-
-
Save AdamMescher/4934fee9e8f7dda605f551f60a03d4bd to your computer and use it in GitHub Desktop.
- [Project Oxygen HBR Case Study]
You are not the hero who will save the audience; the audience is your hero.
You must defer to your audience because if they don't engange and believe in your message, you are the one who loses.
Without their help, your idea will fail.
Power Corrupts. PowerPoint Corrupts Absolutely."
- Edward Tufte, Yale Professor Emeritus
Having all the information in the world at our fingertips doesn't make it easier to communicate: it makes it harder. The more information you're dealing with, the more difficult it is to filter down to the most important bits.
- Focus on the message - lose the fancy clipart, graphics, and fonts
- tell a clear story, don't just make a pretty chart
Nobody sets out to make a bad graph. But it happens, again and again.
Why?
No one teaches us how to tell stories with numbers. Few feel adept in this space.
Being able to visualize data and tell stories with it is key to turning it into information that can be used to drive better decision making.
In absence of training, we rely on available tools.
Anyone can put some data into a graphing application and create a graph.
- This is remarkable.
- This is scary because without a clear path to follow, best intentions can lead the user astray.
Tool defaults and general practices tend to leave our data stories lacking.
There is a story in your data. Your tools don't know what that is.
An effective data visualization can mean the difference between success and failure.
Data visualization is only a single step in the data analysis process that your audience ever sees.
Trial and error
Perhaps this publication can help push past pain points.
Anyone who needs to communicate something to someone using data.
- Understand the context
- Choose an appropriate visual display
- Eliminate clutter
- Focus attention on where you want it
- Think like a designer
- Tell a story
Take the time to learn your tool well so that it does not become a limiting factor when it comes to applying these lessons
Basic proficiency with the software is assumed.
What you can do above and beyond that will set you apart.
Being abble to tell stories with data is one area that will give you that edge and position you for success in any role.
Success in data visualization does not start with data visualization.
Before you begin creating visualization or communication, attention and time should be paid to understanding the context for the need to communicate.
Exploratory: what you do to understand the data and figure out what might be noteworthy or interesting to highlight to others.
Example: Hunting for pearls in oysters.
Explanatory: you have a spcific thing you want to explain, a specific story to tell
Example: Displaying the two pearls found during exploratory phase
- Who - To whom are you communicating?
- What - What do you wnat your audience to know or do?
- How can you use data to help make your point?
The more specific you can be about your audience, the better position you will be in for sucessful communnication.
Avoid general audiences, such as "internal and external stakeholders" or "anyone who might be interested".
By trying to communicate to too many different people, you can't communicate to any one of them as effectively as targeting your audience.
What is your (my) relationship with the audience?
Do they trust me as an expert or do I need to establish credibility?
What do you need your audience to know or do?
You should always want your audience to know or do something.
IF you are the one analyzing and communicating the data, you likely know it best - you are the subject matter expert.
Take a confident stance when it comes to making specific observations and recommendations based on analysis.
Start doing it now - only gets easier with practice and time.
If not your place to make final decisions, suggesting possible next steps can be a great way to get conversation going because it gives your audience something to react to rather than starting with a blank slate.
If you ask for action, your audience must make a decision whether to comply or not.
Accept, Agree, Begin, Believe, Change, Collaborate, Commence, Create, Defend, Desire, Differentiate, Do, Empathize, Empower, Encourage, Engage, Establish, Examine, Facilitate, Familiarize, Form, Implement, Include, Influence, Invest, Invigorate, Know, Learn, Like, Persuade, Plan, Promote, Pursue, Recommend, Receive, Remember, Report, Respond, Secure, Support, Simplify, Start, Try, Understand, Validate
How will you communicate with your audience?
The method used has implications on number of factors, including:
- amount of control you have over how audience takes in information
- level of detail that needs to be explicit
Live Presentation
- presenter is in full control
- what audience sees and when they see it
- Can respond to visual cues to speed up or slow down or go into more or less detail
- Level of detail necessary is low because any questions can be asked directly
Written Document or E-Mail
- presenter has less control
- Audience is in full control of how they consume info
- Level of detail required is much higher because no presenter to ask questions, document must answer frequent questions
Slideument: A single document that's meant to solve both needs.
What tone do you want your communication to set?
What data is available that will help make my point?
Data becomes supporting evidence of the story you will build and twll.
Who: The budget committee that can approve funding for continuation of the summer learning program.
What: The summer learning program on science was a success, please approve budget $X to continue.
How: Illustrate success with data collected through the survey conducted before and after the pilot program.
- What background info is relevant or essential?
- Who is the audience or decision maker? What do we know about them?
- What biases does our audience have that might make them supportive of or resistant to our message?
- What data is available that would strengthen our case? Is our audience familiar with this data, or is it new?
- Where are the risks: what factors could weaken our case and do we need to proactively address them?
- What would a successful outcome look like?
- If you only had a limited amount of time or a single sentence to tell your audience what they needed to know, what would you say?
- Knowing the desired outcome is before you start preparing the communication is critical for structuring it well.
- Putting a significant constraint on the message (a short amount of time or a single sentence) can help you to boil the overall communication down to the single, most important message.
Idea is to boil the "so-what" down to a paragraph and ultimately to a single concise statement.
If you only had three minutes to tell your audience what they need to know, what would you say?
Boils "so-what" down to a single sentence.
Has three components:
- It must articulate your unique point of view
- It must convey what's at stake
- It must be a complete sentence
The storyboard establishes a structure for your communication.
It is a visual outline of the content you plan to create.
When you can, get acceptance from client or stakeholder at this step to ensure what you're planning is in line with need.
Don't start with presentation software.
- Easy to lose track of forest through the slide trees.
- Sunk cost fallacy.
Start low-tech (post-its, whiteboard, plain paper, etc)
Being able to concisely articulate exactly who you want to communicate to and what you want to convey before you start to build content reduces iterations and helps ensure that the communication you build meets the intended purpose.
Understanding and employing the 3-min story, big idea, and storyboarding will enable you to clearly and succinctly tell your story and identify the desired flow.
Many types of graphs but a handful will work for the majority of your needs.
Author uses only a dozen different types of visuals.
The fact that you have some numbers does not mean you need a graph!
With just a number or two to share, simple text can be a great way to communicate.
Tables interact with our verbal system - we read them.
Design elements should fade into the background.
Data should be what stands out
- Borders should eb used to improve legibility of your table
A heatmap is a way to visualize data in tabular format, where in place of (or in addition to) the numbers, you leverage colored cells that convey the relative magnitude of the numbers.
Use color saturation to reduce congnitive load by providing visual clues.
Graphs interact with the visual system - we see it and process it faster
Types of graphs must frequently used by the author:
- Points
- Lines
- Bars
- Area
Can be useful for showing the relationship between two things because they allow you to encode data simultaneously on a horizontal x-axis and vertical y-axis to see whether and what relationship exists.
Tend to be more frequently used in scientific fields and are thus sometimes viewed as more complicated.
FIGURE 2.6 FIGURE 2.7
Most commonl used to plot continuous data.
Implies a connection between the points that may not make sense for categorical data.
Continuous data is often in some unit of time:
- days
- months
- quarters
- years
Two frequntly uses charts:
- line graph
- slopegraph
Can show a single series of data or multiple series of data.
Be consistent in the time points you plot.
Can also convey sense of range by showing someting like average line within min and max area color.
Can be useful when you have two periods of comparison and want to quickly show relative increases and decreases or differences across various categories between two data points.
Pack in a lot of information
- Absolute values
- Intuative visual increase or decrease in rate of change without having explain what "rate of change" is
Can take a bit of patience to set up as they aren't usually included as a standard graph.
Lines work well to show data over time, bars tend to be "go-to" graph for plotting categorical data
Should be leveraged because they are common - people naturally understand them.
Bar charts must have zero baseline to appear unbiased.
If you don't
- MAKE IT CLEAR THAT YOU ARE USING NON-ZERO BASELINE
- take context into account so that you don't overzoom and make minor changes or differences appear significant
Misleading in this manner by inaccurately visualizing data is not OK.
Bars should be wider than the whitespace between bars.
"vanilla" bar chart
Can be single series or multiple series.
As you add more series of data, it becomes more difficult to focus on one at a time and pull out insight.
Visual grouping happens as a result of spacing thus relative order of the categorization is important.
Meant to allow you to compare totals across categories and also see the subcomponent pieces with a given category.
Can quicly become visually overwhelming as it is hard to compare subcomponents across the various categories once you get beyond th bottom series because you no longer hav ea consistent baseline to use to compare.
Can be structured as absolute numbers - 100% bar
think about whether it makes senseto also include the absolute numbers for each category total which may aid in th interpretation of the data.
can be used to pull apart the pieces of a stacked bar chart to focus on one at a time, or to show a starting point, increases and drecreases, and the resulting end point.
Author: "If I had to pick a single go-to graph for categorical data, it would be the horizontal bar chart.
Why?
Because it is easy to read. Eyes hit category name before the bar.
Can be single or multiple series.
Canbe structured to show either absolute values or some to 100%
Can work well for visualizing survey data collected along a Likert scale
Generally avoided as the human eye doesn't do a good job of attributing quantitiative value to 2D space.
Use when need to visualize numbers of vastly different magnitudes.
Choose a graph type that will enable you to clearly get your message across to your audience. With less familiar visuals, you likely need to take extra time to make them accessible and understandable.
- Pie charts
- Donut charts
- 3D
- secondary y-axis
Replace pie chart with horizontal bar chart
only exception is if you are actually plotting a third dimension
introduces unecessary chart elements like side and floor panels and does weird things when plotting values.
1Think about whether following approaches will meet needs:
- Don't show the secondary y-axis. Instead, label the data points that belong on this axis directly.
- Pull the graphs apart vertically and have a seperate y-axis for each (both along the left) but leverage the same x-axis across both.
In many cases, there isn't a correct visual display.
Most important question: What do you need your audience to know?
Choose display that will enable to make that clear.
If wondering What is the right graph for my situaiton?
Answer is the same - whatever will be easiest for your audience to read.
Easy way to test this is to create visual and show to friend or colleague and ahve them articulate the following as they process the information:
- Where they focus
- What they see
- What observations they make
- What questions they have
Amazing post.