Designing graphics for Sales Presentations?

[Mary]

Hi,

I'm working on designing some graphics for sales presentations. While some of these graphics may become part of the dashboard for the final product, which is a payment processing system, I'm also wondering if there is a difference between designing graphics for sales versus designing for presentation.

For instance, I've read some Tufte and the appearance he recommends for readability is fairly minimalistic; white backgrounds, maybe one color predominating throughout, with emphasis on making the graph readable and understandable. But is this the same approach graphs for sales should have? Advertising on TV, for instance, has a lot of color and movement. Should graphs for sales have more color and movement than graphs destined for peer-reviewed journals?

I'd be interested in this group's thoughts, since presumably the reason you are a member is that you do a lot of graphs.

Also any articles or books on the subject you can point to would be of interest to me. It also accurs to me that a lot of Tufte's work was a number of years ago; it is still as relavant today? Yes, I understand the difference between exploratory and expository graphs, but are there not different approaches to presentation-quality graphs for different purposes and venues?


-Mary

[Ben Bolker]

Great question.

There's a lot of stuff on Andrew Gelman's blog about "infoviz vs.
statistical graphics", where "statistical graphics" represents the
classical Tufte/Cleveland approach (clean design, maximize data-ink,
convey as much quantitative information as accurately as possible) and
"infoviz" represents the more recent thread of graphical display of
large data sets, often in presentation contexts (flashy, engaging,
dynamic ...) Gelman comes from the statistical graphics school, but
tries to be evenhanded in exploring the different goals and techniques
of the two approaches.

Stephen Few also writes a lot -- from a business intelligence rather
than a sales perspective (i.e. it sounds like the target of the graphics
would be people *within* a corporation who wanted to know what was going
on, rather than sales clients), but nevertheless from the business world
(so much closer/more sympathetic to the goals of a sales person than the
average academic would be)

Without being too long-winded, I will mention a few more ideas:

* I'm assuming that the goal of the sales pitch is to be *honest* while
selling as strongly as possible -- that is, you're not actively trying
to hide data or convey a false message, but you do want to put your
product in the best light possible
* there's always a cultural undercurrent ("but what if my
boss/clients/supervisor needs to see three-dimensional pie
charts/dynamite plots/etc.?")
* animation and dynamic graphics probably *should* or *could* be used
more by academics, but aren't because we're (1) stodgy (2) cheap and (3)
aiming for dissemination in a fairly traditional print or
print-emulating web format.

[Mary]

Gelman seems to lean towards more and brighter color, but perhaps less text, as in the second set of maps on this blog:

http://andrewgelman.com/2011/08/infovis-infographics-and-data-visualization-where-im-coming-from-and-where-id-like-to-go/

Thanks; would love to hear more ideas on this subject.

-Mary

--- bbo...@gmail.com wrote:

--
--
You received this message because you are subscribed to the ggplot2 mailing list.
Please provide a reproducible example: https://github.com/hadley/devtools/wiki/Reproducibility

To post: email ggp...@googlegroups.com
To unsubscribe: email ggplot2+u...@googlegroups.com
More options: http://groups.google.com/group/ggplot2

---
You received this message because you are subscribed to the Google Groups "ggplot2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ggplot2+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Mary]

Dennis,

Yes, they are less knowledgeble about statistics, but very knowledgeable about their fields (though we are very knowledgeable of the field as well). I do see from my internal manager looking at the results that he wants to take plots like boxplots and click on the outliers to get another graph; can something like that be done in ggplot? So some level of interactivity, if not movement, seems to be desirable. I've got some movement so far but not too much interactivity.

Yes, we are putting in some things that animate, but not going overboard on it, and trying to make them very understandable. The idea is still to make the graphics still make the results of the program more understandable in a graphic form, though we are trying to determine if we need to jazz them up more for the sales than for the deliverable final product, as the graphics will mostly be in that as well.


-Mary

--- djm...@gmail.com wrote:

From: Dennis Murphy <djm...@gmail.com>
To: mlho...@avalon.net
Subject: Re: Designing graphics for Sales Presentations?

Date: Mon, 11 Feb 2013 14:25:03 -0800

Hi Mary:

I find this question interesting:

> But is this the same approach graphs
> for sales should have? Advertising on TV, for instance, has a lot
> of color and movement. Should graphs for sales have more color and
> movement than graphs destined for peer-reviewed journals?

TV ads are targeted at [potential] customers in an environment of
visual clutter where you need to stand out somehow in order to be
noticed. In a (corporate) presentation, the audience is of limited
size and captive, presumably with a certain common base of knowledge
and interest, so I don't particularly see why it would be necessary to
jazz up presentation graphics with animation and color unless (i) the
audience has a short attention span; (ii) the audience is easily bored
and distracted; or (iii) animation and color are necessary to deliver
the intended message. [ (i) + (ii) = a class of second graders ]

I prefer the KISS principle. Most people, including corporate
managers, have a difficult time grasping basic statistical concepts,
which is why I would try to keep the visuals simple - or at least to
keep the message of the graphic simple. The more you add to the graph,
the more opportunities you create to confuse/misdirect the consumer of
the graph. Having said that, sometimes animation provides the added
dimension that transforms an average graphic into something visceral
and powerful to an audience that is capable of grasping the message.
So the bottom line, I guess, is to match the graphical sophistication
of the presentation to the sophistication of the audience to whom it
is being presented...which I'm sure is pretty obvious to you.

My 2c,
Dennis

On Mon, Feb 11, 2013 at 12:41 PM, Mary <mlho...@avalon.net> wrote:
>
> Hi,
>
> I'm working on designing some graphics for sales presentations. While some of these graphics may become part of the dashboard for the final product, which is a payment processing system, I'm also wondering if there is a difference between designing graphics for sales versus designing for presentation.
>
> For instance, I've read some Tufte and the appearance he recommends for readability is fairly minimalistic; white backgrounds, maybe one color predominating throughout, with emphasis on making the graph readable and understandable. But is this the same approach graphs for sales should have? Advertising on TV, for instance, has a lot of color and movement. Should graphs for sales have more color and movement than graphs destined for peer-reviewed journals?
>
> I'd be interested in this group's thoughts, since presumably the reason you are a member is that you do a lot of graphs.
>
> Also any articles or books on the subject you can point to would be of interest to me. It also accurs to me that a lot of Tufte's work was a number of years ago; it is still as relavant today? Yes, I understand the difference between exploratory and expository graphs, but are there not different approaches to presentation-quality graphs for different purposes and venues?
>
>
> -Mary
>

[Mary]

Here's an interesting article;

http://www.perceptualedge.com/files/are_infovis_and_statistical_graphics_really_all_that_different.pdf

"By “beautiful,” I’m referring to an extraordinary quality that makes something
stunning in appearance. There are occasions when something must be stunning
to catch an audience’s attention, such as when people are browsing the Web for
something interesting, but only in such circumstances are the effort and graphic
design skills that are required to accomplish this necessary. The overall aesthetic
qualities that statistical graphics should routinely exhibit may be summarized as
“pleasing to the eye.” The display doesn’t look cluttered. The colors work well
together and are not overly bright, except in cases when brightness causes
something important to stand out. The text is legible. Non-data components of
the display (e.g., axes on a graph) do their part subtly, without undue salience.
The most important information stands out above the rest."

I'm not sure I'm at the skill level to make graphs beautiful, but perhaps I can aim to make them very pleasing according to the definition above.

-Mary

--- mlho...@avalon.net wrote:

[Ista Zahn]

Hi Mary,

I'm more on the academic side of things, but I would imagine that most
of the principles of good statistical graphs would also apply to those
for corporate sales graphs, e.g., don't clutter, no meaningless 3d,
etc. But I think one key difference is that in academic graphs the
overall figure is not designed to attract attention. This is similar
to the idea in typography that the typography should be invisible,
i.e., the read should not notice it. Because if the reader is paying
attention to the typography then they are distracted from the
important thing which is the words themselves. (see
http://en.wikipedia.org/wiki/The_Crystal_Goblet for a more articulate
explanation of this idea.) In corporate graphs however I imagine that
the theme and formatting should look good as opposed to be invisible.

So for example:

p <- ggplot(economics, aes(x=date, y=unemploy))

## basic academic type graph
p + geom_line() + theme_bw()

## maybe better for getting attention:
p + geom_area(fill=(alpha("blue", .5)),
color="blue") +
theme(text=element_text(color="white"),
rect = element_rect(fill="black"),
panel.background = element_rect(fill="black"),
panel.grid.major.x = element_blank(),
panel.grid.minor.x = element_blank(),
panel.grid.minor.y = element_blank(),
panel.grid.major.y = element_line(color=alpha("white", .50)),
axis.line = element_line(color="white"))


Best,
Ista

[M. Edward (Ed) Borasky]

My advice is simple - forget statistics, Tufte, infographics vs.
analysis, etc. You are asking someone to spend money for a product or
service, probably in competition with at least two other vendors.
Therefore:

1. Tell me how it solves my problem and saves me money, saves me time
or makes me money.
2. Tell me why *your* solution is better than competitors' for *my* problem.
3. Tell me how easy it is to get started, ask for the business and shut up! ;-)

Twitter: http://twitter.com/znmeb; Computational Journalism on a Stick
http://j.mp/CompJournoStick/

The National Coal Institute reminds you, "There's no fuel like an old fuel."