This document provides an overview of commonly used data visualization types and best practices for designing effective visualizations. It discusses chart types like bar charts, histograms, line charts, area charts, pie charts, scatter plots, bubble charts, box-and-whisker charts, and heatmaps. For each type, it provides examples of appropriate uses and tips on how to design the visualizations for clarity and to tell a story with the data. The document emphasizes selecting the right chart for the data and audience and designing for simplicity and functionality over just aesthetics.

1. DATA VISUALIZATION

2. Why visualize data? It is a good way to
communicate complex information, because we
are highly visual animals, evolved to spot
patterns and make visual comparisons. To
visualize effectively, however, it helps to
understand a little about how our brains process
visual information. The mantra for this week’s
class is: Design for the human brain!
DATA VISUALIZATION: BASIC PRINCIPLES
CLARITY &SIMPLICITY
• Maximize impact, minimize noise.
• If it doesn’t add value or serve a
purpose, get rid of it.
NARRATIVE
• Don’t just show data, tell astory
• Communicate key insights clearly,
quickly and powerfully
DESIGN & FUNCTION
• Selecting the right type of chart
iscritical
• Beautiful is good, functional is
better, BOTH is ideal

3. THE GOOD,THE BAD
AND THE UGLY
VISUALIZATIONS

4. Why it’s good
• Clean
• simple visualization with
animation overtime
THE GOOD:-

5. Why it’s good
• Dynamic formatting helps
to strengthen the story
THE GOOD:-

6. Why it’s good
• Simple
• intuitive
• custom chart design
THE GOOD:-

7. Why it’s bad:
• All design
• no function
THE BAD:-

8. Why it’s bad:
• Busy
• no clear narrative
THE BAD:-

9. Why it’s bad:
• Misleading chart type
THE BAD:-

10. Why it’s ugly:
• Misleading y-axisscale
THE UGLY:-

11. Why it’s ugly:
• Improper use of percentages
• inconsistent scaling
THE UGLY:-

12. Why it’s ugly:
• Too many elements
• distracting 3D design
THE UGLY:-

13. What type of data are you working with?
Integer, real, categorical, time-series, geo-
spatial, etc.
What are you trying to communicate?
Relationship, comparison, composition,
distribution, trending,etc.
Who is the end user consuming this
information?
Analyst, CEO, client, intern, etc.
THE 3 KEY
QUESTIONS
02
01
03

14. COMMONLY USEDFOR:
• Comparing numerical data across
categories
EXAMPLES:
• Total sales by producttype
• Population by country
• Revenue by department, byquarter
PRO TIPS:
• Use stacked or clustered bars/columns to group by
subcategory or compare multiplemetrics
• Create custom formatting rules to color-code
bars/columns based on theirvalues
BAR &
COLUMN CHARTS

15. COMMONLY USEDFOR:
• Showing the distribution of
acontinuous data set
EXAMPLES:
• Frequency of test scores among students
• Distribution of population by age group
• Distribution of heights orweights
PRO TIPS:
• Adjust the bin size tocustomize the grouping of
values
• Use Pareto Charts to show the cumulative impact of
each bin, ordered bysignificance
HISTOGRAMS&
PARETOCHARTS

16. COMMONLY USEDFOR:
• Visualizing trends over time
EXAMPLES:
• Stock price by hour
• Average temperature by month
• Profit by quarter
PRO TIPS:
• Use linear or polynomial trendlines to visualize
patterns or forecast futureperiods
• Combine line& column charts to trend two variables
on different scales
LINE CHARTS

17. COMMONLY USEDFOR:
• Showing changes in data
composition overtime
EXAMPLES:
• Sales by department, bymonth
• % of total downloads by browser, by week
• Population by continent, bydecade
PRO TIPS:
• Keep the number of unique categories relatively low
(<6) tomaintain clarity
• Use data validation and custom formatting to
dynamically highlight specific dataseries
AREA CHARTS

18. COMMONLY USEDFOR:
• Comparing proportions
totaling100%
EXAMPLES:
• Percentage of budget spent bydepartment
• Proportion of internet users by agerange
• Breakdown of site traffic bysource
PRO TIPS:
• Keep the number of slices small (<6) tomaximize
readability
• Use a donut chart to visualize more than one series
at once, or usetransparent
• segments to create a custom “race track”
visualization
PIE &
DONUT CHARTS

19. COMMONLY USEDFOR:
• Exploring correlations or
relationships betweenseries
EXAMPLES:
• Number of home runs and salary by player
• Ice cream sales and average temperature by day
• Hours of television watched byage
PRO TIPS:
• Add atrendline or line of bestfitto quantify the
correlation between variables
• Remember that correlation does not imply
causation
SCATTER PLOTS

20. COMMONLY USEDFOR:
• Adding a third dimension (size)to a
scatterplotformat
EXAMPLES:
• Product sales (X), Revenue (Y), and MarketShare(size)
by Company
• Income per Capita (X), Life Expectancy (Y)
and Population (size) by Country
PRO TIPS:
• Use color as a fourth dimension to differentiate
betweencategories
• Use cell formulas and form controls to create a
dynamic, animated bubblechart
BUBBLE CHARTS

21. COMMONLY USEDFOR:
• Visualizing statistical
characteristics across dataseries
EXAMPLES:
• Comparing historical annual rainfall acrosscities
• Analyzing distributions of values and identifyingoutliers
• Comparing mean and median height/weight bycountry
PRO TIPS:
• By default, quartiles are calculated by excluding the
median; this calculation can be adjusted to
include the median, but may significantly change
the result (particularly forsmaller data samples)
BOX &
WHISKER CHARTS

22. COMMONLY USEDFOR:
• Visualizing trends or relationships
using colorscales
EXAMPLES:
• Accident rates by time of day and day of week
• Average temperature by city, bymonth
• Average sentiment byhashtag
PRO TIPS:
• Use intuitive color scales (i.e. red to green) and
apply custom formatting to hide cell values (;;;)
• Use data validation and cell formulas to create
dynamic heat maps based on user-enteredvalues
HEATMAPS

23. Thanks