While planning for the session, I asked conference attendees to submit examples from their reports, dashboards, and slideshows that I could makeover as part of the talk.

Later, during the live keynote, I shared a few data visualization principles. Then, as a group, we practiced applying those principles to their real projects.

Here’s one of my favorite submissions:

This conference attendee worked at an organization that placed children into foster care homes. Each year, the organization surveyed their foster care parents to gather their feedback about the experience.

Virtually every organization conducts satisfaction surveys of one kind or another, so even if you don’t work for a foster agency, keep reading!

I’m going to show you the before version followed by five makeovers.

The first makeover didn’t work. The second makeover didn’t work. The third makeover didn’t work. The fourth makeover didn’t work. Just as I was about to give up, I found a winning design with my fifth attempt!

Let’s make fun of my first few attempts together. Then, we can celebrate the fifth attempt together.

I’m going to provide a behind-the-scenes peek into my thought process so that you can apply my thinking to your own projects.

What’s Already Working Well: Length and Context

A couple things were already working well in the before version.

First, I was pleasantly surprised to see that they fit all 22 survey questions and the responses on a single page. They wanted an at-a-glance handout, not a full report. All too often, I witness organizations drone on and on about simple survey results. It’s just a survey. Keep simple things simple, please.

Second, I was pleasantly surprised to see two years’ worth of data included: fiscal year 2016 and fiscal year 2017. All too often, I see annual survey results that only provide the current year’s data. Without historical context, we don’t know whether the current year’s data is any better or worse than previous years. Providing patterns over time is always a good thing.

What’s Not Working: Clutter, Order, Clustered Bars, and Analysis Approach

We’ll declutter the one-pager, obviously. We need to remove the gray background shading. How are the foster agency’s leaders supposed to make decisions based on this data if they can’t even see it?

We’ll also re-order the survey questions. Right now, questions are listed in the order they were asked on the survey: 1, 2, 3, and so on. Presenting survey results in the same order as the survey is rarely the best approach. Instead, we’ll group the individual questions into categories.

Next, I wanted to find an alternative to the clustered bars. Clustered bars are my least favorite chart of all time—even more so than 3D exploding pie charts! Clustered bars aren’t inherently evil. They’re just overused.

Finally, the analysis approach was a bit off. The agency asked foster parents whether they were completely satisfied, very satisfied, satisfied, not very satisfied, or not at all satisfied. They coded a completely satisfied as a 5, a very satisfied as a 4, and so on. Then, they calculated the average score. For example, Staff are courteous and respectful got an average score of 4.5 in 2017.

Although this numeric coding approach is common, it’s not correct. Variables can be nominal (favorite ice cream flavors), ordinal (this satisfaction scale), interval (the scale’s points are equidistant), or ratio (the scale has a true zero, e.g., a height of 0 feet means zero height). If this is the first time you’re learning about nominal, ordinal, interval, and ratio scales, check out this article to learn more.

You can only calculate averages on interval or ratio scales, but the survey has an ordinal scale. In other words, the agency should’ve displayed how many foster parents selected completely satisfied, very satisfied, and so on for each of the questions instead of calculating an average score.

I didn’t have access to the raw dataset while designing this makeover, so I’m going to have to display the average scores here. It’s not the end of the world. But I did cringe during the makeover process. And I’m definitely cringing again during the blogging process.

Makeover 1: Slope Graphs Didn’t Slope…

I experimented with a few makeovers before I settled on a winning design. Here’s the first makeover. Wait! Before you start frowning and rolling your eyes, hear me out. Let’s look at what’s working, and then I’ll be honest with you about what’s not working.

I kept all of the makeovers to a single page, which was a fun challenge. I kept two years’ worth of data. I dramatically decluttered the page, removing the background shading, vertical lines, and horizontal lines from the original. And I grouped the survey questions into categories and then color-coded by category (Staff in turquoise, Clients in purple, Caseworkers in red, and so on). You can learn more about color-coding by category here.

I love most aspects of this redesign.

The major shortcoming of this data visualization makeover is the visualization itself—darn!

We have a few options for comparing two points in time, like these two fiscal years. The most obvious choice is a line chart, which was born with the sole purpose of displaying patterns over time. A slope chart is just the line chart’s cousin; it displays exactly two points in time.

The problem is that the slopes didn’t slope. The visuals were too short to show much of a difference.

Most designs require compromise. I decided that it was more important to limit the makeover to a single page than to increase the height of each slope chart (and, therefore, spill the survey results onto a second page).

Makeover 2: Columns Were Too Short…

Here’s my second attempt.

In this iteration, I visualized the data with column charts.

You can already see the problem, right? The columns were too short to see any differences.

If I didn’t tell you that these were supposed to be column charts, then you might’ve assumed they were just funny-looking squares.

I’m not upset that this makeover didn’t work. I wasn’t rooting for the clustered column approach anyway!

Onwards. I’ve still got a few more ideas up my sleeve…

Makeover 3: The Heat Table Was Too Colorful…

Heat tables are helpful when you’re working with limited space, like my self-imposed rule of limiting myself to a single page, just for fun, ha! The colors live on top of the numbers, not beside them, so they take up less space.

I can generally see that the FY17 column is darker than the FY16 column (ratings were higher in FY17 than in FY16). But I have to work to see the darker colors because I’m distracted by the rainbow in front of me.

If I wasn’t stubbornly devoted to a one-page design, then I could’ve added an empty row between the categories. For example, you’d see the turquoise section for staff. Then, there would be a centimeter of white space. Then, you’d see the purple section for clients. But again, every design is a compromise. I was committed to a one-page design. I couldn’t turn back now!

Makeover 4: Check Boxes Provided Overly Positive News…

As I was critiquing the heat tables, I finally realized that the FY17 results were better than the FY16 results. I hadn’t actually noticed that pattern while looking at the original, at my slopes, or at my columns! Spotting this pattern was a game-changing aha moment.

In this redesign, I opted to focus on big-picture results: that foster parents scored the agency higher in FY17 than in FY16 on every survey item except one. The filled-in squares and empty-squares are easy to scan at a glance. I’ve used square icons in dozens of real-life projects, and I’ve talked about them a few times on the blog before, too. You can read this post to learn how I created them. The filled-in square is a lowercase g in the Webdings font and the empty square is a lowercase c in the Webdings font.

The downside was that the check boxes provided overly positive news.

Can you have too much good news? I think so. Pretend that you’re a leader at the foster care agency. You see this handout at a meeting. You’ve improved from one year to the next on nearly everything! This is great news! There’s nothing to fix! Everything’s working! Except for that one survey question, but that’s just one thing, so who cares! No need to try any harder next year! You’re already doing everything perfectly… or are you?

I didn’t want to encourage complacency. I wanted to provide actionable ideas for improvement.

Makeover 5, the Winning Makeover: Deviation Bars

Finally! The winning makeover! It only took four failed attempts to get this one right…

I loved the simplicity of the check boxes. But I was afraid that they only provided good news. I needed to strike a balance: Keep the makeover simple while providing details about where the agency could do even better.

I created deviation bars to show the size of the difference from one fiscal year to the next. I intentionally re-ordered the survey questions yet again. Within each category, the survey questions are ordered by the magnitude of their improvements.

At a glance, you can still see that all but one survey question improved. But now, you can also see how much or little improvement took place.

It’s good to provide leaders with good news, but it’s better to provide leaders with balanced news.

Now, they can still celebrate all the areas where they improved. Then, it’s time to roll up their sleeves and get to work on improving even more.W

Bonus: Download the Materials

Want to see how I created these five makeovers? They all live inside Excel!

That’s right, I typed the survey questions into Excel, created all of the visuals within Excel, and then PDF’d my screen so that I could share the handouts with leaders.

Purchase the files to learn more and to adapt the templates for your own work.

Learn More

You’ll learn all the step-by-step skills inside Dashboard Design.

It’s intimidating to sit in front of a blank screen and try to start developing your report, infographic, or dashboard from scratch. What data sources will you include? How will you analyze the data? What type of graph will best present your results?

The NHVRC provides hundreds of pages of data about home visiting across the United States. Our team works together behind the scenes to create and finesse each visualization. We want to make sure the numbers are accurate, easy to understand, and visually appealing.

As we tweak the design elements of the 2018 Home Visiting Yearbook—the NHVRC’s third major publication since launching in 2017—we’ve revisited some of our earliest conversations around data visualization. This brief presents our process in five simple steps, along with images of our products at various phases. We hope this inside look can help guide readers looking to share their own home visiting data.

Step 1. Select Your Data Sources

Will you use publicly available datasets? Will you need to collect new data? Which specific variables are available? Sit down with your teammates and select data sources together, just like the NHVRC did during one of our earliest brainstorming sessions. Your process doesn’t have to be fancy. Handwritten notes and whiteboards work just fine.

Step 2. Choose Your Unit of Analysis

Next, think about how you’ll analyze your numbers. For the NHVRC, we decided to provide both national- and state-level analysis of the numbers. The 2017 Home Visiting Yearbook and follow-up Data Supplement) present a map of the United States, so that readers can see where home visiting programs operate across the country. Later, we provide individual state profiles, so that readers can learn about the children and families who benefit from home visiting in particular states and territories. We also decided to provide model-level details in a series of model profiles.

Step 3. Brainstorm Your Page Layout

It’s one thing to design a single graph. It’s another to design an entire page or an entire report! The graphs need a logical, intentional flow from one to the next. Start broad and then go deeper. For the NHVRC State Profiles, we started with the broadest details, like the name of each state and the number of children reached by home visiting programs. Then, we dove into demographic details about those children and their caregivers.

Step 4. Choose the Right Chart for Each Variable

Will you use a pie chart, a bar chart, or something else altogether? Here are a few familiar chart types and some that you may not have encountered before.

Pies and Donuts

Let’s start with the most familiar chart of all, the pie chart. Some data visualization trainers advise against using pie charts at all. That’s probably because so many of us accidentally include too many slices, like this impossible-to-read pie. Here’s what not to do.

Two-slice pies are fine. You can also try an alternative, like a two-slice donut, two-slice waffle, or two-slice icon array. Icon arrays contain miniature shapes, like the miniature coffee cups below.

Bar Charts

Bar charts are another familiar chart type. My advice—if you’re not sure which chart to use, start with a bar chart instead of a pie chart. Bars are often easier to read than pies, especially when you’ve got more than two categories.

The only problem with bar charts is that they tend to be overused. Try an alternative, like a lollipop chart, which places a dot at the end of the bar. Lollipops tend to look cleaner than bars, and they focus viewers’ attention where you want it—on the endpoints.

If you decide to continue using bar charts, consider adding valuable context. Does your organization track your progress toward meeting year-end goals or specific outcomes outlined in grants or contracts? If so, add a target line or use overlapping bars to showcase your achievements.

Line Charts

Need to show how something has unfolded over time? Try a line chart to let your viewers see all the peaks and valleys. Want to focus on the big picture? Try a slope chart, in which viewers focus on the steepness of the slope over time.

When you’re using line charts, beware of the spaghetti line! Spaghetti lines crisscross and intersect—just like a pile of tangled spaghetti—making them near-impossible to decipher. If you see a spaghetti line chart in your project, don’t worry. You can try making it easier to read by highlighting a single line of interest in a darker color or testing a small multiples layout in which you build multiple small charts, one per line.

Step 5. Draft Your Visualizations with Sticky Notes

You may be tempted to sit down at your computer and begin creating charts right away. I’ve found it helpful to create drafts with good old paper and pencil before creating anything on the computer. I also like using sticky notes as I’m drafting because they allow me to move graphs around the page and get the placement just right. Your sticky notes might contain some of the pies charts, bar charts, or line charts shown previously.

While drafting the NHVRC State Profiles, we drew one graph covering one variable per sticky note. Then, we arranged and rearranged the sticky notes. As a team, we talked about which variables should come first, second, and third on each page. When we reached consensus, we started building the graphs on the computer. Drafting on paper saved us a lot of time and helped us develop a final product that reflected everyone’s input.

We hope this behind-the-scenes peek into our data visualization process and early images can help you during your own drafting process. Click through the NHVRC website to see our results so far and feel free to send us your data masterpieces!

Before

This attendee worked at a large university library and was responsible for writing an annual report about the library’s operations and accomplishments. The report was full of tables, graphs, and photographs.

This is a screenshot from the beginning of the report that offered background information about the library, like how many visitors came to the library each hour and each day. Library headcounts inform decisions about staffing and about future library hours. For example, if they find that library attendance peaks in the morning, then the university might decide to open the library earlier to accommodate their visitors’ preferences.

After: Reenvisioning the Individual Graphs

The first graph was about average hourly headcounts—how many students, staff, and other visitors were present in the library at any given hour in the day.

In this makeover, we:

• Decluttered the graph by removing the border, tick marks, and horizontal x-axis line;
• Decluttered the vertical y-axis by removing all of the labels except for the smallest value (0 visitors) and largest value (50 visitors);
• Decluttered the horizontal x-axis by only labeling a handful of key hours;
• Nudged the columns closer together (here’s a tutorial on shrinking the gap width) so that viewers could see the smoothed-out shape of the graph rather than the individual columns;
• Applied a mix of darker and lighter colors; and
• Wrote an annotation above the graph to highlight the key finding. This takeaway—that the library’s peak hours are between 10am and 7pm—was hiding in the paragraph above the graph. I wanted to make the report skimmable.

Here’s the second graph, which is about average daily headcounts. I decluttered the graph and then brought their key message into center stage with dark colors and an annotation.

After: Re-envisioning the Page as a Whole

Next, we had to think creatively about the page as a whole. How would we arrange the graphs on the page? How big or small should we make each graph? How would we re-write the existing paragraphs?

While editing the page as a whole, we:

• Adjusted the report’s text hierarchy by making the Heading 1 and Heading 2 text large, bold, and dark.
• Re-wrote the introductory paragraph and moved some of those sentences closer to their respective graphs. The sentences about daily headcounts belong next to the graph about daily headcounts.
• Wrote graph titles. I usually advocate for storytelling titles that explicitly state the graph’s desired takeaway message. But in this makeover, I decided that annotations—Peak Days; Monday through Thursday—would be just as powerful. I didn’t want the graph’s storytelling title to be redundant with the graph’s annotation, so I combined a generic title with a storytelling annotation.
• Added icons because icons can make our graphs more memorable; and
• Paid careful attention to alignment. The words are left-aligned. The graphs are aligned with each other, too. You could draw a single vertical line from the top graph’s y-axis down to the bottom graph’s y-axis. It took a few minutes to get the spacing just right, but alignment is always worth the extra time because it makes the finished product look more professional and purposeful.

Here’s the full before/after data visualization makeover:

Bonus

Would you like to explore the Excel file, Word document, and PowerPoint slides that I used to create this makeover? Purchase the materials and use them as inspiration for your own projects.

Purchase the Templates ($5)

What makes a good graph great? Let’s answer that question by taking a walk down memory lane.

Earlier in my career, I was a researcher for a large nonprofit in Washington, DC. We gathered for all-staff meetings every few months, all 100 of us standing like sardines inside our largest conference room. In those meetings, we needed to know more about the teenagers and families we were working with in the community and whether our afterschool programs were effective.

When it was my turn to report on the data, I would stand in front of my fellow sardines and show a slide like this on our tiny, grainy projector screen.

Staff squinted, straining their eyes to read small, diagonal text. I mumbled through a few talking points, like how we worked with a lot of Latino teenagers. They clapped, sort of, when I was finished, and the next person on the agenda began her presentation.

Nowadays, I know better. Here’s how I would revamp that slide if I were to design it today.

• First, I would remove a lot of unnecessary ink. I would delete the logo and the redundant sentences and remove the gray background shading. I would also remove the 3D effect, which was distorting the data—making the columns look taller or smaller than they really were.
• Next, I would more intentionally use text to help tell a story. As an organization, we spent a lot of time serving Latino and African American youth in the surrounding neighborhoods. In the original column chart, the names of the ethnicities were too close together to be legible to staff in the back of the room. I would rotate the vertical column chart into a horizontal bar chart, then reorder the categories so that the Latino and African American youth would be listed first.
• Finally, I would apply strategic branding and design choices. This includes incorporating the organization’s fonts and color palette and using dark colors and an annotation to help key numbers pop.

What a difference!

I apply these techniques to all of my reports, slide decks, dashboards, and infographics—including a series of state profiles I recently partnered with the National Home Visiting Resource Center for the Data Supplement to the 2017 Home Visiting Yearbook. Here’s a behind-the-scenes peek at how an early draft received the same three-step treatment.

Remove a Lot of Unnecessary Ink

At a bare minimum, we needed to remove unnecessary design elements. In the example below, we removed the graph’s border, generic title, and vertical and horizontal lines. We also adjusted the wording and the colors.

Intentionally Use Text to Help Tell a Story

As data visualization becomes more popular, I often see people focus so much attention on creating good images that they forget about the text. Both can work together to tell your story. Here’s how we translated the bullet points from our spreadsheets into full sentences.

Apply Strategic Branding and Design Choices

We started with the NHVRC color palette. You might recognize these colors from the NHRVC web site. The “before” version of an NHVRC state profile is displayed below on the left. It’s colorful, sure, but the colors are primarily used to decorate the page. In the “after” version (displayed below on the right), we intentionally color-coded items by category. Information about children has turquoise text and graphs, while information about families uses purple text and graphs. We wanted readers to spot the two colors—and therefore distinguish between the two buckets of data—instantly.

We also used dark colors intentionally within sections. Do you see the turquoise bar chart depicting age ranges? Light turquoise corresponds to less age (i.e., younger children) while dark turquoise corresponds to more (i.e., older children). We applied darker colors intentionally in the waffle chart at the bottom of the page, too.

In the visual perception world, we call dark color a preattentive attribute. Our brains are wired to notice dark colors without any conscious thought. We can instantly see the difference between dark and light colors without even having to think about it!

These three steps can go a long way helping translate your data into compelling “data viz.” Visit the Data Supplement’s state profiles section to see these principles in action.

Like most in-house research teams, this team had a lot of data. The researchers had spent time on the ground in each country where their programming was offered. They collected data from children, parents, and program staff in both the treatment and control groups. Then, they did the math to see where the treatment group had outperformed the control group.

Our 30-page report for their leadership began with a short intro about the study. Then, we launched into the fun stuff, the results! We needed to provide an overview of where the program had been effective (where the treatment group had significantly better outcomes than the control group).

Before

Here’s what the original one-page overview of the study’s results looked like. I made up fake values and hid the names of the variables, but you get the idea. There were a lot of p-values and decimal places.

I’m not going to define p-values. Heck, not even scientists can easily explain p-values! Basically, lower values are better.

After, the Bare Minimum

At the very least, we need to declutter the table. We removed the outer border. We removed the vertical lines. Notice how your eyes can still read down each column without the lines. We kept the horizontal lines, but we changed the black ink to light gray ink. We need the results to stand out, and the results can’t stand out if they’re hidden by unnecessary lines.

And at the very least, we need to apply color strategically. You should use your own organization’s colors so that your table (and the rest of your publication) will reinforce your brand. Throughout the report, we talked about the differences between the three countries, so we color-coded by country. In our tables, charts, and maps, Country A was always blue, Country B was always purple, and Country C was always turquoise. I’ve got another example of color-coding by category here.

After, the Winning Makeover

Here’s what our final makeover looked like. We decided to focus on the big-picture findings. So, we used empty squares to represent variables that weren’t statistically significant and filled-in squares to represent variables that were. We used p<.05 as our cutoff here; anything with .05 or lower got filled in and anything above .05 remained empty.

This was the first page of our Results chapter and we wanted readers to know what happened at a glance. Apply the Squint Test to both versions. Go ahead. Squint your eyes so that you’re peering at each table through your eyelashes. The asterisks from the original version are supposed to grab our attention and help us decode the information, but they end up looking so cluttered compared to the squares. Our brains can scan a page of squares faster than they can read a list of numbers and asterisks. Data visualization for the win!

And remember our audience: the organization’s internal leaders who were not researchers by training. These leaders had expertise in running organizations and launching programs around the world. They might’ve taken a couple research methods or statistics courses, but that would’ve been decades ago, and they don’t need to use those skills on a regular basis. They simply needed to know whether their treatment group had done better than their control group so that they could decide whether to expand, shrink, or adjust their approach to programming. They didn’t need the exact numbers in order to make those decisions. And I worry that providing the exact numbers can actually distract busy leaders from the big picture. The exact numbers go in the appendix of the report. The big picture findings go in the body of the report.

How We Created the Squares

Want to know our secret to producing the squares? We saved a million hours of time with Webdings. Yes, that crazy-looking font that you discovered back in 1999 and haven’t used since! In Webdings, a lowercase g gives you a filled-in square and a lowercase c gives you an empty square. I’ve used Webdings icons in a few dashboard projects and the results are easy to follow every time.

It was important to use existing icons within Microsoft Word. This was an internal report, and we didn’t want to slow down the writing process for the research team by making someone else reformat everything in a graphic design software program.

I made a list of all the shapes that are readily available to you through Webdings, Wingdings 1, Wingdings 2, and Wingdings 3 fonts. I highlighted a few of the most promising icons in yellow–but my favorite is the squares. There’s a link at the bottom of this blog post that allows you to download my spreadsheet for free.

The Runners-Up That We Considered and Tossed Aside

We experimented with a few drafts.

Circles: Too Time-Consuming to Create

Circles looked great, but they were a pain to create. We had to create individual circles using Insert –> Shape and make sure they were perfectly aligned within the table. Even making the fake version of the table for this blog post took me at least 15 minutes to get right. In comparison, creating the squares takes less than a minute because you’re just typing g’s and c’s and then changing the font to Webdings.

Check Marks: Too Hard to Scan at a Glance

Check marks and x’s looked too cluttered. In other words, they took longer to scan at a glance than the squares. There were too many diagonal marks all over the page.

And they were so small. We could’ve enlarged the font size of the check marks and x’s to make them more legible. But then we would’ve needed to enlarge the row heights in the table, too. The real version of this table included a couple dozen variables and it filled the entire page of the report. We didn’t want to enlarge the font size, and therefore enlarge the row heights, and therefore have the table spill onto the next page. The goal was to provide the exact right amount of information to the internal leaders. There’s something psychologically stressful about tables that last for more than a full page.

Thumbs Up and Thumbs Down Symbols: The Wrong Vibe

Finally, we experimented with thumbs up and thumbs down symbols. The non-researchers in the room loved the simplicity of this approach. But the researchers and I cringed because it gave off the wrong vibe. The thumbs down represents negativity. We didn’t want to say that something was bad just because it wasn’t statistically significant. The empty square represents an opportunity–an opportunity to do better and fill in the square.

The thumbs up and thumbs down symbols were tiny, too. Can you even see them here? I can’t, and I created the symbols in font size 18! We would’ve needed to enlarge the font size… and therefore enlarge the row height…. and therefore spill the table into a second page.

How are you transforming p-value tables for your non-technical audiences? Do you think any of these approaches would work for you? Leave a comment and let me know.

Bonus

Download the symbols.

Symbols Available with Webdings and Wingdings Fonts (Free Download)

Updates

Bogdan Miku (@trizniak) explored how we might visualize confidence intervals. Read his post: http://trizniak.blogspot.be/2018/02/confidence-intervals-for-effect-size_21.html

Dana Wanzer (@danawanzer) explored how we might visualize effect sizes. Read her post: http://danawanzer.com/visualizing-statistical-significance-and-effect-sizes/

For the first time, TechnoServe’s beneficiaries, partners, donors, and prospective donors can view key stats about TechnoServe’s projects. TechnoServe is offering their data to the public, not because they have to, but because they want to. What’s not to admire about an organization that’s committed to transparency and information-sharing?

The portal includes a map of the countries and regions where TechnoServe works…

…and results by country and by project.

There are three features that make TechnoServe’s Results Portal easy to navigate:

• Consistent color-coding and icons by category;
• Consistent navigation; and a
• Consistent text hierarchy.

Consistent Color-Coding and Icons by Category

I like color-coding by category within a single document. I love color-coding across multiple documents and projects.

Remember TechnoServe’s 2016 Impact Report, which I showed you last time? They reported on three key variables: financial benefits, beneficiaries, and finance mobilized. Throughout the entire 16-page report, information about financial benefits was always displayed in green with the icon of the hand holding paper currency. Information about beneficiaries was always displayed in purple with an icon of several farmers standing together. And information about finance mobilized was always displayed in turquoise with the line graph icon.

We repeated the same colors and icons throughout the Results Portal, too. Can you spot the icons, terms, and definitions in the opening screen of the Results Portal? We added a fourth variable to this project, too. TechnoServe wanted to emphasize how many beneficiaries were female, so information about percent women was displayed in orange.

2016 Impact Report (left) and Results Portal (right)

We also worked on a third product, a series of Google Sheets dashboards that were designed for TechnoServe’s internal audiences. Those dashboards aren’t public. But yes, you guessed it, the internal dashboards follow the same color-coding. Financial benefits are green, beneficiaries are purple, percent women is orange, and finance mobilized is turquoise.

Consistent Navigation

Tableau allows you to insert drill-down menus just about anywhere: in the upper left, upper right, lower left, lower right, or middle of the page.

In earlier drafts, our lists of countries and projects that viewers could explore were in different places on different screens. We placed the lists wherever we found blank space and could squeeze them in.

In the final version, we intentionally placed the lists in the upper left corners. In Western cultures, we read beginning in the upper left corner and then read across and down in a z-shaped pattern. That’s why we placed the country and project lists in the upper left corner—because it’s the most valuable real estate on the page.

Our goal was to make navigation seamless. We wanted viewers to focus on interpreting the data, not on interpreting the dashboard.

Consistent Text Hierarchy

Do you really need font in size 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18?

Sometimes, software programs give you too many different font sizes. You insert a chart and there are different sizes for the title, subtitle, axis labels, numeric labels, and category labels.

Other times, dashboard designers create too many different font sizes. We’re not sure what our final product will look like. We experiment. We try a graph title in size 12 on one screen and a graph title in size 13 on another screen. We try a body font in black on one screen and dark gray on another screen. We try a page title in bold on one screen and italic on another screen.

In the final version of the Results Portal, we paid careful attention to fonts, sizes, colors, and styles. We built a consistent text hierarchy. A text hierarchy tells your viewers which information is at the top of the food chain. The most important information is largest, the information that’s of medium importance is a medium size, and the regular ol’ body font is the smallest size.

Consistent text hierarchies across screens make you look polished and professional. More importantly, text hierarchies make your viewers’ job of interpreting the information faster and easier.

Has your organization built a public-facing dashboard like TechnoServe yet? Link to your websites here!