Don’t ever fork your presentation: don’t break it up into two, three, or seven pieces: “Today we are here to discuss Project X. There are seven things I want to tell you about it. Thing 1 . . ., thing 2 . . ., thing 3. . . .” By Thing 4, they are losing interest. You may never get to Thing 7.

In software development, to fork development of a program means to break it into two versions (as in a “fork in the road” of development); subsequent work on one version will therefore be incompatible with the other. Forking is for this reason considered to be a “bad thing” in software development.

Forking is also a bad thing in presentations. Time is linear. Your presentation, although it may appear to meander due to interruptions, will in reality be presented over time, which flows in a straight line. So design it along a straight line. If you have several complications that are all resolved with one resolution, that’ s fine—mention all of them and then present the resolution.

However, if you have several resolutions that are needed to solve one complication, don’t present the complication and then say “and the solution to this has five parts. Let me start with Part 1. . . .” This is too boring. It is not storytelling, it is presenting a list. And, as we saw in Chapter 5, lists are less memorable than stories. Instead, what you do is present the resolution that makes the strongest contribution first, even though it only addresses part of the complication. And then the next complication will be about the part that has not been addressed (e.g., “Yes, but this doesn’t cover our Western business. . . .”), and so your next resolution will address that part, and so on.

What do you do if you feel that you have just an apparent random set of points that you want to make, and you don’ t know where to start? First, you have to understand what the one larger problem is that they all roll up to. Then you find a S.Co.R.E. sequence that covers all your points. At all costs, avoid saying, “I have seven points I want to make.” Too boring.

If you want to take them through, say, a seven-step process, you can do this by trying to get each step to generate a complication that is responded to by the next step. Think about what would happen if Step 3 did not exist. What would go wrong? Whatever that is, it is the complication that Step 3 is a resolution to. An outline for a presentation that describes a process would look something like this:

Objective: to determine whether safety procedures have been adopted correctly Step 1: review procedures

Complication: We do not know whether the safety procedures are being adopted correctly, and this could lead to some serious accidents.

Resolution: We are proposing a process for reviewing safety procedures. It begins with a review of the existing procedures.

Complication: But you can’t know whether the procedures are being adopted if you don’t speak to the people responsible.

Example: details about the kind of conversations that will be had Complication: But the people responsible won’t give you the full story.

In this case, the S.Co.R.E. method allows you to take what could otherwise be a dull series of steps into an interesting story.

What to Do with What Doesn’t Fit into Your Storyline—The Role of the Appendix

The main thing is not to put any information into your presentation that does not have a place on your storyline. If it doesn’ t have a place in your storyline, it does not belong in your presentation—period. We need to be ruthless—i f it doesn’ t fit the storyline, it shouldn’t go in the presentation. If you think it might come up, or if you’ve spent so much time on it and you just want to put it in, then put it in the appendix. Typically, you will bring printed copies of your appendix to the presentation, but you will not hand them out until the end of the meeting, unless a question comes up that requires you to refer to the appendix, at which point you can choose to hand them out.

NO MATTER HOW CAREFULLY you think about your audience or plan your presentation, if you end up presenting fifty slides with seven bullet points of seven words each, you are almost guaranteed to numb your audience to the point of incomprehension. We have all sat through this kind of “Death by PowerPoint.” It is not fun, and it does not work very well, so why do we continue to inflict it on others?

The next two chapters are about the graphical dimension of your presentation—your visual aids. For our purposes, we will define “graphics” very expansively: every two-dimensional representation that is not exclusively text. So charts are graphics, but so are photographs, sketches, diagrams, and cartoons. Graphics can contain text within them: they can be annotated, for example. Graphics can also be mostly made up of text: a table of words, for our purposes, is considered a graphic. The location of each word in the table carries as much information as the word itself. (Of course, the location of a word in a sentence carries as much meaning as the word itself; but in text, the words are always ordered in the same way, one after the other, while in a graphic, elements can be placed anywhere on the page).

Chapter 7 explains how to use each of the components of a slide—charts, graphics, and text—so that they strengthen your communication rather than get in the way of it. Chapter 8 is about how to put all those components together on a slide—how to lay them out on the slide so that your audience “gets” what the slide is about right away, and so that it draws them in to your message.

Before we move on to those chapters, we need to address some fundamental questions that will get to the root causes of “Death by PowerPoint.” These questions are:

• If you use PowerPoint, should the ideal slide have seven bullets with seven words each?

These questions might sound almost heretical, with most people’s response being “Of course!” to all three. The answers are not nearly so straightforward. Research does suggest that visuals can improve presentation effectiveness, but not in all cases, and only if they are used properly. Otherwise they can distract from your message and severely inhibit its communication. There also has been a lot of criticism of PowerPoint— according to its critics, it “dumbs down” communication (e.g., Tufte, 2003a). But are there reasonable alternatives? Is there a way for PowerPoint be used constructively?

¹See Vogel (1986) and Vogel & Morrison (1998). On visual metaphors see McQuarrie & Mick (2003) and Morgan & Reichert (1999).

²In several studies, irrelevant images and details were found to hinder effective communication (Bartsch & Cobern, 2003; Edell & Staelin,

1983; Feinberg & Murphy, 2000; Mayer, 2001; Moreno, 2006; Myers-Levy & Peracchio, 1995; Slykhuis, 2005).

Finally—paradoxically—research shows that the standard advice of seven bullets of seven words each is about the worst approach to slide design. Communication improves dramatically if you either reduce the amount of text to much less than that—or increase it to much more. The next three sections will explain each of these counter-i ntuitive points in turn.

This is a question. You should ask yourself whether your visual aids are adding anything to your presentation. Too often the first act of preparing a presentation is to launch PowerPoint. You now know better: you will go through Steps 1 to 7 out of the ten-step process before you even think about launching any kind of presentation graphics program. The word “presentation” means both to show something and to help people to imagine something in their heads. So technically, whether you use visuals or not, you are still making a presentation.

Research suggests that visuals do help improve presentation effectiveness. One study found that visuals can increase your audience’s understanding of your material, help convince them of your recommendations, and help them remember what you have presented. But not in every case: another study found that using slides with transitions (fades, wipes, etc.) actually gave worse results than using no slides at all, in terms of persuasion. Print advertising research experiments indicate that visual metaphors are more memorable than written ones and more easily understood.¹

However, certain kinds of visual aids can actually cause harm to the effectiveness of your presentation. Projecting slides with text bullet points and/or irrelevant graphics such as clip art during your presentation will likely have worse results than speaking without any visual aids at all.²

Also, if your audience is not engaged in your presentation, perhaps because they have been forced to sit through it and are just trying to get it over with, visual aids can hurt your communication efforts. In such cases, it’s better to toss out the idea of slides and just talk to your audience.¹

PowerPoint is accused of “dumbing down” communication. Graphics guru Edward Tufte, in an essay entitled “The Cognitive Style of PowerPoint,” argued that PowerPoint, as well as programs like it, “reduces the analytical quality of presentations” and that PowerPoint templates “usually weaken verbal and spatial reasoning and almost always corrupt statistical analysis” (Tufte, 2003b). In a subsequent edition of this essay, Tufte concluded that “PowerPoint, compared to other common presentation tools, reduces the analytical quality of serious presentations of evidence” (Tufte, 2006).

In a Wired magazine article that popularized this position (titled “ PowerPoint Is Evil” and subtitled “Power Corrupts: PowerPoint Corrupts Absolutely” S , Tufte (2003a) claimed that “The PowerPoint style routinely disrupts, dominates, and trivializes content.” Several others have also criticized PowerPoint presentations, from politicians to military personnel (Ricks, 2006).

Why should the software be blamed, rather than its users? The problem with a program like PowerPoint, according to its critics, is that it leads the user to develop presentations in a particular way, and this way may not be ideal. This criticism can be applied not just to PowerPoint, but also to other presentation programs, including Apple’s Keynote and Google Docs.⁴

⁴“Software engineers . . . necessarily implement fairly distinct ideas and concepts into any software . . . of how the final product is to be used” (Voswinckel, 2005, p. 45). Presentation software therefore provides “. . . an obvious predisposition towards a certain, default workflow” (p. 50).

Other authorities on presentation have responded that it is possible to create effective slides with PowerPoint or other presentation tools, and therefore that the criticism of these tools is misplaced. While the tools may not be perfect, they assert, these tools do not deserve all the blame that is heaped on them, since there are clear guidelines that can be followed for making effective presentations using them.⁵

Most importantly, perhaps, the debate is a clash of opinions more than anything else, because neither side has provided much evidence to support its position.⁶ Fortunately, though, there is some research on the effectiveness of PowerPoint as a presentation tool. This research is summarized in the box below.

⁵See Norman (2004), Holmes (2004), and Doumont (2005). The website http://sooper.org/misc/ ppt/ provides an extensive list of links, in chronological order, to contributors to the debate over the past several years.

⁶Few (2006) criticizes Tufte for not providing any empirical evidence to support his argument.

Most of the PowerPoint effectiveness research described here comes from the field of education. Initial research in this area was optimistic, suggesting that PowerPoint had a positive effect on student understanding, recall, and grades. Specific conclusions included that, as a result of using PowerPoint in lectures:

• Their overall grades improved (Lowry, 1999; Mantei, 2000; Szabo & Hastings, 2000)

• They liked the lectures more and said that they helped them learn (Simpson, Pollacia, Speers, Willis, & Tarver, 2003).

Other studies, however, found mixed results, no effect, or negative results where students who had not been exposed to PowerPoint-based lectures scored higher on tests than those who had.

• Among students who received PowerPoint lectures, female students had improved grades, but the male students did not (Kask, 2000).

• No difference in results between lectures with and without PowerPoint (Daniels, 1999; Mines, 2001; Ranking & Hoaas, 2001).

• Kapoun (2003) taught multiple fifteen-minute instructional sessions, with and without PowerPoint, and found that students scored higher on tests afterward for the sessions without PowerPoint.

The problem with the earlier research, it turns out, was that most of the studies compared results from classes taught in years prior to the introduction of PowerPoint to those taught afterward, or classes for which the content of PowerPoint lectures was different from those of non-PowerPoint lectures.

In more recent research, when otherwise identical classes were run simultaneously, some with PowerPoint and some without, student attitudes to the class improved (as well as—not surprisingly—their ability to use PowerPoint), but no effect on their academic performance was found—either positive or negative.

So according to this research, PowerPoint is neither helpful nor harmful, at least in an educational environment.

• Susskind (2005) reports on an experiment during two sections of an Introduction to Psychology course. For the first five weeks, one section received PowerPoint lectures and the other one did not. Students were then tested, and the teaching methods were switched for the next five weeks. In each case, no differences in academic performance was found as a result of the PowerPoint lectures.

It is worth noting that in much of this research the PowerPoint presentations studied were largely text, typically lists of bullet points. As we saw above, this is PowerPoint at its worst. Bullet point slides are often less effective than no slides at all, so the mixed results from these studies should not be surprising.

The authorities are divided, and so is the research. Where does this leave us? The authorities—and the researchers—seem to be working at cross-purposes. In a certain sense, both sides are correct, because they appear to have different approaches to presentation in mind.⁷

To understand this, it helps to understand the idea of presentation idiom. A presentation idiom is a form of expression and an associated set of design principles. I call the two main types of presentation idiom Ballroom style and Conference Room style. Ballroom style presentations are what most typical PowerPoint presentations are trying to be: colorful, vibrant, attention-grabbing, and (sometimes) noisy. They typically take place in a large, dark room—such as a hotel ballroom. Conference Room style presentations are more understated: they have less color, with more details on each page; they are more likely to be on printed handouts than projected slides, and they are more suited to your average corporate conference room.

The biggest single mistake that presenters make—and the root cause of the PowerPoint debate, it seems to me—is confusing the two idioms, and particularly, using ballroom style where conference room style is more appropriate. Almost all PowerPoint presentations are given using ballroom style—yet most of the time presentation conditions call for conference room style. Ballroom style is appropriate for when the objective is to inform, impress, and/or entertain a large audience and when the information flow is largely expected to be one-way (presenter to audience). Conference room style presentations are more suited to meetings for which the objective is to engage, persuade, come to some conclusion, and

drive action. Conference room style presentations are therefore appropriate for any of the following:

Information flow in this idiom is expected to be two-way—it’s more interactive. Table IV.1 summarizes and compares the main characteristics of ballroom and conference room style presentations.

From this perspective, Tufte is criticizing—correctly—the use of ballroom style presentations in situations that need conference room style presentations. His critics are defending ballroom style presentations in situations in which they are appropriate. Therefore, in that sense, both are correct.

This advice, widely offered, is based on a faulty reading of George Miller’s 1956 paper, “The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information.” Miller himself debunks this flawed interpretation in an email to Mike Halpern. The conclusion of his paper, summarized in this letter, was that “. . . seven was a limit for the discrimination of unidimensional stimuli (pitch, loudness, brightness, etc.) and also a limit for immediate recall, neither of which has anything to do with a person’ s capacity to comprehend printed text.”⁸

Now that we understand the differences between ballroom style and conference room style presentations, we can see how slides should contain either much more, or much less, text. Ironically, therefore, seven bullets of seven words each actually gives you the worst possible slide, because it is too many for a good ballroom style slide and too few for a good conference room style slide. Ballroom style slides should have much less text than

(seven times seven equals) forty-nine words, because they are heavily visual. Conference room style slides could have a lot more text than that; you could fill up an 8-^-by-11-inch sheet of paper with well over a thousand words in nine-point font, and—if the page is properly laid out—you will still have a very effective presentation page. (Chapter 8 will explain how to design this kind of page.)

The core problem here is that spoken and printed words can conflict with each other in a presentation. There is extensive research evidence, from multiple fields, that when you speak and present text at the same time, communication is harmed because the text and your voice compete with each other in the minds of your audience. This research is summarized here.

The “redundancy principle,” that people “learn better from graphics and narration than from graphics and redundant narration and text” is well established in research (see especially Mayer, 2001, and Moreno, 2006, p. 65). Specific findings include:

• Mousavi, Low, and Sweller (1995), summarizing six experiments about teaching geometry, found that visuals combined with spoken words were more effective for student learning than either visuals combined with spoken words and text, or visuals and text without spoken words.

• Images plus narration is better than images with text and no narration, and better than images with text and narration (Mayer. 2001, p. 134).

• “If learners are required to coordinate and simultaneously process redundant material such as written and spoken text, an excessive working memory load is generated” (Kalyuga, Chandler, & Swelling, 2004, p. 567)

• Eye-tracking research found that when web viewers received pictures, audio, and text simultaneously, they tended to ignore the text (Outing & Ruel, 2004).

Researchers explain these results by suggesting that humans have multiple short- term memories for different types of information. Visual and auditory are different types of information, and so receiving both at the same time actually improves overall short- term memory. But written and spoken words use the same short-term memory, so when both are received at the same time, the audience suffers from cognitive overload.

The one exception is when you are trying to get people to remember specific facts, names, or places, in which case you should use text. Unfamiliar concepts are recalled better when presented with a multimedia graphic, while facts, names, and places are recalled better when presented with text.⁹

Not surprisingly, therefore, using graphics or a graphical layout is superior to a text slide in many ways, as the following research indicates.

In several experiments, researchers have found that using graphics in visuals increases recall, persuasion, and positive attitudes

• Use of visuals in print advertising increases positive attitudes (Rossiter & Percy, 1980).

• Data communicated through graphics (rather than tables) improved both the speed and quality of decision making in a laboratory experiment simulating a product management budget allocation exercise (Benbasat & Dexter, 1985).

• Visuals and/or animation increased persuasion over text alone (King, Dent, & Miles, 1991).

Photographs, in particular, tend to be more interesting and attractive to audiences.

• In print, the eye goes to large pictures, even before the title or headline (Garcia, 1991).

• Addition of photographs made news articles more likely to be read, and made facts from the articles more memorable (Zillman, Knobloch, & Yu, 2001).

Chapter 7, on visual presentation elements, explains what kinds of graphics to use in your presentation, which charts to use and when, whether to use color and animation, and which fonts and type sizes to use. Chapter 8 , on layout, explains how to take all those elements and put them together on each slide so that the result is impressive but not overwhelming. In particular, it will show you how you could put a lot of words on a page and still have a very effective slide.

Step 7: Identify the Most Effective Graphical Elements to Use in Your Presentation

In Step 7, you will decide which charts will provide the most convincing display of your quantitative evidence (if you are using any), and whether to use any other graphics to emphasize or clarify your qualitative evidence. In Chapter 6 you created your storyline on a series of S.Co.R.E. cards. For Step 7, you will draw a thumbnail sketch of which chart or graphic you will use on each S.Co.R.E. card that represents data that you would like to present graphically. To help you do this, we will now cover:

You will want to use a variety of different graphical types to maintain the interest of your audience: use charts, diagrams, photographs and cartoons. Threatening images can be particularly eye-catching.²

The most important thing, though, is that each graphic you use must be relevant. We know from research that when interesting but irrelevant pictures (or sounds, or music)

¹See research done on print and online magazines by Knobloch, Hastall, Zillman, & Callison (2003) and Outing & Ruel (2004). Images of victimization also draw people’s attention (Zillman, Knobloch, & Yu, 2001).

²See Mayer (2001). Also, recall is lower, and likelihood of distraction higher (Edell & Staelin, 1983).

This is known as the coherence principle: “Students learn better when extraneous material is excluded rather than included in a lesson” (Moreno, 2006, p. 65, building on Mayer, 2001).

In two related studies, students performed worse on quizzes and recall and recognition tasks when irrelevant pictures were used in presentations (Bartsch & Cobern, 2003). In an eye-tracking test, students eyes were drawn more to relevant photographs on PowerPoint slides than to irrelevant ones (Slykhuis, 2005).

⁴A study by Bergen, Grimes, & Potter (2005) showed that audiences recalled about 10 percent fewer facts from news stories communicated in a format in which graphical clutter is added to the video of the announcer than from news stories without the visual clutter.

are included in communication, effectiveness of that communication (measured in terms of ability to apply what’s been learned) is reduced.²

Your audience has limited processing capacity; irrelevant information can overload this capacity and reduce the effectiveness of your communication.³ On television, clutter reduces the amount of information conveyed; the success of MTV, which included a cluttered format, led others, particularly CNN, to emulate that clutter. But researchers now conclude that other factors—the youth of the announcers, the language used, and the music itself—contributed to the success, and that the visual clutter actually harms communication.

Should you use clip art? No. Never. Clip art is an example of irrelevant graphics, and should be avoided, without exception.

To help you select a good chart, use the chart selector guide in Figure 7.1. This diagram helps you think about which kinds of chart to consider, depending on what you want your data to demonstrate. Your choice of chart will depend first on what task you want the chart to accomplish on the slide. Each chart could accomplish any of four tasks on a slide: it can show a relationship (e.g., when advertising goes up, sales go up too); it can make a comparison (e.g., retention rates are higher among women than among men, or this year’s sales are higher than last year’s); it can display the distribution of your data (e.g., there is a broad range of prices that people are willing to pay for car warranties); and it can show the composition of your data—what the component parts are (e.g., final cost of the product is made up of manufacturing cost, transportation cost, and insurance).

Next, you must consider various characteristics of your data set, such as the number of observations, the number of variables per observation, and whether you are looking at static (point- in- time) data or data over time (time series). Which characteristics will drive your choice of chart depends on your initial choice of what you are intending to demonstrate.

The way to use this diagram is as follows. Start in the middle, with the question “What would you like to show?” Then decide which of the four choices you want your data to demonstrate: relationship, comparison, distribution, or composition. If you want to show more than one of these, you will probably end up drawing more than one chart.

When choosing the right chart for displaying a relationship, the important characteristic is the number of variables in your data; for bivariate data, a scatter plot is appropriate, while for trivariate data, a bubble chart works well. The relationship between advertising and unit sales involves two variables (advertising and unit sales), and therefore a scatter plot would be a good choice here.

If you wanted to show the relationship of advertising, price, and unit sales (three variables), then the suggestion is a bubble chart, where the x-axis, the y-axis, and the size of the bubbles represent the three variables.

If you want to show the distribution of your data, move to the top of Figure 7.1. If you have univariate data, move upward and left. If your observations are graphed into few intervals, the suggested chart is a column histogram; if many intervals, a line histogram. For example, if you are showing age breakdown of a particular sample of people in brackets of twenty years (0 to 19, 20 to 39, 40 to 59, etc., that is, few intervals), then a column histogram would be appropriate. If you are showing instead age breakdown by year (how many people are age 0, 1, 2, 3, etc., that is, many intervals) then a line histogram is suggested. To the right of that, if you have bivariate data, a scatter chart is proposed, and if trivariate data, a 3D area chart.

To show a comparison, move to the left of the figure. The first question is whether you want to show how your data is changing over time, in which case follow the chart to the upper left, or if it is static data, just a single period, which is at the lower left. For static data, use a bar chart for univariate data, with each bar representing one observation. Sales for five different companies would be shown on five different bars. Bar charts (bars lie horizontally) are for comparisons among things, while column charts (columns rise vertically) are for comparisons over time.

For static bivariate data, use a variable width column chart. The height of each column corresponds to one variable, and the width, the other. So if you had two variables representing sales—dollar sales and sales growth—each column would be drawn using those two variables. (Although drawing this chart in column form would seem to contradict the conclusion that columns are only for showing change over time, I have never seen it drawn in bar form.) For static multivariate data with few subjects, use multiple bar charts, lined up across the page, where each chart represents one subject and each of the bars correspond to the different variables. For example, if you wanted to compare GDP per capita, GPD growth, population size, population growth, and population density (variables) for four different countries (subjects), then each of the bar charts would represent one country, while each of the bars would represent one of the variables for that country. This way you can compare variables across subjects by reading across the page, and variables within each subject by looking at each bar chart as a whole.

Finally, where you have many items and many variables, use a trellis chart, which is a table of charts. You can either use the table as a space saver—allowing you to fit, say, twenty bar charts on a page in a four-by-five matrix—or you can use it to highlight one or two of the more important variables by using those variables to determine which cell in the table you place each chart on. (Of course, there are limits to this approach, because you cannot place more then one chart in each cell, and the intervals on each axis have to be roughly equal.)

For time-series data, if you have only a few periods of data, then you could use a column chart if you also have only a few data items (with each data item represented by a different set of columns, in different colors or shades of gray), or else a line chart if you have many data items, with several lines, one for each data item.

If you have many periods of data, then also use a line chart if you have non-cyclical data or a circular area chart (sometimes called a spider chart) for cyclical data.

The last option, on the right side of Figure 7.1, is composition: this is when you want to highlight the components of your data. Your first choice here is again whether your data is static or changing over time. Time-series data is also a comparison of sorts—except that you are comparing the same items over time, rather than different items—so the options are similar to the choices for comparisons. If you have few time periods to display, then choose a stacked column chart to show both the relative and absolute differences between periods, and if only the relative differences matter, than use a stacked 100 percent column chart, where the height of each column is fixed and the components are shown as percentages rather than absolute numbers.

If you have many periods to show, then you will again use lines rather than columns. To show relative and absolute differences over time, use a stacked area chart, and to show relative differences only, use a stacked 100 percent area chart. Options for displaying static data are at the upper right. For showing a simple share of total, the pie chart is very popular, although some people have found it to be less effective than other options. If you are trying to show how components of your data add up to and subtract from a total, use a waterfall chart, and if you want to show how some of your subcomponents also have subcomponents, then a stacked 100 percent column chart with subcomponents is ideal.

The chart selector diagram is only a guide. The best way to decide which chart to use, once you have decided what you want to show with your data, is to use the diagram for suggestions and then try a few alternatives—have your colleagues take a look at the options—to see which one works best.⁵

Fortunately, one factor that does not seem to be important in choosing a good chart is audience personality type, so while audience personality type is critical for several other aspects of your presentation design (see Chapter 1), it does not appear to be relevant to choosing charts.⁶

Perhaps the most thoroughly researched aspect of presentation design is the use of charts. There are several relevant findings:

• Line graphs are better than bar graphs for trends (Shah, Mayer, & Hegarty, 1999).

• Based on an extensive review of empirical work on graphs, Jarvenpaa and Dickson (1988) recommend using a horizontal (bar) chart rather than a vertical (column) chart when making comparisons among variables.

• Lewandowsky and Spence (1989) found that in scatterplots, discriminating different data series with color provided the fastest comprehension.

• Cleveland (1984) found that people make less accurate judgments about the quantities in a pie chart than in a bar or column chart.

• For multiple comparisons, divided bars are better than pies, so long as the height is kept the same (Hollands & Spence, 2001).

• Horizontal (bar) charts are more effective for showing parts of a whole than either pie charts or cumulative (stacked) column charts (Jarvenpaa & Dickson, 1988), but Spence and Lewandowsky (1991) found that pies are more effective when the reader has to mentally add two or more quantities together (likely because if the quantities are adjacent in a pie, they can be read as a single quantity—but this is also true of stacked bars).

For additional research on the use of charts, see the Graphics section of Appendix D.

As you consider which charts to use, make sure that you include lots of relevant detail: detail improves the persuasiveness of your presentation. (Chapter 8 will discuss the importance of adding lots of detail—properly organized-to each of your slides, particularly for conference room style presentations.)

The research here is inconclusive. While several studies (such as Wheildon, 2005) claimed to show that serif fonts (such as Times New Roman) are more legible than sans serif fonts (for example, Arial), many other studies have found no difference, and more recent analysis has found methodological problems with the studies that conclude that one type of font is better than the other (Lund, 1999). Therefore, do not spend any time worrying about fonts; just pick a readable one and stay with it.

People seem to remember information presented in bullets better than information in paragraphs⁷. But don’t rely on this, because—as we saw in the introduction to the Rhetoric section of this book—stories are more memorable than bullet points. Use bullet points to deliver information that cannot fit into a story form, such as lists of items.

It is very important, though, that the bullet list not be the central feature of your slide, otherwise you will run into the problem of verbal redundancy described in the introduction to this part of the book (see “Should each slide have seven bullets with seven words per bullet?”). The bullet list is just one element in your overall slide—and therefore is usually only appropriate in a conference room style slide, because in ballroom style it would take up the whole page. (Chapter 8 will explain how to incorporate bullet lists, and text in general, in conference room style slides in a way that avoids verbal redundancy.)

The results of the research on the use of color in communication is mixed on whether color adds any benefit. This suggests that there are more factors involved in deciding the answer than have been captured in any particular research study. One study indicates that the different results depend on how involved the audience is. For an uninvolved audience, color can be useful for attracting attention and persuading, but for a more involved audience, irrelevant color is harmful because it distracts your audience and wastes their mental effort processing material that is extraneous to your message. The most effective use of color therefore seems to be for highlighting specific items on your slide.

• Vogel (1986) found that color is more persuasive than black and white; this study was done using projection of acetate slides.

• Butler and Mautz (1996) found that use of animated color graphics improved recall among those who are visually inclined, versus use of text-based visuals only. However, it is unclear from this study whether the improved recall is attributable to the color, the graphics, or the animation.

• Kelly and Hoel (1991), studying Yellow Pages advertising, found that in three out of four test ads, the addition of color had no effect, while in one case it did appear to increase the likelihood of the consumer selecting the advertised business over others.

• Color can attract attention and help convey specific information (Finn, 1988; Fernandez & Rosen, 2000).

• In print advertising and direct mail, color gains attention (Rossiter & Percy, 1997) and response (Woodside, Beretich, & Lauricella, 1993)

• People tend to select color ads from Yellow Pages directories over black and white (Lohse, 1997)

• A review of the research on the role of color in speeding up identification of information, from the 1950s to the 1970s, found that irrelevant color harms both speed and accuracy of information access (Christ, 1975).

Color helps speed up both information extraction (from certain kinds of charts) and decision making.

• A laboratory experiment showed that color helped decision making when under time pressure for audiences who are “field dependent,” that is, more holistic, “big picture” thinkers, who tend to have less attention to detail (Benbasat & Dexter,

• A laboratory experiment showed that color sped up information extraction from bar and pie charts (Hoadley, 1990), but the alternative tested was a black-and-white chart with awful cross-hatch shading—arguably anything would be better than that.

• One study found that using color symbols to discriminate among different data series in a scatter plot provided faster comprehension than any other type of symbol differences, such as shading or letters (Lewandowsky & Spence, 1989).

• Christ's (1975) meta-analysis concluded that color used for reinforcement or highlighting improved both access speed and accuracy, so long as the viewer knew what color to look for.

Perhaps the most useful guidance on the role of color is provided by a study of print advertising, which concluded that, when your audience is not very motivated, color helps persuade, likely because color leads them to make a superficial judgment that you have a higher quality presentation. When the audience is motivated, however, full color is harmful to persuasion, because the audience has to spend brain effort processing the meaning of the color, where black and white would be easier to comprehend. If the color is used only to highlight and reinforce relevant aspects of the presentation, though, (and the audience is motivated), then persuasion is improved.

• Myers-Levy and Peracchio (1995) conducted two experiments with students who viewed print advertising for bicycles. They measured the impact of color on the students' favorability ratings of the bikes in each ad; presumably, more persuasive advertising would lead the viewer to rate the bike more favorably.

Finally, there are two practical considerations, both of which suggest avoiding color. One is that a non-trivial proportion of the population is colorblind, and the other is that if people make copies of your presentation, chances are that they will be in black and white.

The question of whether to use color, therefore, along with the question of how much detail to include, which is covered in the next chapter, depends on whether you are making a ballroom style or conference room style presentation. Ballroom style presentations, which are used to inform or entertain, would seem to have less motivated audiences, and therefore would benefit from the use of color. In any case, the use of color photographs is appropriate, because all the colors in photograph are “relevant.” You should avoid using color to embellish your slides, though, because it will be distracting. Conference room style presentations, which are used to engage and persuade, have more motivated audiences (and use printed presentations, which might be photocopied), and therefore should use color just for reinforcement or not at all.

Transitions and animation are two forms of moving content. Transitions are when one slide changes to another, while animation is any moving content within an individual slide. Typical examples of transitions include dissolving, shrinking, or page turning effects. Transitions alone are generally a very bad idea. Research suggests that using transitions alone, without animations within the slides, is worse than using no visuals at all, in terms of getting attention and agreement. This is probably because, as so much of the research has shown, irrelevant material tends to harm communication, and transitions are largely irrelevant to your message. Transitions and animation together might be better than visuals alone, but the research is not conclusive here.³

The research on animation is not very encouraging, either, which suggests that both transitions and animation should be avoided. The only time when animation is warranted appears to be when the animation effect communicates something more than could be shown through static images only. (And this is good news, because developing an animated sequence for a presentation takes a lot of work.)

Research on the benefits of animation in communication, similar to that of color, is very mixed. Where benefits are found, they tend to be only for particular types of people (e.g., those who are visually inclined, or those who are novices in the subject matter being presented).

• No difference was found between static visuals and animation in several studies (Hegarty, Narayanan, & Freitas, 2002; King, Dent, & miles, 1991; Mayer, Mathias, & Wetzell, 2002; Narayanan & Hegarty, 2002; Ricer, Filak, & Short, 2005).

• Animation, when combined with graphics and color, did improve recall, versus text-based visuals, but only among those who are visually inclined. Recall among audience members who were not visually inclined was actually reduced with the animated version (Butler & Mautz, 1996).

• Lai (2000a; 2000b) found that animation with audio is superior to static graphics and text with narration in computer-aided instruction.

• One recent study found that animated charts were superior to printed statistics; however, this could have proven only that charts are superior to numbers, which we already know to be the case. Further, the finding held only for “novice” audience members; expert audience members were not swayed by the technology (Guadagno, Sundie, Asher, & Cialdini, 2006).

The only time that research finds animation to be consistently superior to static display is when animation provides additional information about changes in the process or system being described.

• In a review of recent empirical research on the use of animation, Tversky, Morrison, & Betrancourt (2002, p. 21) concluded that where animation is found to be superior to static visuals, it is usually because the animation contains some additional information, so it is not a fair comparison. They suggest that animation might be most useful for showing the “qualitative aspects of motion or . . . the exact sequence and timing of complex operations,” but they are unable to say whether even in these cases animation would be superior to static graphics.

At this point you know what elements you are going to use on each slide or page. But how exactly do you lay them out all together to communicate them most effectively? That is what Chapter 8 is all about.

Step 8: Create Slides That Communicate Your Information Concisely and Effectively

The previous chapter was about how to select and create all the elements of a good presentation page or slide: the graphics, charts, color, animation, and fonts. In this chapter, we will look at how to put all those elements together—how to lay them out on each page.

Before you begin laying out your slides, you need to decide definitively whether you are going to use ballroom style or conference room style. Steps 1 to 7 of the Extreme Presentation method work equally well for both idioms. Once you start step 8, however, you have to decide on—and commit to—which idiom to use for your presentation. (If you do not know the difference between conference room style and ballroom style, then go back and read the introduction to Part IV: Graphics).

The main determinant of which style to use is whether you are trying to persuade a small audience, in which case you should use conference room style, or whether you are trying to inform or entertain a larger audience, which would instead call for ballroom style. If you recall the discussion in Chapter 3, we argued that “purely informational” presentations have a very hard time attracting and keeping audience attention (which is why they need the vibrant imagery and the emotional appeal of ballroom style). The recommendation in that chapter was that you should always try to identify a problem that your

audience has, and then focus on persuading them to take action to solve it. We will assume that you have followed this advice, and therefore this chapter will focus mostly on conference room style.

A critical question in this chapter is, “How do you lay out each slide so that it grabs the audience ’s interest and persuasively conveys the intended message?” The answer to this question is to design each page so that the page layout itself reinforces the main message of the page, For example, if the slide is talking about two alternatives, then draw the alternatives on either side of the page (Figure 8.1a). If your slide is about how you had several ideas, which you put thorough a number of screens, then draw it that way (Figure 8.1b; Figure 1.3 in Chapter 1 is an example of this layout). If you’re explaining how several factors are converging to create a new situation, then draw the factors around the side of the page and the new situation in the middle (Figure 8.1c).

FIGURE 8.1B. Layout Showing Screening of Ideas FIGURE 8.1C. Layout Showing Factors Converging

One of the advantages of the greater detail that conference style presentations allow is that you can combine what would otherwise be several slides into one, with the added benefit that the relationships within the content of those formerly different slides is emphasized visually. Figure 8.1c is an example of this: each of the four factors and the conclusion in the center might have been drawn as five different slides; instead, we draw them as one slide to visually reinforce the notion of the four factors coming together to create the conclusion. (Figure 8.1d is a fully drawn variation on Figure 8.1c.) Always prefer diagrams and images to text, because, as we saw in Chapter 7, pictures reinforce a spoken presentation, while text competes with it.

What you should do now is work through your S.Co.R.E. cards and see how many you think you can fit on each slide by grouping your cards. As you take each card, in order, think about how much space each will take on a slide, and then keep on adding cards until you think you will have filled your first slide. Put those cards in a pile, and then take the next card and start a new pile. Keep doing this until you have used all your cards. At this point, you should have a series of piles of cards, each pile representing a slide.

Then think about what the central message of each slide should be and make a rough sketch of the layout of each slide so that the layout of the page supports the main message of the page. Once you have a rough sketch of the layout of each page, place the sketches side-by-side across your desk, and you will have a storyboard for your presentation. Scan this storyboard to make sure that there is enough variety in page layout, from slide to slide. If there isn’t, this may be an indication that your presentation is repetitive; consider combining some of the pages or changing their design. Note that you could create this storyboard in PowerPoint, but if so, be sure that each slide is only a very rough layout for now. You do not want to be spending much time on any given slide until you are sure that you have the right layout.

Once you are happy with your storyboard, you can then create your actual slides or pages, in PowerPoint or whatever your preferred layout program is. While this is still one of the most time-consuming steps in the presentation-design process, having gone through all the preceding steps you should find that you do spend a lot less time actually drawing your pages than you did before. This is because you are clear on what you want to show and how you will show it, so you will spend less time redrawing pages or drawing pages that you do not end up using. Also, you will have far fewer slides to draw because the method you have followed creates much shorter presentations.

* How to make sure that your slide layout reinforces the main message of the slide

How to Make Sure That Your Slide Layout Reinforces the Main Message of the Slide

The most important message of this chapter has already been introduced, namely, that if you want to design slides that really grab your audience ’s attention, you need to make sure that the layout of each page itself reinforces the point you’ re trying to make on the page. But how exactly do you do this? The answer is to use what designers call the “ squint test. ”

Hotes off	Lotts off	Lotts off	Lotts off	Lotssoff
sttufff	stufff	stufff	stUff	stUff
happening	happening	happening	happening	happening
ini each off	ini each off	in<eaChoff	in<eaChoff	in<eaChoff
these	tthheessee	tthheessee	tthheessee	tthheessee
phases	phases	pphhaasseess	pphhaasseess	pphhaasseess

To apply the squint test, squint at the slide, so that all the text is blurred and illegible. Do you understand anything about the slide, without having to read the text? If you can see, already, what the main message of the page is—that the page is showing a process, for example—then your page passes the squint test (see Figure 8.2). The squint test seems to simulate what the brain registers in the first fraction of a second when the image from your slide hits the viewer’ s retina. If this introduces the main point of your page, the rest of your message from that page can be received more easily than if the brain has to spend the next few seconds trying to decipher what your point is.

Does your typical bullet point page pass the squint test? Not really. At best, this kind of page layout suggests a list. If the message of the page is a list, then this page layout may be appropriate. But if so, help the viewer understand what kind of a list it is. If it’s a checklist, use checkboxes instead of bullets. If it’ s a series of steps, wouldn’ t a more effective layout be a process diagram, across the page? At worst, when the brain registers a series of bullets—perhaps for the fiftieth time that hour—it wanders off to some other more pleasing image, such as next summer’s vacation.

Here are some examples of slides that fail the squint test being transformed into slides that pass the squint test. In the first example, Figure 8.3 shows a series of four slides from a real presentation, disguised. The slides describe the objective and a sequence of steps in a procedure. If you squint at them, though, all you see is a bunch of bullet points and therefore these slides fail the squint test. Consider now the slide in Figure 8.4. This is the

- Determine if facility has developed & operationalized polices and procedures that xxxxxx

4. Request evidence of how the facility has handled alleged violations. Select 2-3 alleged violations since last survey

Objective: Determine whether facility has developed and operationalized procedures that xxxx

same content, on one slide. If you squint at this slide, you can tell, even without reading any of the content, that it represents a process, just by the layout of the page. So it is an ¹S_everal d_esign_{s o}f layouts _tha_t improvement, because it passes the squint test. And as an added benefit, page count is

pass the squint test are available reduced by 75 percent, from four pages down to one.¹

A study of Air Force Academy students taking a mandatory class on engineering mechanics found that the visuals used in the class were considered to be a “confusing, intimidating waste of time” (Bowe, Jensen, Feland, & Self, 2000, p. 12).

f you want to show that your data provides evidence of a relationship, for example, between advertising and sales revenues, then you would move to the bottom of Figure 7.1.

See Vogel & Morrison (1998) on the effects of transitions. On using transitions and animations together, the study identified a directional, but not statistically significant, improvement in perceptions of the speaker, in terms of preparedness, professionalism, clarity, strength, and quality of support data.

Purpose	Inform, impress, or entertain a large audience	Engage, persuade, drive action in a smaller audience
Look	Colorful, vibrant, attention-grabbing, noisy	Black and white, lots of detail
Typical information flow	One-way (presenter to audience)	Two-way (interactive)
Delivery	Projected	Printed handout
Typical location	Hotel ballroom	Office or conference room


Tie to client problem	89%
Tie to client problem
Variety of charts	67%		22%
■
Broad range of	56%		33%
Simplicity of
design/complexity	33%	33%
of data	Regularly	Occasionally

	Procedures
	1. Obtain and Review policies and procedures for key components
	2. Interview individuals responsible for coordinating policies and procedures
	3. Probes
	How staff is monitored to assure xxx?
	How to determine xxx?
	How do you ensure xxx?

	Procedures (Cont.)
	7. Interview at least three frontline supervisors or staff to determine how they monitor xxx
	- Xxxxxxx
	- xxxxxxx
	8. Obtain a list of all employees hired within the past four months, select five and ask the facility for written evidence of pre-employment screening.
	- Xxxxxxx

1. Obtain	\ 2. Interview	3. Probe	\4. Request	5. Under-	\6. Deter-	\ 7. Evaluate	\8. Confirm pre-
and review	\ policy and \ procedures\ evidence of \ stand				mine staff	\ monitoring \ employment
policies	procedure		violation	customer	knowledge-		screening
and	coordinators		handling handling	awareness awareness	ability ability		/
procedures							/
Obtain and	Interview in-	Ask how staff	Request	Interview	Interview	Interview at	Obtain a
Review	dividuals re-	is monitored	evidence	several	five staff over	least three	list of all
Policies and	sponsible for	to assure xxx	of how the	customers	three shifts	frontline	employees
Procedures	coordinating		facility has	about their	to determine	supervisors	hired within
for key com-	Policies and		handled	awareness	if they are	or staff to de-	the past four
ponents	Procedures		alleged	of reporting	knowledge-	termine how	months,
			violations	procedures	able about how to xxx, and what, when, & to whom to report according to policy	they monitor xxxx	select five and ask the facility for written evidence of preemployment screening.