标签：Visualization Reveal Data Florida chart sales total data

Visualization Upgrades Reveal Hidden cosmetic trends in Florida

December 11, 2024 by Jingyu Guo, Yingqi Liu, Zhenyu He, Qinya Zhu

Data visualizations are an essential tool for communicating complex information in a clear and impactful way. However, not all visualizations achieve this goal. In fact, poor visualizations are surprisingly common and can do more harm than good.

Imagine staring at a chart cluttered with unnecessary elements, inconsistent scales, or confusing colors. Instead of gaining clarity, you’re left with frustration and confusion.

This blog addresses these challenges with a step-by-step guide to identifying and fixing common flaws in visualizations. By the end, you'll learn how to transform unclear charts into intuitive and impactful visual stories. Let’s move from chaos to clarity!

Table of Contents

Visualization Upgrades Reveal Hidden cosmetic trends in Florida

1. Understanding the Importance of Good Visualization

Bad data visualization can mislead the reality of data. As Edward Tufte noted in The Visual Display of Quantitative Information, poor visualizations often have these issues:

A. Chartjunk: Unnecessary elements that distract and fail to convey information effectively.
B. Distorted Proportions: Designs mislead by not displaying accurate data proportions, like through non-zero baselines.
C. Misleading Use of Visual Variables: Improper use can confuse, with inadequate color contrast or semantically misaligned colors.
D. Low Data-Ink Ratio: "Data ink" should dominate, not be overshadowed by excessive decoration.
E. Overloading or Oversimplification: Too much or too little information can hinder key insights communication.

Suggested reading: The Visual Display of Quantitative Information by Edward Tufte (1984)

2. Challenges in Enhancing Cosmetics Data Visualization

Context of the chart : US Cosmetics Industry

The U.S. cosmetics industry, as the largest market in the world, which has reached $15.53 billion in 2022, offers significant business opportunities. Increased demand for convenient color makeup, rising disposable income, and influencers influence are driving the growth. The beauty industry creates jobs, shapes beauty trends, and influences consumer behavior and social culture around the world.

In this blog, we use a chart from AnyChart as an example. Its source is: https://www.anychart.com/products/anychart/gallery/3D_Bar_Charts/Stacked_3D_Bar_Chart.php?theme=darkBlue

The chart is shown below:

bad example

The original image presents the sales situation of cosmetics in different states in the form of a stacked bar chart. It is divided into five sections, depicting the specific sales data of the top 10 products ranked by total sales in Florida, Texas, Arizona, and Nevada. It helps the audience understand which products are more popular in different regions.

This essay analyzes sales data from Florida, Texas, Arizona, and Nevada, highlighting distinct top-ten product characteristics and regional market demands. Recognizing these differences enables cosmetics companies to tailor marketing strategies, enhancing local consumer service and business growth.

Key impacts:

Optimized Production & Inventory: Match production and stock to sales.
Targeted Marketing: Develop region-specific campaigns.
Consumer Insights: Use data for demand forecasting and product refinement.
Economic Benefits: Promote economic growth and job creation.
Social Responsibility: Improve efficiency and support sustainability.

How to Read the chart

The vertical axis shows product types and lists the top 10 cosmetics by total sales from top to bottom
The horizontal axis represents sales in US dollars. In the same row, each column represents the sales of that cosmetic product in a particular state, and when they are stacked together, they show the total sales of that cosmetic product in the four states.
Each row, from left to right, represents Florida (light blue), Texas (dark blue), Arizona (light purple), and Nevada (dark purple) in a fixed order. The different colored sections represent different states.

Visual Variables

Color: Different colors distinguish the data of each state.
Length: The total length of the stacked bars indicates the total sales of each product type, while the length of a single color represents the sales of a specific state. The longer the length, the greater the sales.
Position: The arrangement order of each cosmetic bar in the chart indicates the ranking of that cosmetic's total sales.

Problems in the Chart

Do you think this chart is already perfect? No, it still has many flaws.

A. High Cognitive Load:

Stacked bar charts are hard to read, making interstate and intrastate comparisons difficult.
Ambiguous titles like “products” and “revenue” cause misunderstandings and increase cognitive load.

B. Low Data-to-Ink Ratio:

Unnecessary 3D effects, grids, and lines add clutter, reduce clarity, and constitute Chartjunk.

C. Indirect Look-Up:

Legends require back-and-forth referencing, making data harder to process.
Bar lengths lack precise labels, making comparisons difficult.

D. Design Issues:

Vertical and small text is hard to read.
Similar colors for bars may cause confusion.

These issues obscure insights, hindering effective sales strategies and consumer satisfaction.

3. Step-by-Step Improvement

Improvement Process Overview

To address the issues mentioned above, we will tackle them step by step:

By dividing the stacked bar chart into four 2*2 subplots (one for each state), using a 2D format, a soft purple color scheme, and removing the 3D effects and deep background, we simplified the chart. Then, we adjusted it to a horizontal layout, added a total sales bar chart on the left side, standardized the scale to enhance comparability, and optimized details like increasing the title font size, labeling specific values, and rotating the axis labels. These changes ultimately improved the clarity and readability of the chart.

Below is the detailed improvement process.

Step1. Simplify the Design: Stacked Bar Chart → Four Subplots

This step involves dividing the stacked bars into four subplots by state. The specific improvements are as follows:

Removed some Non-Data-Ink: replaced the unnecessary 3D effect with a 2D chart and removed the deep background color.
Divided the stacked bars into four 2*2 subplots, with each subplot representing the cosmetic sales by state, which allows for an intuitive comparison of sales within states and between states.
The original color scheme was blue and purple. We changed it to pink and purple, which aligns more with the public's perception of cosmetic colors, thus reducing cognitive load.

Step2. Enhance Comparability: Arrange Subplots Horizontally and Add "Total" Column

This step mainly involves changing the layout of the subplots to a horizontal arrangement and adding a total sales bar chart. The specific improvements are as follows:

The original 2*2 subplot layout had low data ink, was not intuitive, and did not show the total sales. Therefore, we changed the layout to horizontal and added a gray total sales bar chart on the far left.
Removed some Non-Data-Ink: deleted unnecessary gridlines and borders.
Used state names as subplot titles for clarity.

Step3. Standardize Axes: Normalize Axes for Consistency and Eliminate Misinterpretation

This step is about standardizing the scales for all subplots, allowing for direct comparison of the bar lengths between them. The specific improvements are as follows:

Used a consistent scale for each subplot (equal lengths represent equal values), which eliminates visual traps caused by different scales.
After standardizing the scales, adjusted the maximum value on each subplot's x-axis according to the maximum value in that subplot. This adjustment made the bar width occupy the maximum proportion of the subplot width, reducing empty space and increasing the effective information ratio.
Rotated the vertical axis labels (Cosmetic) to horizontal for easier reading.

Step4. Improve Readability: Fine-Tune Titles, Colors, and Other Design Elements

This step focuses on improving various details of the chart. The specific improvements are as follows:

Updated the title with a more accurate one: Distribution of Top 10 Cosmetics Sales in Four States.
Made the title font bold and larger, with slightly larger font sizes for subplot labels and axis labels. This makes it easier for the viewer to quickly grasp the key information.
Added value labels: Placed the specific values on the right end of each bar on the x-axis, eliminating the need for indirect referencing and reducing the reader's cognitive load.
Removed the x-axis: To avoid redundant information, we removed the x-axis.

4. Insights into Enhanced Visualization

How to Read the Enhanced Chart

The chart shows the top 10 cosmetics by total and state sales.

The leftmost subplot represents total sales; the others show state-specific sales.
Cosmetics are sorted by total sales and share a vertical axis.
A consistent horizontal scale allows direct comparison across states.

The Visual Variables

This visualization uses the following variables:

Color: Differentiates total sales and state-specific data.
Length: Bar length represents sales revenue, the key quantitative variable.
Position: Bars represent cosmetics, ordered by sales, with horizontally arranged subplots for state comparisons.

What is the Story?

Florida's lip gloss sales stand out significantly, reaching around $20,000, far surpassing lipstick sales at $5,000. While other states show balanced distributions, Florida's exceptional lip gloss sales make it a noteworthy case for analysis.

Why does this occur?

A. Beach Culture: Florida's beach lifestyle makes vibrant, lightweight lip gloss ideal for quick, lively makeup.
B. Convenience: Lip gloss is simple and quick to apply, suiting busy lifestyles.
C. Personalization: Floridians value freedom and variety, with lip gloss offering more options than lipstick.

These cultural factors contribute to the significantly higher sales of lip gloss.

Impact:

Insights from the visualization can guide recommendations for Florida's cosmetics sales:
A. Increase Inventory: Boost lip gloss stock to meet high demand.
B. Diversify Products: Offer more colors, lip care, and sunscreen products.
C. Promotional Campaigns: Launch seasonal events to attract consumers and tourists.
D. Boost Other Products: Bundle mascara and eyeliner with lip gloss or offer makeup tutorials to increase interest.

More insights beyond those mentioned above:

A. Total Sales: Powder is the most popular, showing high demand for base makeup. Foundation and eyeliner also perform well, while nail polish sales are low, likely due to preference for professional nail services.

B. Texas: Eyeshadows lead sales, reflecting a trend toward fashionable looks. Introduce more colors and styles to align with consumer preferences and enhance promotion.

C. Arizona: Overall sales are low, with no standout products, likely due to weak marketing. Conduct market research to understand local preferences and adjust product offerings to meet diverse needs.

D. Nevada: Trends favor defined brows and nail decoration, boosting eyebrow pencil and nail polish sales. Offer varied colors and textures to match consumer preferences and makeup styles.

5. Conclusion: The Value of Effective Data Visualization

This project showcases the importance of improving flawed visualizations to uncover meaningful insights. By breaking down a complex stacked bar chart into simpler subplots, standardizing scales for clear comparisons, and adding storytelling elements, we turned an initially confusing chart into an intuitive and impactful tool for analyzing cosmetics sales.

Key-home Messages and Next Steps

Simplify designs for clarity, ensure consistency in variables, and use storytelling to highlight key insights. Moving forward, focus on interactive visualizations, real-time data integration, and advanced analytics like trend prediction.

Reference

Brendan Conway-Smith, Metacognitive Proceduralization.
http://cognitivemodelinglab.com/index.php/proceduralization/
Jiajing Guo, How Is Data Visualization Influenced By
Our Cognitive Processes? 2017.
https://medium.com/@jiajingguo/how-is-data-visualization-
influenced-by-our-cognitive-processes-281d8486abfe
Colin Ware. Information Visualization – Perception
for Design, 2013.
J. J. van Wijk, "The value of visualization," VIS 05. IEEE Visualization, 2005., Minneapolis, MN, USA, 2005, pp. 79-86, doi: 10.1109/VISUAL.2005.1532781.
Danielle Albers Szafir, The good, the bad, and the biased: five ways visualizations can mislead (and how to fix them), 2018, https://doi.org/10.1145/3231772.
Leo Yu-Ho Lo, Ayush Gupta, Kento Shigyo, Aoyu Wu, Enrico Bertini, Huamin Qu, Misinformed by Visualization: What Do We Learn From Misinformative Visualizations? 2022, https://doi.org/10.1111/cgf.14559.
H. K. Bako, X. Liu, L. Battle and Z. Liu, "Understanding how Designers Find and Use Data Visualization Examples," in IEEE Transactions on Visualization and Computer Graphics, vol. 29, no. 1, pp. 1048-1058, Jan. 2023, doi: 10.1109/TVCG.2022.3209490.
Bako H K, Liu X, Battle L, et al. Understanding how designers find and use data visualization examples[J]. IEEE Transactions on Visualization and Computer Graphics, 2022, 29(1): 1048-1058.

标签：Visualization,Reveal,Data,Florida,chart,sales,total,data
From： https://www.cnblogs.com/qinya/p/18605488

Visualization Upgrades Reveal Hidden cosmetic trends in Florida

Visualization Upgrades Reveal Hidden cosmetic trends in Florida

1. Understanding the Importance of Good Visualization

2. Challenges in Enhancing Cosmetics Data Visualization