to measure the roi). Size of the dot stands for the cost (spend) of a certain ad. ideally, advertisers want the ad to be away from the y-axis (high roi) and small (low cost).

It seems that over time, brands are spending more but getting less in return (a good example is nfl - some of the largest dots but also closest to the y axis).

The above results occurred for the following 3 reasons:

1. Advertising investment grows too fast, leading to resource waste. In recent years, major brands' investment in Super Bowl advertising has grown rapidly, but advertising effectiveness has not improved synchronously, leading to a decline in return on investment. Brands overinvest and misuse resources, which is an important reason for the decline in advertising effectiveness.

2. Insufficient use of new media. Super Bowl advertising relies too much on television platforms, and the use and investment of new media platforms is insufficient, limiting the reach and influence of advertising. To increase return on investment, brands need to increase investment in new media and expand the audience reach of Super Bowl advertising.

3. Limited data-driven operations and advertising optimization. Many brands rely too much on personal experience and subjective judgment in Super Bowl advertising. Lack of data-driven operations, advertising effectiveness evaluation and optimization make it impossible to truly optimize the use of advertising budgets. To succeed, brands must establish a data analysis mechanism and continuously improve data operations and optimize advertising creativity and delivery.

In summary, uncontrolled growth in advertising investment, insufficient use of new media, and lack of data-driven operations and optimization are the three major reasons for the decline in return on advertising investment. Brands must continue to learn and improve in these three areas to control the rhythm of advertising investment, increase investment in new media, establish a data-driven advertising optimization mechanism, and truly improve the effectiveness of Super Bowl advertising to maximize return on investment.

The case of the NFL brand also warns other brands to prevent over-investment from causing serious waste of resources and low efficiency. Brands must understand the characteristics of Super Bowl advertising, develop scientific investment strategies, invest moderately and accurately, choose the most potential advertising forms and platforms, continuously optimize and improve, and optimize the use of advertising budgets. This is the key to improving return on investment.

How are views/roi affected by different elements of ads?

How are views/roi affected by different elements of an ad?

Dots sized by views per dollar (spent)

Views do not seem to be heavily associated with different elements in an ad. However some elements are clearly associated with less than ideal roi (views per dollar spent) - patriotic and celebrity. On the contrary, ads that are funny and shows product quickly seem to be associated with higher roi.

Advertisers should keep the ad placement cost below 15M (meaning shorter ads), avoid elements like patriotism, celebs and make their ads funny, sexy while showing the products early on.

With no significant increase in viewership over the past 20 years, advertisers are still spending more cash than ever - what's the catch?

One possible reason i could think of is that with the rise of social media platforms over the past years, advertisers are not only getting views from the actual game being televised, but also publicity from possibly trending on social media. our data does not capture the entirety of the return.

Conclusion and Future Work

Conclusion

Time is money. Over the 21-year period, the cost of video advertising on the Super Bowl has increased and has increased with the length of the ad. 10 brands have the highest number of ads in the beverage category and account for more than 50% of the ad cost over the 21-year period. 30-second and 60-second ads have high engagement rates, with 30-second ads costing around $2.1M to $5.6M and 60-second generation ads costing around $4.2M to $11.20M, with 30-second ads being more cost-effective. 4.2M to $11.20M for the 60 second generation, with the 30 second ad being more cost effective. Viewer engagement is significant in video ads that feature animals, funny and quick display of the advertised product, so funny, animal and quick display of the product is where advertisers need to invest more of their advertising dollars.

Future Work

Future work, continue to improve data integration and analysis work, study the Super Bowl data dimensional correlation, complete data pre-processing through feature engineering including data transformation, feature validity analysis, derived variables, downscaling and feature selection, observe data characteristics, classify data characteristics and correlation according to hierarchical analysis, study cause-effect relationships, correlate actual situations, explore the commercial value presented by Super Bowl ads, summarise The data was analysed to determine the commercial value of the Super Bowl advertisements, the video characteristics of the advertisements and the investment patterns of brands in the advertisements, and to answer the question of which characteristics make the advertisements more sticky and fresh and can be converted into economic benefits. The results of the correlation data analysis were synthesised using visualisation techniques such as D3, which was able to stylise and decorate the simple and cluttered charts created in the early steps. The colour schemes used in the earlier data analysis explorations (to show as many values as possible) are replaced by more considered colours that emphasise the key themes in the data being analysed. The formatting of labels on elements of the chart (such as axes) is more considered and weaker (less visually conspicuous), as it is not enough for the data visualisation to be understood by those who build it, it needs to be understood and believed by those involved in making decisions around the visualised data without hindrance.

References

The references will follow an IEEE style (or any other standard referencing format such as Harvard, APA, among others).

Books

[1] Mittal, B. “A theoretical analysis of two recent measures of involvement.” In Advances in consumer research, Eds.Chris T. Allen and Deborah Roedder John, vol.21, p239-244, Provo, UT : Association for Consumer Research, 1994.

Journals

[2] Chang, W. C., Simpson, T. W., & Liu, F. C. "Evolution of Super Bowl commercials: A longitudinal study of brand commercials." International Journal of Sport Finance, 4(4), 195-205, 2009.

[3] Du, J., Hao, Y., Feng, Y., Liao, X., & Wang, C."How humorous can you get? The effects of humorous advertising on brand attitude and memory." Journal of Marketing Communications. 22(5), 576-594, 2016.

[4] Eisend, M. "Explaining the joint effect of source credibility and negativity of information in two-sided messages." Psychology & Marketing, 23(11): 923-940, 2006.

[5] Yi, Y. "Cognitive and affective priming effects of the context for print advertisements." Journal of Advertising, 19(2), 40-48, 1990.

[6] Pickett, G. M., & Kahle, L. R. "Social values and salespeople's use of emotion in selling." Journal of Marketing Theory and Practice, 20(2), 197-208, 2012.

Conference Technical Articles

[7] DeLorme, D. E., Huh, J., & Reid, L. N. "Cost effectiveness of Super Bowl advertising: 1990-1999." Paper presented at the 1999 American Academy of Advertising (conference paper), 1999.

[8] Baker, W. E., Hutchinson, W. S., Moore, D., &Nedungadi, P. "Brand familiarity and advertising: Effects on the evoked set and brand preference." Advances in Consumer Research, 29(1), 637-642, 2002.

Online Sources

[9] Cunningham, Scott M., Shirish C. Srivastava, and Thomas L. Baker. “Impact of interactive marketing on Super Bowl advertising.” Journal of interactive marketing 21, no. 2: 26-37, 2007. 2007. Available:https://doi.org/10.1002/dir.20071.

[10] Narechania, A., Coscia, A., ... Endert, A., 2022. Lumos: Increasing Awareness of Analytic Behavior during Visual Data Analysis. IEEE Transactions on Visualization and Computer Graphics 28, 1009–1018. doi:10.1109/TVCG.2021.3114827

[11]Skender, F., Manevska, V., 2022. Data Visualization Tools-Preview and Comparison, Journal of Emerging Computer Technologies. APA.

[12]Dowdy, A., Jessel, J., ... Peltier, C., 2022. Structured visual analysis of single-case experimental design data: Developments and technological advancements. Journal of Applied Behavior Analysis 55, 451–462. doi:10.1002/jaba.899

[13]Batt, S., Grealis, T., ... Tomolonis, P., 2020. Learning Tableau: A data visualization tool. Journal of Economic Education 51, 317–328. doi:10.1080/00220485.2020.1804503

[14]Sariasih, F.A., 2022. Implementasi Business Intelligence Dashboard dengan Tableau Public untuk Visualisasi Propinsi Rawan Banjir di Indonesia. Jurnal Pendidikan Tambusai 6, 14424–14431.

[15]Jena, B., 2019. An Approach for Forecast Prediction in Data Analytics Field by Tableau Software. International Journal of Information Engineering and Electronic Business 11, 19–26. doi:10.5815/ijieeb.2019.01.03

[16]Noh, Y., Kim, H., ... Cheong, Y., 2021. The relationship between consumer engagements with Super Bowl ad tweets and the advertisers’ official Twitter accounts: a panel data analysis of 2019 Super Bowl advertisers. International Journal of Advertising 40, 852–869. doi:10.1080/02650487.2020.1819006

[17]Li, D., Watanabe, N.M., 2022. Effects of Super Bowl advertising on online brand search: ten years of insights from 2011 to 2020. International Journal of Sports Marketing and Sponsorship 23, 841–854. doi:10.1108/IJSMS-07-2021-0151

[18]Hartmann, W.R., Klapper, D., 2018. Super bowl ads. Marketing Science 37, 78–96. doi:10.1287/mksc.2017.1055

[19]Lee, J.G., Ko, K.A., 2021. The market responses to super bowl advertising: The role of product type and multiple executions. Sustainability (Switzerland) 13. doi:10.3390/su13137127

[20]Pedregosa, F., Varoquaux, G., ... Duchesnay, É., 2011. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research 12, 2825–2830.

[21]Python? (2019). Seibutsu-Kogaku Kaishi, 97(4), 226–229. https://doi.org/10.1145/2544173.2509536

[22]Hoque, E., Agrawala, M., 2020. Searching the Visual Style and Structure of D3 Visualizations. IEEE Transactions on Visualization and Computer Graphics 26, 1236–1245. doi:10.1109/TVCG.2019.2934431

[23]Peng, D., 2020. TargetingVis: visual exploration and analysis of targeted advertising data. Journal of Visualization 23, 1113–1127. doi:10.1007/s12650-020-00671-w

[24]Oh, C., Sasser, S., Almahmoud, S., 2015. SOCIAL MEDIA ANALYTICS FRAMEWORK Journal of Information Technology Management SOCIAL MEDIA ANALYTICS FRAMEWORK: THE CASE OF TWITTER AND SUPER BOWL ADS. Journal of Information Technology Management XXVI, 1–18.

[25]Hartmann, W.R., Klapper, D., 2014. Do Superbowl Ads Affect Brand Share? SSRN Electronic Journal. doi:10.2139/ssrn.2385058

Appendix A. Code

Data reading and data description

Data reading and data description are the first steps in data analysis. Before reading the data, we first need to determine the source and location of the data. The data can come from databases, data warehouses, text files, spreadsheets or other formats. Once we find and read the data, we need to spend time fully understanding and describing the data. This includes checking data types, possible values, and data structure. We need to determine if there are any missing or incorrect values, duplicates or outliers. We also need to understand the definition and meaning of each variable and the relationships between them.

The process of describing the data requires us to draw summary statistics of the data, such as mean, median, maximum and minimum values. We can make joint frequency tables and interaction plots between variables to better understand the breadth and depth of the data. Visual representations of the data, such as scatter plots, box plots and histograms, can provide an intuitive expression of the data distribution. Data reading and description are important first steps to any data analysis or modeling. Taking the time to thoroughly understand the data can help us better clean, transform and model the data.

Data reading and data description are important starting points for getting data analysis work done. Before proceeding with actual statistical analysis or machine learning modeling, we must first read, check and describe our data. This helps us gain a deeper understanding of the data so we can make better analytical decisions.

Table A1 Most frequent type (descending order)

Figure A1 Code of Read Data and Most frequent type

Figure A2 Code of the Brands appear every year and Describe

The code for drawing data charts

Drawing data charts is a key part of data analysis and visualization. There are many options to create different types of charts, such as bar charts, pie charts, scatter plots, boxplots and line charts. For any chart, we first need to prepare the data - this usually includes reading data files, cleaning the data and calculating any necessary summary statistics.

Once the data is ready, we can use various programming languages and data visualization libraries to draw charts. For example, using Python, we can use the Matplotlib and Seaborn libraries. To draw a bar chart, we can call plt.bar() and pass the data for the x-axis (bar width) and y-axis (bar height). To draw a pie chart, we can use plt.pie() and pass a list containing the values for each pie chart slice.