Your Communication Toolbox: Visualizations, Logic, and Stories

Visualization Science

Using Visualization Science to Influence Business Decisions

• What people is tightly intertwined with what people decide.
• What you can do to your graphs to draw peoples' eyes to where you want them to be.

两个需要避免的操作：

1. Default formatting
2. Visually busy slides

Will leaving decisions to chance, giving up your power to direct your audience's attention and gaze

一定要用到的操作：
Visual contrast is one of the best ways to attract people's gaze to data related images.

Storyboarding Your Data Story

The Storyboarding Hourglass

will learn: How to structure a business presentation
Hourglass:

• start：wide
• detail： narrow
• end(recommendation): wide
• surrounding the data you show with motivation context

How to do?

1. Why? So what?
   • In first 30-60s: introduce the business problem you will solve
   • Use compelling story element(引号” ，dramatic pictures， actual stories about someone involved in problem) to draw people in by eliciting emotion and motivation (emotional hooks).
   • 让你的听众在一开始就有兴趣听你讲话是非常重要的，从某些方面来讲是最重要的
   • “So what?”比“What？”更重要
2. SMART recommendation
   • Present to your audience the recommendation in a format, for how you are going to solve the business problem they are now very emotionally invested in.
   • In business, your take-home message needs to come very early in your presentation (people want to know that their time with you is going to be well spent as soon as they enter the room)
3. What am I going to show
   • Agenda slide could go here
4. Supporting evidence
   • Show the thoughtfully organized during storyboarding process
5. Link evidence to SMART recommendation
6. What you could gain
   • Show the strong benefits that will result if your recommendation is followed
7. Why? So what?
   • Reenergize the audience's enthusiasm and motivation by using story elements again
   • Leave your audience with a strong, positive emotion
8. Definitely be well prepared for all the questions in slides of an appendix at the end of presentation
   • Answer how you would address potential pitfalls
   • Articulate what assumptions you made when design your recommendation
9. Also prepare a Technical Details White Paper
   • A written handout with large amount of information rather than slides
   • Wait to hand the white paper out after the presentation

Making Your Data Story Come Alive

Story Element

故事的四要素：

1. Characters
   • Can mot be a person, rather a thing
   • The most effective way to draw your audience in is to describe the experience of a character related to your problem or location related to your problem. Using details that make your audience actually picture the person or place in their mind.
   • Another particular powerful method is to tell a story about your own experience.
2. Location
3. Conflict
4. Resolution

其他方式开头：

• 互动
• 引用其他人说的话

Tips and Tricks

1. Try starting in the middle of the story
   The character is in the mid of action, which create a sense of momentum and expectation for what will come next)
2. Use details
   When possible, describe your characters and give them texture. Keep you description short but use specific details, which add credibility and improve belief.
3. Stick to positive emotions
   When possible, use positive message stories rather than negative stories.
4. Use large, high-quality pictures

Storyboarding Your Presentation

Storyboarding is the process of identifying the key scenes in your processor story and putting them in a logical order that conveys your message compellingly.
Storyboarding is, basically, a way of having a plan for your presentation that you can articulate and communicate with others. So that you can get feedback and work in the presentation together.

Reasons to Storyboard

1. Clarifies your logic of your hypothesis and assertions
2. Highlight the gaps in your logic
3. Facilitates communication
4. Streamlines slide-making
5. Companies are starting to ask for storyboards instead of traditional dashboards

Difference between storyboarding and analysis plan

1. Trying to narrow in on the minimum number it seems necessary to convey your data story, rather than include any scene that might be important
2. Think about what kinds of visualizations were communicating information in each scene, rather than what kinds of visualizations will help you analyze vast amount of information
3. A critical aspect of your presentation storyboard will be the precise order, how to organize your scenes and how you link them together for your audience

Physical process of storyboarding with Post-it

1. Brainstorming: writing down every single insight you discover during your analysis, you think was important at helping you arrive at your conclusion and business recommendation. Each insight needs to go on its own Post-it, index card or on its own box in your software program.

   Each point or detail on a Post-it should be considered a story point, that will eventually get its own graph and its own slide. Each story point should be able to be summarized in one sentence. If it takes more than that, it's too complicated and should be broken down into separate story points.

2. Whittle down your story points until you have only the ones that are absolutely critical for justifying your recommendation for a business process change.

   As a rule of thumb, you should try to have no more than three main story points. And each of those main story points should have no more than three sub-points. Three items is about the limit of complexity most people can handle in one sitting.

   So, each point you add over that needs to add so much value that it will be able to overcome the lack of attention your audience is likely to exhibit. Be warned, this narrowing down step is deceptively difficult. Which illustrates how difficult it is to make a streamline presentation in the first place and why it takes data teams so long to make them. There's also where you will get the most bang for your buck, though. So, it's worth giving it considerable energy and attention.

3. Organize the points according to the order which you are planning on telling them.

   The order should reflect what you believe is the most compelling logical argument you can give to support your recommendation. Not the order in which you actually did the analysis. Remember, this is a presentation about what they should do. Not a presentation about what you did over the past 6 to 12 months.
   Here are a few guidelines to help you figure out your story point order:

   • when your recommendation IS NOT controversial:
     First story point = most compelling story point
     Get your audiences buy in as quickly as possible
     The art of persuasion isn't the same as the art of entertainment, even though they sometimes overlap.
   • when your recommendation IS controversial:
     First story point = least controversial story point
     People are more likely to be persuaded by an argument if you get them into a general feeling of agreement first. So by starting with your least controversial point, you will get your audience used to saying yes before you hit them with something they might wanna say no to.
   • Make sure you show the data necessary for being convinced by an argument, either before or at the same time as making the argument.
   • Draw a quick sketch on the Post-it of what kind of visualization will convey each story point best.
     This is a good point in the process to remind yourself the primary context surrounding your presentation is your audience. When you write down your story points, make sure they are at the appropriate level detail for your audience(parameters on your models, your confidence intervals, or pretty much any other details you've had to worry).
   • Share your storyboard with as many people as you can before you present to the people who will ultimately decide the verdict of your recommendation. In particular, you want to ask other people to check the logic of your argument and make sure it is sound. I call this, stress testing your story.

Stress-testing Your Story

The Best Stress-testers are Teams

need to take classes or read books of reason and logic

Common Errors to Avoid

Overgeneralization and Sample Bias

• Sample bias and data missing will lead to wrong decision.

Tips and Tricks

1. Ask questions of how your data was collected (collection methods)
   Listen to hints about how the collection methods might have biased your data and look for ways you test the data you have for bias against specific demographic groups.
2. Always check how many data points you have in all the groups
   If you don't have many data points from a specific group, subcategory, or time point, don't put much weight in the effects you see there.
3. Test subsets of data for consistency
   If have a lot of data, split your full data set into three to five random subsets, see if you observe the same effects in each one as you see in a group as a whole. If not, take caution when interpreting the results from the group as a whole. It's likely that the effect is either due to chance, isn't that large, or it is only found in a certain subset of your data that you should track down and characterize.
4. Look for common characteristics of outliers and missing data
   Do that before removing the outliers and missing data, if the data have something in common, you will wanna try to collect more data with those characteristics to fill in what you will then exclude. At the very least, you will wanna be aware how you are biasing your result when you remove those entries from your data set.

Misinterpretations Due to Lack of Controls

设置合适的对照组将使我们对此项目有更加正确的认识

• Always include carefully designed comparison groups that should not have the effect you were looking for in your analyses to make sure the effects you observe are due to the events you think they are due to.
• If a key part of your data story does not have a control group, go run some control analyses, it's the only way to make sure you are interpreting your data correctly.

Correlation Does Not Equal Causation

两条曲线看起来相关不一定代表两件事之间互为因果
Concepts:

1. correlation 相关
2. causation 原因
3. coincidence 一致
4. spurious relationship 伪关系

The coincidence or spurious relationship we observed might due to a variable we didn't pay attention to.

How Correlations Impact Business Decisions

What to do to know how much confidence you should have in the correlation you see:

Tip 1:

1. Ask whether there is any other third or fourth or fifth variable that can explain the relationship you see
2. Look for data that allows you to test whether that third or forth ofr fifth variable is a better measure of the phenomena you are interested in.
3. Examine whether the correlation you're basing your business recommendation on exists in other contexts or datasets. The more you can duplicate the effect, the less likely the first correlation you saw was due to random chance.

Tip 2:

• Try to come up with different, complementary angles to assess the causal relationship you're hypothesizing about.

  Example: If your hypothesis was that more security engineers causes more security breaches, do you see more increases in security breaches when security engineers are fired? In addition, if the number of security engineers causes more security breaches, increases in security breaches should happen after more security engineers are hired. Not before they are hired. Does your data give you enough resolution to address that question? If not, try to get data that does have high enough time resolution.

Other correlation does not equal the causation issue

1. The likelihood that you will get into trouble inferring causation from correlation, increases as the size of your data sets increase.
The more data you have, the more opportunity you will have to find coincidental relationships that just happen by chance.

2. The likelihood for getting into trouble increase as the complexity of your data sets increases as well.
When many variables are highly related, you can get some strange effects, where sometimes a variable you are interested in correlates with an increase in a metric you care about. But other times the same variable correlates with a decrease in the variable you care about. These seemingly contradictory effects are due to what's happening with the other variables you may or may not be measuring. I've seen seemingly contradictory effects a lot in my own research, and you are likely to see them as well. I don't wanna go into the statistical details of why this happens, but I do want you to remember that the bigger and more complex your data set, the more aware you should be of investing a lot of capital in a single correlation.

Two situations in which you don't care as much if correlation represents causation is when you are trying to measure phenomenon that you don't have a reliable way to measure directly or when you are trying to simply predict how likely something is to happen. If a correlation between two variables is consistent and reliable, one variable can be used to both measure and predict the other even if one doesn't cause the other.

The problem is that if you don't know why one variable is correlating with another it's hard to anticipate when they will stop correlating with each other.

So if a business is going to strongly invest in a correlational phenomenon they don't understand the cause of, they need to also be prepared to invest in the infrastructure necessary to continuously monitor the correlation, and to make the adjustments if necessary. The recommendations you suggest to your stakeholders should reflect these principles.

Tools for Conveying Your data Story

Choosing Visualizations for Story Points

Stick to bar and line graphs

When you're giving business presentations bar charts or line charts will almost always be a good way and maybe even the best way to communicate the implications of your analysis. This is even true, maybe even more true when you're running very complicated statistical models. I suggest using bar charts and line charts as your defaults, and only considering other types when you have had some more time to dig into the details of visualization science.
Leave the keeping things interesting part to the way you motivate your presentation with story elements, the actual content of what you were telling your audience, and how you format your non data slides.

1. Bar charts

• Should be used for comparing measures in different groups or categories.
• Better to show aggregated rather than raw data.

2. Line charts

• Should show variables and categories vary over time or any ordered category.
• One danger with line charts is that our eyes naturally follow the lines in the chart, and interpret them as if the points in the line have some kind of direct, ordered relationship. So when the points don't have an obvious sequentially relationship, it takes longer for us to understand what the chart is showing, and there's actually a good chance we will misinterpret the point the graph is trying to make.
• Line charts should be made into bar charts when the ticks on the x-axis don't a specific, inherent, sequential order.

3. Pie charts

• You are communicating categories that add up to 100%.
• You are going to highlight no more than 4 categories (Eyes can not detect differences in spatial area).

Do not use: Scatter plots and 3D charts

Scatter plots

• Show the relationship between two variables particularly continuous variables.
• Usually show raw data, aggregated data as well.
• Could show a trend in the picture, but not the data underlying it.
• Do not use until you have technical audience

Never use 3D charts.

The Neuroscience of Visual Perception Can Make or Break Your Visualization

Misinterpretations Caused by Colorbars

Do not use color to：

• Convey detailed quantitative differences in the value of continuous variables.
  Do use color to：
• If you have to use color scale, use grey scale, from black to white, which tend to have more even transitions than other color scales do.
• Use for continuous variables, illustrate very general and obvious patterns.
• Different colors can be used efficiently to represent different categories within a categorical variable.
• Draw attention to something you want your audience to pay attention to.

Visual Contrast Directs Where Your Audience Looks

Best visualization for data analysis ≠ Best visualization to tell persuasive data stories
squint test 斜视测试

• Color contrast can be a very effective way of communicating what you should pay attention to, but not a good way for communicating detailed information.
• If you really wanna call attention to a certain area of the plot, it's good practice to take away the borders of the plot or at least make them the same color as the background of your slide.
• Grey out the unimportant part in the chart.
• Consider whether your chart is truly the best way to describe your story (合适的顺序和对照)

Putting Compelling Data Visualizations into Persuasive Business Presentations

Formatting Slides to Communicate Data Stories

1. Maximize the Data-ink ratio: Take out everything in the chart and everything on the slide that doesn't have a clear and unique purpose.

   • Data ink: ink that represent the actual data, which can't removed without also removing information
   • Non-data ink: everything else!

2. Understanding at a glance:

   • Use full words unless abbreviations are extremely common.
   • Include units.
   • Label directly on the chart instead of using a legend, and try to use the background color of your slide for your labels.
   • Also note that horizontal labels are easier to read than vertical labels.

3. Fonts
   Easy to read:

   • Times New Roman
   • Bodoni
   • Garamond

   Resolution problem or audience sit far away from the screen:

   • Helvetica
   • Calibri
   • Arial

• Test your slides readable from the end of the room, make the elements bigger (the minimum size you should use is 30 points).
• Whenever you finish a presentation, put your slides in sorter mode and reduce their size to 66%. If you can't read them well at that size, your font is probably too small for your audience to read from the back of the room. As a general rule of thumb, if you were worried that your font is too small, then it's definitely too small.

4. Don't make your audience do visual math
   • Make sure all your axis have similar scales
   • Try to use the same limits for your axis across your entire presentation (If you are asked to make a plot with two y axis, such as this one, where the two measurements have completely different types of units that can't easily be put on the same scale. Either transform the two variables to a common scale, such as a percentage so that they can be plotted on a single graph, simply or separate the measurements into two separate graphs that are easy to compare. 两张图上下排列)
   • Use white grid lines to make it easier to see exactly how tall a bar is on a bar graph.
5. Use titles to convey take-home messages
   • Slide titles shouldn't be there unless they add extra information.

Formatting Presentations to Communicate Data Stories

1. Visual formatting
   Consistency:

   • That means you should aim to use consistent formatting throughout your presentation, including everything from slide background to fonts to general color scheme, except when you purposely use a different presentation style for dramatic effect.
   • If you want to liven up your presentation, you can try things like using one background or font for the main points of your presentation and another background color or font for the transition slides or summary slides.

2. Transition Slides

   • 10 Minute Rule: people generally only stay intensively involved in learning experiences for about 10 minutes. After that they start to tune out.
   • To the extent that's true, it's a good idea to build what I call soft breaks into your presentation. During these soft breaks you will either summarize what you just went over, provide something visually engaging to your audience, or shift gears and prepare the audience to go in a different direction.
   • Transition slides can be useful for these soft breaks. The key is to make sure the transitions are smooth, not disruptive.
     • If you were in a more formal presentation, you can use your agenda slide as your transition slide.
     • If you have a little more flexibility, you can use one of the pictures from your story of your introduction or a slight modifications of that picture that correspond with the transition you're about to make.
   • Since it's often a good idea to try to make your transition slides as visually compelling as possible to help wake up your audience, you might try applying the rule of thirds to your transition slides.

   The rule of third says that if you divide a slide up into three equal compartments vertically and three equal compartments horizontally, people consistently look at the intersection of those lines more than the center of the screen.

   The rule of third is not recommended in the data slide, but only the attraction slides (transition slides)

3. Animation

   • Just make sure that when you do use animations, use them with an explicit purpose in mind, and don't use anything too flashy or distracting.
   • Fade in and fade out animations are usually very good, motion paths that allow words move can be good too.
   • checkerboard and spinning usually not work well.

4. Other tips

   • Check your typos! then check again! (at least three times) Nothing undermines your credibility more than typos in your presentation.
   • Bold is more readable than italics or underlining as for emphasizing
   • Always make sure your pictures and figures are high resolution. Nothing should be fuzzy or distorted.
   • Use the same 2 to 3 colors for the basic presentation. Other colors should only be used to highlight something in a particular slide.

Delivering Your Data Stories

1. What you will say
   • Remember to be prepared with things that will get your audience participating.
   • Transition sentences to attract audience's attention: the words you say to connect the content of one side to another (like glue in your story).
   • Don't assume your audience has any specific knowledge, and only use language that everybody in the general audience would understand
2. How you say them
   • Make a conscious effort to convey your enthusiasm and energy for what you've done in your voice.
   • Make sure to vary your voice speed
     • speak slowly: emphasize something
     • speak quickly: convey how incredible excited you are about this extremely impactful recommendation you're about to make.
   • Make sure to vary your tone of your voice
     • higher tone: often approachable
     • lower tone: turns people into the fact that you are saying something very important they really need to pay attention to
3. What you will do before the presentation
   • Practice, Practice, Practice
   • One practice technique that can be particularly effective is to practice just one small piece of your presentation over and over again until you are completely comfortable with it. Then move to practicing another small piece, and keep doing that until you've covered the entire presentation.
4. What you will do during the presentation
   • Make sure you are conscious of your body posture: your posture can affect how your audience responds to you
   • Remain physically open by keeping your arms away from the front of your body. So don't cross your arms. Directly face your audience. And of course, smile.
   • Make sure you look at different people in the room, so that you don't exclude anyone.
   • Don't look down or read your slides.
   • Try to be natural in your movements, to the best of your ability, and don't feel too constrained.
5. What you will do after the presentation
   • You ask for feedback from your colleagues and your peers. Cuz giving presentations is hard and it's even harder to objectively measure how well you did.
   • In the end, the success of your presentation is determined how your audience reacts. So the only way to improve is to hear from audience members what they thought.
6. How you will feel
   Public speaking is nerve-wracking for many people. So much so, that psychologists use it in experiments as a way to actively stress people out. Winston Churchill was reported to have had terrible speaking anxiety, as well as Warren Buffett. But they are great proof that speech or performance anxiety can be overcome. So practice, practice, practice. Confidence is the best antidote to anxiety.