Graph Wisely: An Interview With Cole Nussbaumer Knaflic

Graph Wisely: An Interview With Cole Nussbaumer Knaflic

Isa Gautschi
Authored byIsabel Gautschi
Sean Campbell
Authored bySean Campbell

In this episode of the B2B Market Research Podcast, Cascade Insights CEO Sean Campbell spoke with Cole Nussbaumer Knaflic, author of the book Storytelling with Data: A Data Visualization Guide for Business Professionals and the blog Storytelling With Data.

Graph Wisely: An Interview With Cole Nussbaumer Knaflic

Their conversation covers:

  • Why the audience of your presentation needs to be defined beyond “internal audience” or “external stakeholders.”
  • Why graphs aren’t always the most effective way to communicate your point.
  • The best times to use line graphs, bar graphs, and tables. (And why pie charts are evil.)
  • How to structure an effective presentation.
Episode 123: Graph Wisely: An Interview With Cole Nussbaumer Knaflic
You can listen to the episode or read an edited print version below. The transcript has been modified for clarity and readability.[sc_embed_player_template1 fileurl=”″]


Today on the podcast I have with me Cole Nussbaumer Knaflic. She is the author of a great book called Storytelling with Data, which we’ll discuss on today’s podcast.

Cole, welcome to the podcast.


Hi, Sean. Thanks for having me.


Great to have you here. Let’s dive in.

When defining your audience, why do you think people should go further than internal or external stakeholders?


Probably first and foremost because that “stakeholders” term is such a big glob. When people target an external audience, they often design presentations for their product, or project, or data without pausing to think about the people they will be presenting to. They don’t consider how they are, what needs they have, and how to successfully communicate with them.

For me, it’s important to have a really clear picture of who that somebody on the other end of your communication is. When we talk about defining audience, there’s a real strategy in being able to pinpoint a very narrow audience, a specific person if you can get it that narrow.

If you can identify a specific person or group of people, you can ask:

  • What does that person care about?
  • What keeps them up at night?
  • How can I frame what I need from them in terms of what resonates with them?

Rather than preparing your presentation for yourself or for your work you should prepare it for the audience member. For me, “internal audience” or “external stakeholders” are so broad that we lose that personal connection with our audience in our presentations and the data included in them.


To take that one extra step, if you’re going to give a research presentation to a business group, you might think through how you would split that main presentation into separate presentations tailored to each of the different personas you might encounter. You may want to prepare a presentation for the product management team, a different presentation for the marketers, and a separate the presentation for the sales team.


Absolutely. You can imagine how each of those teams are going to have different needs. They’re going to have different things that they care about, different things that they don’t care about. By thinking of them as distinct and separate audiences you really put yourself in a better position for success.

If we try to communicate to too many at once, we don’t exactly meet the needs of any one of those audiences. When you start covering something that’s less relevant, you run the risk of people tuning out. At that point, it’s hard to get their attention back.


Absolutely. In the book, you say that the fact that you have numbers does not mean you need a graph. I’m smiling now and I almost laughed out loud when I read it. Unpack that for me a bit.


You want graphs to create an ah-ha moment!

Like a lot of things that I cover in the book, when you say it, it sounds obvious, yet we don’t put these obvious practices into practice so often. I think there’s a growing amount of data out there, certainly. Along with that comes a growing desire to be data-driven, to inform decisions with data, and to demonstrate robust analysis.

There’s a huge impulse to put data in everything that we do. It’s a great goal, but we don’t want to let it cripple us. Sometimes I’ll see a graph that, when you unpack it, really only has one or two numbers. The power of those 1-2 numbers may be lost because they’ve been packaged in this way that doesn’t make the information come across in a way that creates that magical, ah-ha moment.

You want graphs to create an ah-ha moment!

My advice is:

  • If you just have a number or two that you want to communicate, think about using that number directly.
  • You may not always need to use data. Question that assumption.

I come from a highly technical, analytical background where data drives everything. As an analyst, it’s hard to hear that data is not always the answer. It’s important to pause and ask yourself, “What is my message? What do I need to get across? How do I do that in the best way possible?” It may involve data, or it may not. For example, a story or a quote from a customer may be more effective in helping you to drive change.


These are some really good points. Your comment about using a story is spot on. I often notice that folks will get so focused on the data that they forget that there’s a narrative to it. They forget that the data and the narrative need to be shared jointly to really drive a meaningful impact.

You mention at another point in the book that using tables in presentations fires off the reading center of your brain.

That part made me think of all the times I’ve put tables in presentations. I’d like to say that most of the time I was doing it because I wanted stakeholders to read what was in the table. But that’s not a good enough answer.


When it comes to presentations, my general rule is don’t ever use a table…There is usually another way to visualize the information that will allow you to make your point more quickly and effectively for your audience.

People interact very differently with tables and graphs. Tables, as you mentioned, interact with our verbal system, which means that we read them. So, if I put a table up in front of my audience, I’ve immediately lost their attention to what I’m saying because they’re stopping to read the table.

They’re looking at different data points and mentally storing numbers in their heads so they can compare with other numbers. It’s a highly taxing process. When that happens, our audience’s ears turn off to whatever we’re saying. Even if we have a fantastic point we were going to make with the table, we might lose the ability to get that point across because we’ve lost our audience to the reading of the table.

Graphs, on the other hand, interact with our visual system. Our visual system is much faster at processing information than our verbal system.

When it comes to presentations, my general rule is don’t ever use a table. (There are exceptions to this.) If you catch yourself about to use a table, just pause and ask yourself why. What point are you trying to make with that table? There is usually another way to visualize the information that will allow you to make your point more quickly and effectively for your audience.


Really good points. I also wanted to ask you about your take on pie charts. You say that they’re evil. Walk us through why you think they’re evil.


Let me preface this by saying reasonable people may disagree with me on this point. My view is that pie charts are not a good way to show data. There is good reason for that. When we’re showing data in a pie, we’re asking our audience to try to compare angles and areas. Our eyes don’t do a good job of ascribing quantitative value to 2-dimensional space. Basically, pie charts are hard for people to read.

With pie charts, it’s almost impossible for people to compare values within two segments. At best, you can make some general observations like, “This piece is bigger,” or, “This piece is smaller.” However, when segments are close in size, it’s nearly impossible to tell which is bigger, which is smaller, and to make nuanced comparisons.

With pie charts, it’s almost impossible for people to compare values within two segments.

Pie charts also lend themselves really easily to other bad things when it comes to graphing data like 3-D, meaningless color, and the sort of glitz that distracts from the point we’re trying to make with the data. (Donuts, by the way, are even worse than pie charts.) There is usually a better way of getting the information across than a pie chart.

Getting to the second part of your question on types of charts, I actually took a look at my work over the course of a year and categorized every single graph, every single data visual that I made.

I was surprised, because I thought it was going to be a really long list with lots of different types of charts. Actually, there were only 12 categories of visuals that I found myself using on a regular basis. (In the book, I talk about the pros and cons of using each category of visual.)

Probably the biggest take away is that, for the most part, the graphs that I was using were not anything crazy. I used a lot of your standard bar graphs and line graphs. Not pies, though.

The reason for that is that people instinctively know how to read bar graphs and line graphs. There is less of a learning curve for your audience.

In general:

  • Line charts are fantastic for continuous data, typically for some unit of time.
  • Bar charts are fantastic for showing categorical data, any data that’s organized into categories. Bar charts are really easy for our eyes to read. You just have to compare the end point of those bars. It’s easy to see quickly which category is the biggest, which is the smallest, and also incremental differences between categories.


Excellent. Hopefully we’ve reduced pie chart usages across America by at least 1 percent based on that.

Why do people tend to fear white space on a page? I found this part of the book fascinating because we’ve got a client that is horrendously bad in this way. We will get decks from them as they try to get us up to speed on a business unit or whatever. It’s like reading a short novel per slide. So a related question is, how do you educate clients on how to separate the presentation from the read-only deliverables?


Studies have shown we have something like 3-8 seconds with our audience when we put something in front of them.

It’s an interesting phenomenon. I can totally relate to it. My first career was in banking. You’re handed situations where you get one or two slides. You want to pack as much information onto those slides as you can. You end up with 10-point font, five different graphs, and a lot going on. I can understand the motivation, but this actually does a great disservice to our story and our data.

With so much on the page, our audience may say, “Whoa, that’s going to take me a lot of energy or a lot of time to figure out what’s going on. You know what? I don’t even want to deal with it.” At that point, we’ve lost the ability to communicate with them altogether.

Studies have shown we have something like 3-8 seconds with our audience when we put something in front of them. When we project a really dense slide, that 3-8 seconds is spent deciding whether they’re going to continue to pay attention to what we’ve put in front of them or move on to the next thing. You want to consider how to structure your slides, presentations, and data visualizations so that if you only get 3-8 seconds of your audience’s attention, you can still get the main point across. Probably the biggest lesson is that less is more.

White space, in particular, can be really useful for emphasizing the pieces on the page that are not white space. If you have one thing that’s really important to get across to your audience, you want to think about making that the only thing that’s on that page.

I sometimes think of the parallel between pauses in public speaking. If you get somebody up in front of you, and they’re talking a mile a minute, it’s unclear they’ve even paused to breathe, and you want to ask a question but they keep going… Well, that’s just uncomfortable.

Now, think of a speaker who says, “Death to pie charts.” Before moving on, they pause to let that statement sink in. Like that silence, negative space like white space can really work strategically to emphasize the message you’re trying to get across.


In the book, you discuss about pre-attentive attributes. The way I understood it, pre-attentive attributes are necessary for those circumstances where you just have to have a large block of text.

This is something that an average qualitative researcher deals with all the time. You have this meaty quote that you can’t really break down because it needs the context.


Pre-attentive attributes are sort of like setting the stage. Strategic and sparing use of things like color, size, and a visual on a page can allow you to create a hierarchy of information. Let’s take the example that you brought up. You’ve done some qualitative research. You have this really awesome verbatim quote that you want to use, but it’s long. You decide to use the whole quote, but to present it wisely.

Consider what you want to jump out at them when they take that 3-8 seconds to scan.

So, you start out by making all of the text gray. You consider what you want to jump out at them when they take that 3-8 seconds to scan. Then, you select the pieces of that quote that you really want your audience to notice. Maybe you highlight a key word or a phrase within the quote. Leverage those pre-attentive attributes and make them a different color, make them big, make them bold so that they’ll stand out. If you’ve successfully captured the audience’s attention, then the rest of the quote and context is there for them to read as well.


Those are great points, especially the point about making it all gray in your mind and then up-leveling what you think is the most important thing to highlight.


I’ll often start by doing that with a page in general, whether it’s a graph or text.

It’s a good practice to:

  • Start by pushing everything into the background.
  • Determine whether there is anything there that you can actually get rid of altogether.
  • Consider where you want to draw your audience’s attention and in what relative order. Then bring those pieces forward through color, size, bolding, etc.


You discuss horizontal logic in the book, which really resonated with me because it’s something I’ve always been passionate about. I’ve built a lot of presentations over the years and done a fair amount of training. I very consciously do this with my own decks, but I never really had a phrase for it. You say that when you read the titles of the slides together, the story should make sense.

I tell our own team that they need to think about that. I tell them, “Yes, the deck will be presented, but the deck will be shared to a much larger audience who will open it up and go page down, page down, page down. Some people will just be glancing at the titles. If you don’t have a narrative that grabs them and makes them go, ‘I need to read this stack of slides,’ you’ve missed an opportunity.”


First off, horizontal logic is this idea that you can lay out the path that you’re going to take with your presentation or your report up front. You can think of this as the executive summary for your presentation. For example, you may want to start with a list where each bullet point describes the content you’re going to cover. When you read those bullet points together they should tell a story, a narrative.

Organize your following slides in such a way that every slide title corresponds, in the same order, with one of those bullet points from the executive summary. Basically, introduce what you’re going to say, and then you say it and go through the detail on each of the slides. To your point, you can read just the slide titles and it tells that same story as the executive summary up front.

Horizontal logic is this idea that you can lay out the path that you’re going to take with your presentation or your report up front.

Often times I’ll end with a reiteration of those same points along with strong emphasis on where a decision is needed or where I’m trying to drive people to action. The power of repetition is key. The more things are repeated, the more likely they are to be transferred into our long-term memory. Setting up the horizontal logic in this way, you basically go through different iterations of the same content multiple times. You pick up that executive summary up front, then the slides that follow, and then recap at the end.

Designing presentations in this way helps your audience to form bridges between their short and long-term memories. This, of course, increases the chances that they’re going to be able to remember your message. It can feel really repetitive to the person putting together the presentation, since they are already familiar with the information and the point of their story. However, the audience is much less close to the information, so reminders are really nice for them.

You should:

  • Introduce the topic, letting the audience know what you will be talking about. Outline the story you’re going to tell.
  • Fill in the narrative with further detail.
  • Recap the story.

If anybody missed something at one point, you’ve covered it a couple of other times. Your audience won’t feel like they missed the takeaways.

The workshops and conferences I attend are usually structured this way: “All right, folks, here is what we’re going to cover today. Here are the goals.” You’ve got this expectation upfront of what we’re going to be going through, then go through the details, then recap. To your point, one critical component of this structure is action titles. The title bar in PowerPoint is precious real estate. Don’t under-utilize that space with boring descriptive titles like “2016 budget.” Heather is not going to get anything out of that other than “we’ve got some data on what the 2016 budget is. “

Instead, try a title like “We’re Trending Above Budget” or “We’re Trending Below Budget.” You want a title that gives some context into what you are actually going to talk about on the slide. For the person who is just flipping through and reading those slide titles, you want them to get an idea of what you’re going to say about the budget, not just that a budget exists. (They don’t need a slide to know that.) You want to communicate the story through your titles.


Fantastic points, and ones I think everybody could use. I’m going to guess that with our audience, it’s highly likely that in the next 48 hours, most of them will be editing a PowerPoint. What would you hope that they do as they edit their presentation?


That’s an interesting one. The first thing I’d say is, “Step entirely away from the PowerPoint.” Close your laptop, get yourself in front of white board or a blank piece of paper. Pause and think about your audience. Identify who they are. Be specific. Clarify for yourself the main point you need to communicate to your audience through this presentation. In other words, if they were going to walk out of that room after the meeting and they only remember one thing, this would be the one thing you want them to remember. Actually write it down in a single sentence.

The first thing I’d say is, “Step entirely away from the PowerPoint.”

This sounds like an easy thing to do, but it’s actually very difficult. It’s especially difficult when you’re close to your work. You have put a lot of effort into finding this information and there is a lot you could say about it. However, there is only so much an audience can remember, so choose the most important story to communicate.

Once you’ve clarified your primary message, go back to the PowerPoint and consider how to arrange your information to communicate your point as clearly as possible. It might mean starting over with a blank template, you may need to rearrange things to create the desired narrative structure, or you may want to push some information to the appendix so that your narrative isn’t cluttered.


These are fantastic points. For more on how to effectively tell stories with data, check out Cole’s excellent book. Cole, thanks again for being on the podcast.

This podcast is brought to you by Cascade Insights. We specialize in market research and competitive intelligence for B2B technology companies. Our focus allows us to deliver detailed insights that generalist firms simply can’t match. Check out our podcast for more B2B Market Research episodes and articles.

Image courtesy of Shutterstock.

Home » Blog Posts » Graph Wisely: An Interview With Cole Nussbaumer Knaflic
Share this entry

Get in Touch

"*" indicates required fields

Cascade Insights will never share your information with third parties. View our privacy policy.
This field is for validation purposes and should be left unchanged.