Category: Uncategorized

Posted on

Adaptive Path’s Leah Buley interviews Dave Gray using index cards

“Adaptive Path has a brain crush on Dave Gray. He’s the founder of XPLANE, a compay that creates visualizations to clarify complex business problems. More importantly, Dave is that rare combination: super smart and super down to earth. A couple years ago, Dave did a presentation called Free the Facts, in which every slide is a scan of a handwritten index card. The cards start out simply, and get increasingly more detailed and fascinating as the presentation progresses. See? Super smart and super down to earth!”

Read the interview here.

Posted on

Chris Brogan interviews Dave Gray

“Dave Gray is productive. This man puts out more graphics in more media than a graffiti tagger with a hardware store full of paint. Besides inventing his own visual language (you do that in your spare time, right?), Dave has written a book, taught class in the physical world, started a visual thinking school, and landed a spot in the Top 10 popular lenses at Seth Godin’s Squidoo project. Dave was kind enough to give me an interview recently, and there’s lots to cover.”

Read the interview here.

Posted on

Toward a theory of information relativity

Visual thinking, originally uploaded by dgray_xplane.

People often ask me how to visualize information. They ask things like “How can I visualize my industry ecosystem?” or “How can I visualize how my product works?” My first instinct is to try and back them up a bit. This is because they are already defining their project in terms of an answer or solution, and before you can deliver an answer you need to know the question. Getting the question right is the most important component in information design, and it’s the most common point where information design goes wrong.

This is because information is always relative. Always. Before you can undertake any kind of visualization exercise, you need to know what question you want to answer, and for whom. A look at the history of information will confirm this point.

Science is a process by which we attempt to compare our perceptions with something we call “reality” but in fact reality is something we can never really know for sure. Like the flickering shadows on the wall of Plato’s cave, reality is something we can only see dimly, because it’s distorted by our perceptions and beliefs. Every observation and fact has a margin of error, which is directly related to the observer’s background, beliefs, culture and limitations. It is written in the Talmud, “We see things not as they are, but as we are.”1

By beginning with an audience and a question you give yourself a focusing device. Like a flashlight, the audience and question will illuminate the information that’s relevant to your goal, while leaving the rest in the dark. Good information design is as much about what you leave out as what you put in.

What to put in: Information that’s relevant to your target audience, and that answers a real question that they have. What to leave out: Everything else. The best rule of thumb is “When in doubt, leave it out.”

So if you want to create a visual in order to explain something, ask yourself the following questions first:

“Who am I explaining it to?”
“What do I want them to do?”

At XPLANE we call this the WHODO, and it’s a required input to any project we undertake. Once you understand the WHO (your audience) you will have a sense of their level of existing knowledge of the subject.

For example if you are explaining scientific or technical information to engineers or scientists you can assume a high level of sophistication and readiness. Based on the cultural expectations in science and engineering fields, you can also assume a high level of skepticism and a need for evidence and proof.

Explaining the same information to a group of executives, or salespeople, would be a completely different exercise. You can expect that they will have a different set of questions and probably will be more focused on practical applications and will get impatient with scientific or technical explanations.

Thus, the same information will need to be presented very differently based on the audience that you are talking to.

Understanding the audience is only part of the equation. The other half of the WHODO is DO. Before you can undertake any explanatory task you need to know what outcome you expect. Describing this as a change in understanding is not enough. Understanding is difficult to observe. People often will say they understand something just to get you out of their hair.

People will also believe that they understand something when they don’t really understand it. Have you ever left a meeting where everyone seemed to be in agreement, yet their later actions made it clear that they didn’t agree after all? It’s common to see nodding heads in a room when people don’t agree – they think they agree, but in reality they don’t. This is because when an explanation is sufficiently vague, people are free to believe what they want to believe. Politicians often use this rhetorical principle to great effect. Words like “freedom, justice and fairness” mean different things to different people. Vague explanations are common in business, and they can give the illusion of agreement. But they don’t get results.

Here’s the key: When people understand things differently, they do things differently. What they say is less revelatory than what they do. So if you want to build a rock-solid explanation, focus on what you want people to do. If they understand what you are saying, what changes in behavior would you expect to see?

Once you have defined your WHODO, next you need to anticipate the questions they will ask. This again will depend on your audience and the information they will need to make a decision. Part of this is also cultural. Scientists will want to see scientific evidence. They will want technical explanations and probably a lot of detailed analysis. Busy executives may want different kinds of proof, such as what customers are buying and what competitors are doing. They will also be less generous with their time and expect you to get quickly to the point.

But you don’t have to go in cold. If you understand the mind of your audience, you will be able to generate a list of questions that they are likely to ask.

Once you have defined your WHODO and generated a list of questions, you can start thinking about how to visualize the information. Will they need a broad overview or detailed charts and specifications? Will they need to see the value to the customer, or the technical operational details?

In the fields of information science and knowledge management there is a model known as the Data Information Knowledge Wisdom Hierarchy, or DIKW for short. This has become a standard for defining the terms and how they relate to each other.2 Here’s how it works:

Data has no particular significance beyond representation. It consists of symbols that stand for objects, events or their properties. Data is a collection of facts3 – also called “know-nothing”4 to reinforce the point that data, by itself is dumb; it has no meaning.

Information is data that has been organized so that it is useful, usually because of relational connections – also called “know-what.” Information answers questions like who, what, where, when and how many.

Knowledge is information that has been integrated into the mind, memory and body, such that it can be applied to doing and making things, also called “know-how.” Knowledge is usually acquired through experience, or through stories about other people’s experiences.

Wisdom is the ability to perceive value, make judgments, and evaluate long-term consequences. Russell Ackoff describes the difference between knowledge and wisdom as the difference between doing things right and doing the right things. Wisdom requires values, and values are perhaps the most relative thing of all.

So I propose the beginnings of a theory of information relativity:

1. All information is relative, and it’s always relative: relative to the observer and the observer’s point of view; relative to the culture and its values; relative to the situation; relative to what has come before, and to what will come next.

2. The value of information is always relative because it is directly related to its usefulness, which depends on the user, the context and the situation.

3. Information design must therefore be driven by the context within which it will be experienced. Information design must serve the needs of real human beings doing real things. Information wants to be used.

At its heart, information design is about change. It’s about increasing the amount of useful information in the world. Good information design should result in changes to understanding – increases in knowledge and wisdom – which can be directly measured by observable changes in human behavior.

Notes:

1. Also attributed to John Milton, H.M. Tomlinson, Anais Nin and others.

2. Harlan Cleveland first wrote about the DIKW hierarchy in a December, 1982 article “Information as Resource” in The Futurist, citing inspiration from the following lines written by T.S. Eliot:

Where is the Life we have lost in living?
Where is the wisdom we have lost in knowledge?
Where is the knowledge we have lost in information?

~T.S. Eliot, The Rock, 1934

3. Even the things we think of as facts are relative to the observer and a particular point of view. The problem of facts is as old as science itself and is still unresolved. See “Free the Facts” by Dave Gray, 2009.

4. Milan Zeleny, “Management Support Systems: Toward Integrated Knowledge Management,” Human Systems Management 7, no 1, 1987.

Posted on

How to draw a car

Part of my exploration of visual literacy is looking at how we might teach people to express themselves visually. Part of this is learning how to draw simple objects. This video shows how you can draw a car using very simple shapes.

Posted on

Some rules for effective business communication

Here are some basic rules of communication that could improve communication flow in a business. Some of them are fundamental “rules of engagement” that could set the overall tone for a company. Others are specific concepts for categorizing and measuring communication flow.

The idea is that the link between communication and action can be measured and reported. We do this for financial and product flows: Why not do it for our communication?

These ideas map to my concept of a new kind of email tool. Part of the idea requires that the sender classify every message as one of the following types: Information, Request, Order or Confirmation.

Rule 1: IROC. Classify all communications as one of the following:

  • Information: No reply required.
  • Request: Reply options are “Yes” or “No” (System asks “why?”). No response is considered “No”
  • Order: Reply options are “Accepted” or “Rejected” (System asks “why?”). System follows up aggressively when it gets no response.
  • Confirmation: Reply options are “Yes” and “No” (System asks “Why?)

Rule 2: Passive approval. “Yes” is assumed for all intra-company requests unless you hear “no” within 48 hours. “No” requires a rationale.

Rule 3: Brevity. Use short words. Use short sentences. Use short paragraphs. Be clear.

Rule 4: If it wasn’t said by email, it wasn’t said. “I told you on the phone last week,” “I told you in the hall” etc., are unacceptable.

Posted on

Q-tools: An approach for discovery and knowledge work

What is information design? And in a world where, slowly but surely, everything is being translated into, and expressed as, information, how do you begin to think about organizing the massive amounts of information we’re generating every day? What are the most successful strategies for thinking with, manipulating and organizing information?

Google home page

Google, one of the superstars of the information age, has recognized something important: that the internet does not need to be organized until you want to ask it a question. The entire Google business model is built around this premise: just-in-time organization. As soon as you have a question, Google will attempt to organize all available information in rank order, from most relevant to least relevant.

This would seem to be an ideal method for organizing information, and Google’s business success is a testament to this.

You can think of a question as a tool that you can use to increase your knowledge or reduce uncertainty. In fact, a question may be the most basic tool for gaining knowledge and working with information. If this is so, then it makes sense to ask which kinds of questions are best suited to different kinds of information challenges. A set of standard questions that can be applied consistently to different situations might form the basis of a standard toolkit for information discovery and design.

Q-tools

The list below attempts to define a set of “Q-tools” that may be used to generate, sort, classify and perform operations on information. This is not intended to be an exhaustive list, but more of a starting point for discussion. I have also added some alternative names for each Q-tool.

PrismA prism is a question that divides information into smaller groups. The purpose of a prism is to break down information into categories or subgroups. An example might be “What are the parts of this system?” Prisms are used extensively in scientific inquiry. They are also used in organization design to map the departments and sub-departments of a company. An example question used in this activity might be “What roles are required to deliver this functionality?” To create a prism, define a question that can be used to divide a unit of information into its constituent parts. Alternative names: Divider, separator, splitter, brancher.

Examples:
“What are the components?”
“What is it made of?”
“What resources are required to make this happen?”

RazorA razor is a question that divides information into two categories, based on relevance. The purpose of a razor is to define a boundary which can be used to sort any incoming information , and determine whether it lies inside our outside the boundary. A simple example: “Is this information relevant to me right now?” Productivity expert David Allen uses razors such as “Is it actionable?” as a method to process inbox information and increase productivity. A razor can also be used to sort things into rank order by asking of any two items “Which is more important?” To create a razor, define a question that can be answered only two ways, such as yes/no, or in/out. Alternative names: Cutter, fin, blade.

Examples:
“Is this relevant to the issue at hand?”
“Can it be used to…?”
“Can you eat it?”
“Can you find it in a grocery store?”
“Can it be used to make a cake?”
“Which is more important?”

GeneratorA generator is a question that has a potentially unlimited number of answers. The purpose of a generator is to explore new or unknown territory and to find knowledge that may be hidden and make it explicit. An example might be “What kinds of things are possible?” Open questions are used by inventors to stimulate discovery, for example: “How might we transport sound over great distances?” To create a generator, define a question that has an unlimited set of possible answers. Alternative names: Spout, spring, opener.

Examples:
“How might we…?”
“What might happen if…?”
“What kinds of things could…?”
“What is the meaning of life?”

PeelerA peeler is a single question that, when repeated, drives attention to deeper and deeper levels, like the peeling of an onion. The purpose of a peeler is to penetrate beyond superficial levels in order to get to the heart of the matter. An example might be “Why?” Taiichi Ono, the designer of Toyota’s production system, described “Five whys” as the basis of Toyota’s approach, because by the time you have asked why five times, the nature of the problem as well as its solution becomes clear. But a peeler does not have to use the word “why.” Another peeler results from using the question “How does that work?” This approach has served science for many years and has led to deeper and deeper levels of the onion, from atoms to particles and beyond [Is it fair to call this reductionism?]. To create a peeler, define a single question that can be asked repeatedly to drive to deeper and deeper levels of an issue. Alternative names: Ladder, drill, hammer, spike, repeater.

Examples:
“What else?”
“And?”
“So what?” (For developing a selling story)

FlankerA flanker is a question that seeks patterns or ideas that are similar. The purpose of a flanker is to think laterally and find an analogous situation that may help you think about things differently. An example might be “How might you approach this problem if you were a chef?” Flankers are used to break out of existing thought patterns that may cause blind spots. To create a flanker, define a question that approaches your situation from an oblique or unexpected angle. Flanker may be a subcategory of generator. Alternative names: Similator, Analogizer, slider, sider, shifter, sidestepper, lateral.

Examples:
“How might this look from the opposite side?”
“What would we do if we wanted to create this problem instead of solve it?”

SplicerA splicer is a question that combines information, or sets, into groups. The purpose of a splicer is to find larger categories and simplify collections of information. An example might be “What do these things have in common?” Splicers are used to classify information and build hierarchical information structures. For example, any invertebrate animal that has four pairs of legs, no antennae and no wings may be considered as a members of a single class which we call arachnids. Splicers are also used in a process called affinity mapping, a process designed to find similarities in large groups of information. Another example might be “What things feel like they belong together?” To create a splicer, define a question that can be used to find commonality. Alternative names: Merger, on-ramp, combiner, fuser.

Examples:
“What are the similarities?”
“What group does this belong to?”

A pointer is a question that has a well-defined or broadly agreed-upon answer, or set of answers. The purpose of a pointer is to gather specific information, for example: “What is your name?” Pointers are used extensively to collect information for the purposes of comparison and quantitative analysis. Census data would be an example. To create a pointer, define a question that is likely to have one, and only one, answer. Alternative names: Finder, getter, coupler.

Examples:
“What is your gender?”
“How old are you?”
“What is your primary address?”
“How many languages do you speak?”

Other Q-tools that might be worth exploring:

The essay question is particularly interesting. I think there’s a benefit to a tool whose purpose is to generate a story or narrative. For example, “Tell me about a time when…”

Another might be an estimator, to use when you need to make an educated guess. Such cases are quite common in business, because most business plans are actually a combination of goals and educated guesses. Budgets, agendas and project plans are examples. An example of an estimator might be “How much will this cost?”

Another way that q-tools can be useful is to help define the purpose of a question. For example, “What is the scope of the project” could be a generator (open question, exploratory) or it could be a pointer (specific request for information). There are many occasions when a pointer could look like a generator and vice versa, causing a lot of potential confusion. Common use of q-tools would make such distinctions explicit.

1. The questions we use in tests and school might be worth exploring:
– a. Word problem
– b. Essay question
– c. Multiple choice
– d. True/false (falls under razor)
2. The Zen koan, exploring the nature of paradox
3. The riddle

I’d appreciate any thoughts that you have for additional Q-tools, as well as your thoughts about what they might be called. It seems to me that tool names should reflect the use of the tool as much as possible (a drill drills, a screwdriver drives screws, a hammer hammers, etc.). I have chosen the names with that in mind.

I’ll try to keep this list updated as new Q-tools are defined.

Update: Q-tools are mentioned in “Formality in sketches and informal representation: Some informal reflections” and I was listed as an author. My first academic paper!

Posted on

Why PowerPoint rules the business world

A CALL FOR VISUAL LITERACY

In many organizations, the beginning and end of any business activity is marked by the PowerPoint presentation. In the early stages of an initiative, PowerPoint is used in strategy sessions, to present proposals and put forth plans. Later, it’s used for updates and progress reports. In the final stages, it’s used to report back and to present findings and conclusions. PowerPoint is everywhere, and it shows no signs of going away anytime soon.

Why is it used so broadly? And how did this simple tool become so entrenched in business? Here are a few thoughts:

  1. PowerPoint is accessible. For the novice, it’s easy to learn and use.
  2. PowerPoint is everywhere. Pretty much everyone has it, or has the ability to view a file. This makes it easy to share ideas and generally move meaning around. Slides can be borrowed, stolen, recycled and re-used.
  3. PowerPoint is flexible. The same document that is used to present information in a meeting or conference, can, with little or no modification, be emailed as a document or shared online, retaining much of its meaning.
  4. PowerPoint is easy to read. PowerPoint documents can be scanned and understood more rapidly than text documents. Because they are primarily visual they tend to be more easily understood and remembered.
  5. PowerPoint is modular. It can be broken down into single slides, which can be arranged and rearranged into numerous different sequences. Over time you can build up a storehouse of slides that represent your – or your team’s – collective knowledge about any subject, which can be distributed, shared, discussed and modified as things evolve over time.
  6. PowerPoint is powerful. For the more experienced user, it’s a powerful multimedia tool, with animation and other advanced effects. It’s easy to add information of any kind: Video, charts, photographs, maps – just about anything that can be digitized can be added to PowerPoint.

People use PowerPoint to represent knowledge, and the main element is relatively small and useful atomic unit we call the slide.

So what’s the problem?
“Death by PowerPoint” is the popular term for the much-dreaded meeting where a presenter subjects his audience to slide after slide, each one densely packed with bullet points or complex, confusing information, leaving the audience bored, frustrated confused.

But PowerPoint is not the enemy. When used appropriately, slides, and short sequences of slides, are excellent tools to represent knowledge. A good slide contains visual and verbal information in equal measure, and as an “information container,” a slide is just about the perfect size for memory and retention: big enough to hold meaningful information, but not so big that it’s likely to become overwhelming. A well-designed slide – one that’s comfortable to view and read – holds just about the same amount of information that you can hold in your short-term memory.

The problem is much, much deeper than PowerPoint. The issue is this: PowerPoint is a visual tool, and we are a visually illiterate society.

What do I mean by this?
In the modern world we are constantly confronted – you might say bombarded – with visual information: Television and film are the primary culprits, followed closely by billboards, brochures, and, yes, bullet points. Advertisers have long known that visualizing an idea is one of the quickest and most reliable ways to insert it into a human brain.

Due to this visual assault, we have, over time, become more sophisticated in our reading of visual information. In a world where information is digital, where photos can be altered in Photoshop and where films can show impossible things like dinosaurs and talking animals with a high degree of realism, we understand that seeing is no longer believing.

But this kind of visual sophistication is not literacy. Literacy is the ability to both read and write. If a child could read written language but not write it – if he could read a mathematical equation but not perform such operations himself – then we would not consider him prepared for success in the world.

In our school systems we teach our children the three R’s – reading, writing and arithmetic, because we believe them to be fundamental skills for successful integration in society. But the three R’s are no longer enough. Our world is changing fast – faster than we can keep up with our historical modes of thinking and communicating. Visual literacy – the ability to both read and write visual information; the ability to learn visually; to think and solve problems in the visual domain – will, as the information revolution evolves, become a requirement for success in business and in life.

PowerPoint has risen to its current position for two reasons:

  1. We’re processing more information than ever before, at unprecedented volumes
  2. We don’t have as much time to read anymore, and
  3. Much of the information we need to share is non-linear in nature.

PowerPoint is a visual medium. If you want to convey information visually, it’s the most accessible and ubiquitous tool there is. The answer to bad PowerPoint is not to eliminate the tool, but to improve our visual literacy. We need to teach visual literacy in our schools, and to our business people. We need an ABC book of visual language (a project I am working on).

We’re leaving an industrial age and entering an information age, yet we continue to teach, and operate our schools, as if they were factories. In an information age, visually literate societies will succeed and thrive. Shouldn’t we be one of them?

Posted on

The whirl

I talk about some of the cognitive challenges of the information age, and why visual language is an important tool for dealing with them.