Evergreen Data Visualization Blog

Guest Post: Getting Started with Graphic Recording

By Lydia Hooper, Fountain Visual Communications

Data visualization professionals often focus on numbers, helping to tell the story of what, when, where, and how much. But more often than not organizations first need to better understand the why and how.

Before there is a need to communicate key insights from quantitative data, there is always a need to communicate first about what insights will matter most to the people involved, and therefore what data to plan, collect, analyze, and present.

Graphic recording is a method for using visuals to support communication and understanding during real-time dialogue. Hand-drawn illustrations allow teams to simultaneously collect, analyze, and report information about people and groups.

There are several reasons why graphic recording is hands-down a great method for visualizing information:

It’s an opportunity for thorough reporting. Graphic recording captures both broad themes and the actual language used.
The method mirrors the content. By portraying social processes, the visual both reflects and highlights the interactions themselves.
Relationships are kept central – and they are supported. Real-time graphic recording allows partners and stakeholders to better communicate and understand one another in the very moments when decision making is happening.
It’s exploratory in nature. Discussions are inherently dynamic, not concrete, singular, or fixed. Graphic recording allows for multiple realities to be explored simultaneously, helping everyone better understand the complete, evolving context.
It allows for remembering and reflecting over time. Groups of course change over time. Participants can revisit the visuals to not only recall what was discussed but to remember their personal experience was. They serve as a fun, ongoing feedback loop for continuous, thoughtful decision making.

When I first started helping organizations with communicating about data and complex topics, I quickly gathered that they needed more help with internal and partner communications to reach meaningful, shared understanding than they did with external and mass communications to persuade anyone else.

Unfortunately, too often these internal communications are overlooked, side-stepped, or rushed through. Social dynamics are complex and navigating them can be overwhelming. As with any good data visualization, graphic recording makes understanding this complexity easier and communicating about it transformational.

For example, during one meeting I graphically recorded, the two groups that had been convened had so little understanding of one another that they spent almost an entire hour debating about whether to proceed with the agenda for the meeting. By the end of the meeting, they could clearly see that the solutions to the questions and concerns that they had initially raised were easily found among one another.

The process of graphic recording can support internal dialogue as well as any communications that follow. Many times a graphic recording I created during a team’s process became a visual that they proudly shared with others as a representation of not only the “who” and “what” but the “why” and “how.”

Perhaps the best news of all is that graphic recording is a tool that doesn’t require extensive planning or even knowing software to begin using it. Like all data visualization, it does require a certain degree of visual clarity and, above all, the willingness to stay true to the data and to consciously engage in a thoughtful process in order to share it in a way that will be meaningful and effective.

Here’s some tips for getting started:

Practice listening objectively. As always, content is king. To make sure you are capturing the most important ideas and themes, you need to both listen without judgement and allow time for thinking and synthesizing before making marks.
Learn the basics. Whether you are using markers on paper or a stylus on a tablet, you’ll need to slowly master lettering, using bullets, lines, arrows, and boxes, and drawing people and faces. If you are, like most people, unfamiliar with and/or intimidated by drawing, start with Dave Gray’s visual alphabet below.

Build your vocabulary of visuals over time. Again, focus on developing this based upon key concepts rather than exciting icons or complex metaphors. For example, when people are describing emotions I often use hearts and when they are talking about ideas or “aha”s I often use lightbulbs.
Mind the big picture. Draw connections between ideas, use size to emphasize ideas or themes, and consider layout ahead of time (many folks start using graphic recording by using pre-existing, pre-drawn templates).
Practice, practice, practice. Only real-time practice will flex the muscles you need to write and draw faster, capture content more accurately, and become more helpful to the folks in the room. The easiest way to start may be to capture one-person talks or podcasts in a private notebook (also known as sketchnoting) and then work your way up to capturing discussions publicly.
Leverage our supportive community. Connect with a local chapter of org, follow graphic recorders on social media, and ask us questions. We love helping others learn more about this incredible, little-known tool!

Lydia Hooper specializes in helping organizations collaborate and communicate about complex topics. She has partnered with more than 50 organizations and networks, offering services and trainings in data storytelling, graphic recording, and communications strategy. You can read more blog articles and get a free copy of her ebook about graphic recording “Using Visuals to Support Collaborative Work” at www.fountainvisualcommunications.com.

Qualitative Chart Chooser 3.0

Stephanie’s Note: A year-ish ago, we released our qualitative chart chooser and we loved it to pieces but, like most things in life, we immediately wanted to change some things. We collaborated with some researchers steeped in qualitative data collection and analysis to rethink the chart chooser and released it in this post.

We continued to evolve the chart chooser and now it is in Effective Data Visualization, where Chapter 8 includes the largest compendium of qualitative chart choices available. Join us at the Evergreen Data Visualization Academy where we teach you how to make these graphs in Excel, Tableau, and R.

Less attention has been paid to how to display qualitative or text data visually. Here at Evergreen Data, we are trying to tackle that. Qualitative data gives us more power to engage people’s hearts and minds. We are able to extend our data story to a more personal level.

When we are being told a story, not only are the language processing parts in our brain activated, but so is any other area in our brain that we would use when experiencing the actual events of the story. If someone tells us about how delicious certain foods were, our sensory cortex lights up. If it’s about motion, our motor cortex gets active.

Our main goal in communicating with our audience is that they will remember our main point, right? Well, when we tell people stories they are better at remembering what we say because more parts of their brain are engaged. Stories inspire people to take action.

All of that to say, qualitative data gives us more power to tell our audiences engaging data stories.

Sounds like it should be simple to effectively visualize qualitative data but it can be tricky. That is why we have all seen SO MANY bad qualitative visual representations, like endless pages of bulleted quotes. Yuck! No one likes that. It is not engaging, it is cognitively demanding, and it is super boring. At Evergreen Data, we are committed to providing you with useful tools to help discern which visual fits your audience’s needs and story.

That’s why we are proud to introduce the Qualitative Chart Chooser 3.0!

What’s different?

In the new version, it was important for us to incorporate (1) the overall nature of the data, which drives the kind of story you can tell, (2) account for whether, in your analysis, you want to quantify the qualitative data or keep it purely qualitative, and (3) whether you want to highlight a word/phrase or display some kind of thematic analysis.

Our goal is to make this chart chooser useful for all, whether you are a die-heart qualitative person or a quantitative person who periodically asks open-ended questions.

All other visual techniques we teach at Evergreen Data still apply to all these qualitative visuals. Let’s take something like a timeline (one of my favorite qualitative visuals).

Here is a nicely designed timeline (done all in PowerPoint). It visualizes different stages of an project overtime. No matter what way you slice it, this is slightly overwhelming. Remember your data viz secret weapon, whenever a visual gets overwhelming, break it up into small multiples. This is one of the most useful techniques to apply to qualitative visuals. So when you include a project timeline like this in a scope of work, present it in whole form first. Then as you talk through each part of the project, show just a small portion in the margin. I also threaded color coding into the different sections of the scope of work.

We hope this refined qualitative chart chooser will make it easier for you to explore effective qualitative visual options. We are always learning and developing this tool to better fit your visualizing needs, so keep in touch with us on your feedback as you put this to use. And stay tuned for more blogs on qualitative visualization. If there’s something on here that you haven’t heard of before (journey mapping, anyone?) we will tell you all about it. You can get started with our growing collection of qualitative visuals.

Learn in the Academy!

You can find step-by-step instructions on how to make 60+ awesome visuals in my Evergreen Data Visualization Academy.

Video tutorials, worksheets, templates, fun, and a big-hearted super-supportive community. Learn Excel, Tableau, R or all three. Come join us.

Enrollment opens to a limited number of students only twice a year. Our next enrollment window opens April 1. Get on the wait list for access a week earlier than everyone else!

Master Dataviz with Graph Guides!

Our newest program, Graph Guides, is a custom-built, year-long sprint through 50 Academy tutorials.

When you enroll, we’ll assess your current data viz skill set, build you a customized learning path, and hold your hand as you blaze your way to new talents.

We open enrollment to 12 students at a time and only twice a year. Get on the waitlist for early access to our next enrollment window.

Book to Read: Innovative Evaluation Reporting

When I first started talking about presenting data effectively, Kylie Hutchinson came up to me in her bad ass black leather jacket and said if I needed a mentor, she was available. That’s how cool she is.

When I talk about how effective it can be to introduce the element of fun into our data presentations, I always talk about Kylie’s work. That’s how cool she is.

And now she wrote up her ideas into a book. That’s how cool she is.

I love this book so much. You should read it. Here are a few of my favorite parts:

Develop a Communications Plan

Kylie guides readers through a process of detailing each of their main audiences, how those people want to receive information, how much that could cost, when they need it, and where they fall on the priority list.

In my workshops, clients usually express frustration that they have multiple audiences and don’t know how to meet everyone’s reporting needs. Kylie’s straightforward activity can clarify who gets what, when, so that the action plan is clear.

Layer Your Content

Kylie and I both talk about how we need to multiple report formats, each with a different number of details, to form a bit of a bread crumb trail so that audiences can always get as much information as they would like from us without having to wrestle with a long report. Kylie refers to this as layering the content and she uses a hamburger metaphor to explain what layering looks like.

Not everyone can digest the burger, ya know? So let’s give them some other edible pieces.

Physical Items

Perhaps my favorite part of Kylie’s work is her focus on the use of physical items to get audiences engaged with data. Sometimes a dot plot just won’t be enough to get your audience geeked. Sometimes they need to have fun. And Kylie has loads of ideas, many of which are available for free on her website.

This is Kylie’s reporting cube, a tool I have used with clients in the past. You type your data into the textboxes, print out the sheet, and cut and fold as directed until you have formed this 3D physical item. Then audiences roll the cube (it’s not really a cube, and this is where I have a bone to pick with Kylie) and whatever they land on, they talk about. What a great way to make data digestible for people who are normally data scared.

Kylie’s new book has loads of just pure gold. It’ll make you better at reporting, whether or not you work in the field of evaluation. Her ideas are widely applicable.

PS. Her section on flow charts is my new mantra. It’s incredibly insightful.

PPS. Don’t be put off that the book’s title includes the word “evaluation.” Even if that is not your field, this book is for you. If you ever need to report your work out to people and you find yourself having a hard time engaging them, you need this book.

Order here.

Announcing! Kauffman Foundation Evaluation Report Guidance

The folks at the Ewing Marion Kauffman Foundation are serious about investing in the greater good and equally as serious about making sure those dollars have an impact. To that end, the hire organizations to investigate and evaluate what’s going on in Kauffman-funded programs. Those organizations apply the most appropriate research methods, collect a bunch of data, and report back to Kauffman what they found.

To help those organizations (and you!) create the kind of report that is going to be most useful for decision-making and action, we collaborated on a set of guidelines around report layout and design. Are you part of this same tribe, needing to get information to decision-makers in a way that’s useful and functional? Grab this guidance doc.

We intentionally left room for your own organizational branding. I’m not going to try to tell you what colors or fonts you should use. But the guidance doc does provide some basic structural advice for reporting. We elaborate the order in which the report sections should occur so that readers get the bottom line up front. We detail what (Kauffman) decision-makers want to see in an Executive Summary – and I share what a sample exec sum could look like.

I go in depth about how a report should state data-based claims and then provide visualizations of that data to support the claim. In other words, let’s graph! The guidance doc provides several graph examples, including how a graph fits in with the narrative text around it. And I show you how to keep the graphs from moving around on you in Word (such a pain, but an easy solution).

The guidance doc even includes an example of good table design. And how to make and use call out boxes and sidebars. And what graph report covers can look like. And so much more.

So while this guidance document was specifically developed with Kauffman consultants in mind, they are good people who know that many of us could use this kind of boost and that’s why is it accessible to anyone, for free. High fives.

In that guidance document, we point people to some of my other resources that give up even more detailed direction. See:

Evaluation Report Layout Checklist

Data Visualization Checklist

Effective Data Visualization

How to Make Dumbbell Dot Plots in Excel

Data visualization is so cool because it helps you see things that would otherwise take a looooooot of effort. Here’s an example.

Some very sweet clients in Maricopa County, Arizona (that’s the greater Phoenix area, friends) had a habit of presenting super important data in the most hard to digest way: a table.

Tables are super hard for people’s brains. Our brains can’t really process more than 3-5 chunks of information at once and this table has, uh, a little more than that. Our brains have a hard time tracking a row of numbers, comparing back and forth to see where there were increases or decreases, which race’s uninsured rates are higher than another, and when. Sheesh. It’s just too much.

So we plotted this data into a vertical dumbbell dot plot. God, it’s amazing.

Just look at this thing. For the first two races, the dots are clustered close together, meaning people of those races in Maricopa had similar uninsurance rates as compared to the whole state of Arizona and the whole United States. Ok. Then things get totally crazy. I mean, the blue Maricopa dots are all over the place for American Indian/Alaska Native. The very high blue dot in 2009 for Asian is so stark, it actually makes me think there was something up with data collection that year. Seriously, wow! A table ain’t going to tell you that!

So. Here’s how to make that dumbbell dot plot:

Hold down the Control key while you select all the data for one race. It’ll be in three different rows in this table. Don’t worry about the locations in column A or the years in row 1. Then insert a line graph with markers.

Then right-click in the graph and click Select Data. You’ll see this new box open called Select Data Source. On the left are the three series of data you highlighted. But they all say the name of the race, not the location. So click the Edit button in the middle and select the location cell for Series Name. Do this for each of the three series.

On the right side of this box, it says Horizontal (Category) Axis Labels and right now the box just has 1-6 listed. This is where the years should go. So click the Edit button and select the cells with the years.

Now that the right labels are in place, let’s adjust the line with markers. We essentially want to get rid of the line and keep the markers. So right-click on a line, and select Format Data Series. In the Line menu, select No Line. Then in the Marker menu, make the marker larger and adjust the colors. Repeat these steps for each line.

The critical hack to make this into a dumbbell dot plot is to add the vertical line that connects the dots. It’s easier than you think. With the chart highlighted, you should see a dark green tab that says Chart Tools. Under it you’ll see 2 tabs (newer versions of Excel) or 3 tabs (older versions). If you see 2 tabs, click in Design and look for the Add Chart Element button on the left. Open that and select Lines and then High-Low Lines. Boom! If you see 3 tabs, click the center Layout tab, look for the Lines button, then select High-Low Lines. Old school Boom!

Adjust the y-axis scale. Since this graph just shows White data, we know we will need many more graphs. So look for the highest value in the table. It’s 40%. Set the max for the y-axis scale to 0.40. Then adjust fonts and legend placement and title label.

Copy and paste this graph next door. Right-click on your new graph and go back into Select Data. In the 3 series listed on the left, edit each one and select new series values for your next set of data. This will keep the colors, fonts, and scale intact, but the graph will visualize uninsured rates for a new race and the whole process will take you probably 30 seconds. How very cool.

One of the many benefits of data visualization is that it helps you gain insights in the data that you wouldn’t otherwise be able to see without significant effort. Tables bury information because they are so hard to digest. If our job is in any way to communicate our data, a table won’t cut it. Visualize! Maybe with a dumbbell dot plot!

We have video help for this tutorial plus instructions for Tableau and R in our Academy and Graph Guides programs.

Learn in the Academy!

You can find step-by-step instructions on how to make 60+ awesome visuals in my Evergreen Data Visualization Academy.

Video tutorials, worksheets, templates, fun, and a big-hearted super-supportive community. Learn Excel, Tableau, R or all three. Come join us.

Enrollment opens to a limited number of students only twice a year. Our next enrollment window opens April 1. Get on the wait list for access a week earlier than everyone else!

Master Dataviz with Graph Guides!

Our newest program, Graph Guides, is a custom-built, year-long sprint through 50 Academy tutorials.

When you enroll, we’ll assess your current data viz skill set, build you a customized learning path, and hold your hand as you blaze your way to new talents.

We open enrollment to 12 students at a time and only twice a year. Get on the waitlist for early access to our next enrollment window.

Data Nerd Thank You Cards

In years past, I designed for you cards to help you celebrate Valentine’s Day and the winter holidays like the data nerd you really are. But truth be told, my favorite holiday is Thanksgiving. I love the food, the focus on family, and the deep expressions of gratitude. Does your family do that? Yes, I am the one that makes everyone go around the table and list off the things they are thankful for. I usually tear up a little. Then food. I mean, what better holiday is there??

So I made you some thank you cards so that you can express your gratitude to friends, family, sweethearts, and clients.

All cards are purchased through Zazzle, who prints and ships you as many as you want. Enjoy, and thanks for being a part of this with me.

Back to Back Bar Chart Thank You Card

My Friend Venn Diagram Thank You Card This is Significant Thank You Card You Score High on All My Survey Questions Thank You Card Dot Plot Thank You Card

Evergreen Data Visualization Academy subscribers – heads up, you have a 10% discount on these. Coming soon through our Academy newsletter, so sit tight and just obsess over which one you love the most.

I’ve made data nerd cards and gifts for other occasions, too. Check out my Holiday swag (flasks!) and Valentine’s cards.

Anyone Can Do Graphic Design, Even You

This is the kind of post you either love or hate. You hate it if you are a graphic designer.

When graphic designers are in my workshops, they usually spend the first part nodding their heads in passionate agreement because I’m educating their research team on things like font choice and color psychology – things they’ve been trying to communicate for years.

We get to the part of the workshop where I talk about making chart choices and graphic designers listen carefully because this is new information for them.

Then I get to the part about slide and page layout and I say things like “graphic design is not rocket science” and the graphic designers in the room turn icy cold. I say “this is just about inserting some circles or rectangles – anyone can do it” and the graphic designers start drafting their protest letters.

But it’s true.

At it’s core, graphic design is really about using some very basic elements, like shapes and lines, adding in colors and textures, and playing with size. Anyone who has ever clicked Insert>Shape inside PowerPoint can do some basic graphic design.

I punched up this short report template by adding in some incredibly basic rectangles along the left hand side of the page, varying their heights and colors.

Or in this next example, showing a two page spread, we added in some large rectangles behind certain text to section it off. Then we layered in some semi-transparent circles of varying sizes in the Resources box and on top of the stock photo. Just shapes of different colors and sizes.

All that said, the real skill lies in the thoughtful combination of those basic elements and that’s what graphic designers have in both education and experience. So, no, you shouldn’t rip up your existing contracts with graphic design firms. But you can learn some of the basic principles of graphic design for those moments when you have no budget but need your slides or handouts or reports to look a bit more polished. You, too, can do graphic design.

I share tons of ideas on how to do this, even what buttons to push, in my latest book, Presenting Data Effectively.

Don’t Visualize

Here’s the deal: People primarily look at pictures. That’s why we visualize data – to get people to look at it.

But if we waste the short amount of attention people are willing to extend to us by showing them visuals that don’t convey a point, we quickly teach them to ignore our pictures. We abuse the power of the visual and our audiences are more likely to miss the visuals that do carry a message because we’ve told them our visuals aren’t important (or are! – it’s a guess!).

So only graph the things that are important.

It took me a long time to learn this lesson. I spent a lot of my upbringing, like many of you, pouring over survey data and studiously graphing the responses to each question (just like SurveyMonkey will do for you). I was making pie charts by the truckload. Because we were very good survey methodologists, we would have been sure to ask our demographic questions upfront, at the start of the survey.

And our reporting habit back then, like many of you, was to report the data in the same general order that we asked the questions on the survey.

All of this meant that we had pages and pages of pie charts for each of those demographic questions at the front of the report. But they usually carried no real point. We were just showing demographics. Which means people spent their precious attention span looking at meaningless pie charts, initially investigating a little to see if there was something important in the data visualization, but quickly learning our visuals meant nothing. Then by the time we got to the good stuff, if our readers will still even with us, they were visually exhausted.

So let’s reserve that power and only apply it to the things that matter. Like our findings. Like our action items.

It’s a win for us because we save time not making graphs that don’t matter. And it’s a win for the audience who wants to see what we think is important.

Evergreen Collections

The end of August marks an important moment for me because it is when I quit my salaried job to work on data visualization and design full time. 5 years! It remains one of my best decisions. So, I like to celebrate it with you.

Last year, I launched the Evergreen Data Visualization Academy. Enrollment will open again in the next couple of months, so you might want to get on the waiting list.

This year, I’m giving you Evergreen Collections.

On this page, you’ll find a collection of all of the blog posts I’ve ever written that are step-by-step guides on how to make data visualizations in Excel. These rolled over, with lots of other content, into one of my books, Effective Data Visualization. I have 14 tutorials in that collection right now and it’ll grow over time.

On this page, I’ve collected our resources on qualitative data visualization. The Qualitative Chart Chooser is also there and you can expect new additions every quarter.

On this page, have a blast with your data using ideas from my Fun and Games Collection.

On this page, you can snag one of my super helpful handouts or download a template.

My blog has been a labor of love and I hope you are able to use the resources I’ve developed over the years and collected here to grow your visualization and communication skills. Let’s keep on going together.💜

Visualizing Not Applicable or Missing Data

Yes, I know the jig is up. All of my examples in books and workshops are pretty tidy, as if every response options was addressed by every single respondent. The truth is that life and data collection are messy. How can we show that different questions have different sample sizes? The most appropriate visualization method will depend on how severe and inconsistent your problem is.

Note Small Consistent Missing Data

The easiest solution when data is acting like a tiny but equal opportunity absentee for every response option is to just make a note of it somewhere in the graph. My preferred location would be in a subheading, underneath the main takeaway point of the visualization. Make it smaller (in a report, like 9 point font) and gray.

If the data is missing consistently it is good to note but not a super critical issue. This treatment marks the issue but relegates its importance to a background matter.

Add Sample Size for Large Consistent Missing Data

Here’s everyone’s favorite nightmare: You sent your reliable research assistants out to collect data with some paper and pencil surveys… only one page of the survey was missing. It’ll be okay because you have more data but what to do with those questions that have many fewer responses? Note the sample size in the data label.

To do this in Excel, you can add the sample size to each data label cell in your table. Make a line break within a cell (on a PC) by holding down the Alt key while you hit Enter. In the formula bar, it appears broken onto two lines.

In the graph, it shows up as part of your data label, under its corresponding question. In this case, the prenatal question was asked on a different page but it was grouped with other healthcare data from the missing page for reporting purposes. The data label makes it clear that the sizes are inconsistent among the questions in this graph.

The same strategy can be used for large amounts of not applicable data. You can delete it from the graph but the remaining data labels need to note the sample size. Again, the strategy here is to note the absence in a diminished way, but still present and not ignored.

Add a Graph on the Side for Large Inconsistent Missing Data

The messiest data to deal with are those cases when lots of respondents skipped questions that weren’t applicable to them. The most accurate way to handle such data is to be super honest that it is gone and show as much in the graph – just off to the side.

To do this in Excel, we are going to combine visualizations: diverging stacked bars and small multiples.

The title here is important because it needs to explain why both graphs are included and necessary for proper interpretation. It’s a complicated concept. If it becomes difficult to encapsulate both graphs in a single title, it’s probably a good sign that you are trying to convey too much information and that perhaps the Not Applicable data should be dropped and mentioned in another way. If you do drop the Not Applicable data, go back to the previous suggestion and be sure to add the sample size to each label.

The overarching point is that you should treat the missing or Not Applicable data differently from the main data you collected. Your primary data needs to be seen in its own light.

This blog post is an excerpt from my book, Effective Data Visualization, which helps you choose the right chart type and then make that chart right inside Excel.