Got Data, Now What? Storytelling Through Accessible Design

So... First of all, thank you all for being here. I really appreciate your time tonight. Excited about tonight's topic, which is called: Got data -- now what?

So the long and short of tonight's topic is: You've gathered data. You've gathered metrics. You're ready to figure out how to streamline processes based on those numbers. But you're not really sure what to do with the numbers at the moment.

So we're gonna be talking all about that tonight. How to streamline your data. And how to tell a story with your data. And one that other people will find compelling and interesting, and want to hear more about.

So first, about me. I am a queer disabled nonbinary accessibility strategist. My title is advisor, digital solutions, accessibility training and awareness. It's a mouthful.

Basically what it means is that I am leading training and awareness at Navy Federal Credit Union, where I am a contractor. And in that role, I serve on the Enterprise Technical Accessibility Program Team, where we are amplifying accessibility across the organization, ensuring that our members, all of them, every one, no matter what their needs are, or how they access technology, can access their banking needs appropriately, through Navy Federal Credit Union. I am an advocate, a speaker, and an author.

I'm a published co-author of Inclusive Design for Accessibility. The book just came out at the very beginning of August. We're very excited about this. There were 12 of us that came together to write this book.

And it's all about inclusive design and accessibility, and I'll be showing you the book at the end. A great resource. A little bit biased -- chapter 2 is particularly fantastic. But definitely check it out.

And then my passion really is shifting accessibility left and making digital spaces inclusive. And then, if you don't know me on LinkedIn, you can follow me on LinkedIn. I also have my hashtag, #A11yWithAngela, where I usually tag any of my accessibility-related content. So if you're like me and you want to go down a rabbit hole at 2:00 am, that would be the hashtag you want to follow.

So let's talk about what we're gonna talk about tonight. We're gonna start looking at data without barriers. What does that mean? We're also gonna look at: Who are we excluding, and why does that matter?

Then we'll be designing access from the start in our storytelling, and we'll be turning numbers into narratives. And then at the end, I'm gonna be giving you some tools you can use tomorrow. Super quick, no ramp up time, things that you can use immediately to support you on your data gathering journey. And then some key takeaways and actions, and then we'll have time for Q and A.

So let's talk first about what data without barriers would look like. So when we think about data visualization, we're thinking about how the practice of turning those numbers and datasets into visuals that communicate patterns, trends, and meaning are something that we can all do, no matter where we're housed within our organization. So charts, dashboards, infographics -- they all help transform complex concepts into information that we can act on. And when done well, data visualization simplifies complexity, and it amplifies clarity.

So conversely, when you do data poorly, or without accessibility, it obscures meaning and it excludes people from decision making. And that's the last thing you want, especially at your organization where really everyone has valuable feedback they can give you. So why does accessibility in data matter? Well, data drives decisions.

Right? The numbers don't lie. So ultimately, when you gather numbers, it's really gonna tell you where your processes are functioning efficiently, and where you might need to make some tweaks to make things better. So when accessibility is missing, decision making is shaped by only a portion of the available audience.

And at your organization, ideally, the audience is everybody. Right? You don't want to leave anybody out. So exclusion in data visualization reinforces inequality, leaving people unable to contribute the insights or to actually benefit from those outcomes.

So when everyone can access the data, this really establishes equity, trust, and reliance in your brand and your organization. And then accessible data ensures that no one holds or wields all of the power. And this is very important, because if your data isn't accessible, then what you're unconsciously -- I hope -- saying is: Okay. You deserve to make these decisions.

And you don't.

So who gets left behind? When accessibility is ignored, blind and low vision users are locked out of data stories entirely. I had a project recently where our data dashboard was conveyed through a flattened image. All of a sudden, everybody is stressing out, because we had someone who has low vision, who could not access that flattened image.

You know, and the team responsible just didn't have the skillset to do this. So we did some quick ramp-up education. Taught them something new. And we also fixed it for them.

Which they were very helpful about. So not only is it a good opportunity to do somebody a favor at your organization, but it's also a nice teaching moment. And you know, really, here, what we prioritize is empathetic education. We're not coming after anybody.

We're not telling anybody that they're doing something wrong. We're just showing them a way to do it better. So cognitive overload occurs when dashboards are cluttered with too much data. And this also could mean that colorblind users might miss meaning when visuals are relying only on certain gradient color scales such as red or green, which you'll see in a little bit later.

Assistive technology users often find dashboards unusable, because the filters, the charts, and the interactive elements are usually unlabeled or they're hover-dependent. Which is a nightmare for anybody trying to use assistive technology. Now, if I could tell you the number of people who called me out on this next line, when I sent this presentation in, Meryl and Ben -- I was like -- okay. This is intentional.

I promise. But right away... My users with vision noticed it. Right?

So each barrier narrows the diversity of perspectives in critical conversations. Now, on this slide, that particular print is written in a gray that is only slightly offset from the background gray. And I did that on purpose. Because what I'm trying to show you is that when you have one small element that isn't accessible, all of a sudden you're losing information.

So what if that last piece was, like, your conclusion to your whole dataset, right? Here's the point of our dataset. And it's this one bullet point that nobody can read. Right?

So that was kind of the point. So when I came over, and we were looking at my presentation, Ben said: So this last line -- I'm like -- it's intentional. I got it. It's good.

I know it's there. I promise. And then what I usually tell people is: If it doesn't quite make sense to you, or if you're sitting and you're squinting, looking at your dataset, then something is probably wrong. It's probably not accessible.

So... It's not... This is not meant as an insult. It's not rocket science to make something accessible.

Even if you look at something and it gives you just the slightest bit of pause, and you're not sure about something, okay. It's probably not accessible. Let's think about how we can do this differently.

So okay. Now we know we can't leave anybody behind. Now we know we will include everybody. But how will we design access right from the start?

So the very first thing we do is we ensure that our charts utilize a good color contrast. So color can never be the only way of conveying information. The Web Content Accessibility Guidelines, or WCAG, require a minimum of 4.5 to 1 contrast ratio for body text and visuals. Now, what I used to tell people is: If you look at the colors and they don't look right, they're probably not passing color contrast.

However, when I was making this presentation, I caught myself using a green that was just a hair shy of the color contrast ratio. That would have passed. So even myself, I was like... Wait.

I have to actually check my color contrast. Now, I have hex codes memorized that are my go-tos in a pinch. But if I'm looking at colors on a screen and I need to check the color contrast quickly, that's definitely something I do before I present data to my team. Testing tools such as WebAIM's contrast checker make it easy to validate contrasts before publishing.

And again, since you don't want color to be the only way that you're conveying meaning, you really want to ensure that you've got those patterns or textures or some direct labels, alongside the color cues, to support your visualization.

So next, let's talk about labels and legends. Data points without clear labels create confusion and add to that inaccessibility. So legends need to be close to the chart they describe, not hidden across the dashboard. I don't know about you.

I love a nice dashboard. But if it's too crowded and it's not accessible, not everyone is going to love that dashboard. So sometimes it does come down to personal preference, how you're able to acquire that information. So some of us can sit here and spit out SPSS charts and have no problem.

But other people might need something really clear and streamlined, and that's why it's imperative to ensure that your dashboards are accessible to everybody. So you want to ensure that you're not using vague axes like series 1 or value. And you want to actually replace them with descriptive names, so that the person can readily understand what that axis is showing. And when crafting charts, consider: If colors and visual cues totally disappeared, could the data still make sense?

And if it couldn't, then we need to think about how we're going to leverage other elements aside from colors or visual cues to ensure that that data makes sense.

So here are two charts. We have two sales by quarter charts. One is a bad example and one is a good example. So the bad example is bad because it says: Units sold.

And there are bars showing the different numbers of units sold. However, the bars are in red and green. Now, if we're looking at this in gray scale, then we might still have trouble reading where the lines stop, because there are no borders around the lines. And someone with red-green colorblindness totally would not be able to access this chart, because it is just relying on red and green.

So, for example, the colors here could represent different teams, or they could represent two different products. So if you're conveying that information through color, then people on your team might not realize to what team the chart is ascribed or to what team -- what product the bar is signifying. So instead, if you utilize textures, so in the series by quarter good example, the units sold are delineated not only by a blue and orange bar, but within those bars are four different textures.

And the reason why there are four different textures is because it's showing each quarter of sales for different quarters. So that's why those textures are different.

So next let's move on to fonts and readability. Readable text is foundational to accessible visualization. Now, I'm gonna date myself here. But in the '90s, when I had a cell phone, and you could actually change it to one of four different fonts, and one font was like... Little kid handwriting, and everybody thought that was the coolest font ever, right?

But it's not accessible at all. So sans serif fonts are a good practice. At a minimum of 12 to 14 points, within charts. If you can make it even bigger, that's fantastic.

In case you're wondering, sans serif means it doesn't have the little pretty dangly things on the end of each letter. Times New Roman, for example, is a serif font. You also want to avoid all caps. Just... You know... Like we feel when someone texts us in all caps, it feels like someone is yelling at you.

Also, it is a little hard to read. But I know we all have that one relative that only posts on our Facebook page in all caps and types in all caps. Right? And usually we really know how old that relative is, if they sign their name at the end of their post, right?

At the end of their comment. Not their post, but their comment. So... Axis labels and legends must be large, consistent, and clear for everyone to scan quickly.

Especially if you're giving these data charts to executives or management within your company. Okay? We all know how little time they have. So we want to make sure that they can acquire the information in our datasets as quickly as possible.

So next is simplifying the numbers. Nobody wants to see the number π represented in a data chart. Right? So overloading the charts with decimals or raw values or endless metrics overwhelms the user.

This is where storytelling is gonna be your superpower. So if you prioritize what matters most, which is those key trends, not every data point, that's where you're gonna start to formulate this really compelling story for your data. So you can summarize findings into callouts that emphasize the meaning behind your data. So, for example, revenue grew 17% is way more impactful, concise, and powerful than 27 lines of numbers that illustrate the same thing.

And then simplification reduces cognitive burden, and then also increases clarity for everyone. I'm sure we all know someone who loves to call a meeting on a Friday at 4:30. So Friday at 4:30 is not the time to present 27 lines of data. Now, what you can do, if you really do feel that your raw data tells the story, is you can offer something like: If you would like a deeper dive into this, I would be happy to show you my raw input and where this came from.

I apologize for the text being so close to the bullets. I've already moved it several times, and it won't move. So I apologize. Layout and white space.

Which obviously I don't have here. This was on purpose for this slide! Totally intentional! So... Clutter is one of the most common accessibility barriers in data visualization.

Dashboards with dozens of visuals demand more cognitive energy than most users can give. Organize charts into logical groupings that tell a story. So again, this is where you're prioritizing those key trends over time, or grouping your data into contexts that make sense. And whitespace is not wasted space.

It's breathing room that enhances comprehension. Please do not feel like your dashboard has to look like it came from a spaceship. Okay? You want some whitespace to allow your reader's eyes to be able to understand each area of data and where one starts and one finishes.

So let's think about alternative formats. No visualization is complete without an accessible alternative. So data tables must be available for every chart or dashboard. And again, this is where you can offer to go into a deeper dive, if someone is interested.

If you have someone on your team that loves to dig through the data, then send it as an attachment, and just leave it there. If they want it, they'll use it. And if they don't, that's totally fine too. Accessible exports in CSV, Excel, or HTML give users control of how they consume the data.

So when I'm presenting data to my team, I'm giving it to them in several different formats, so that they can then choose the format that they prefer. Avoid PDF-only reports. I usually say this in general. Don't do a PDF.

The caveat on that is: If you know how to appropriately tag a PDF for accessibility, then it's fine. However, that's also a whole other extra step. So you might as well just avoid PDF data reports. Often, this style of report is inaccessible to assistive technology, unless it's reviewed and approved by an accessible document specialist.

Shoutout to IAAP for offering that certification. And also for offering the CAECs for tonight's event.

Here's another example of a great pie chart and a not so great pie chart. So... Here you can see that we have market share that is depicting almost an even split. However, as you can see, one of the areas of the market share take a little bit more percentage than the rest of them.

However, if you have low vision, you might not be able to tell that the gray scale circle actually has different shades of gray at all. Nor can you tell what the exact percentages are, nor can you tell what each slice is depicting. Which region each slice is depicting. So instead, use colors that are really different from one another.

So that you can readily see this. But remember, we don't always convey information with color. So we also have a label, a helpful label, that says: You know, in the North region, we're experiencing 30% market share. In the East, it's 30%.

And the South has a little more of the pie with 40%. Okay. So now we've talked about numbers. And it's time to transform those numbers into a really compelling story that gets your organization excited about taking action.

So why does storytelling matter? Humans remember stories, not raw numbers. And accessibility ensures that everyone remembers your story. So storytelling also helps align teams.

This chart shows a 20% drop in engagement, and here's why that matters. So that's going to get your team energized and enthusiastic about: Where do we go from here? Where do we solution? To identify the issues for this problem.

And how can we move forward and how can we make it not a problem anymore? Inclusive storytelling avoids leaving behind those who can't see or process visuals the same way. So again, it's almost like multimodal learning in education. So not only did you show someone a data chart and did you give them numbers, but now you're transforming that into a story.

So for people who really prioritize narratives and really like that way of learning, and putting those pieces together, it's gonna help support those people as well. So accessible narrative moves beyond equity to prioritize efficiency, because it means enabling more people to understand your information faster, which really is the key.

So there are several layers of accessibility storytelling. And there's a visual layer, such as colors, labels, structures. As long as it's perceivable. There's a language layer.

Captions, summaries, alt text. Which of course must be understandable. And then there's an interaction layer. So controls, filters, and exports that are operable.

Now, if you're picking up what I'm putting down, you'll notice that we've hit perceivable, understandable, and operable within the narrative for WCAG. So when all layers align, data transforms into an inclusive narrative.

So let's talk about some practices for accessible storytelling. Write captions that state the main point of the chart. I used to think that I had to describe the entire chart in detail, as if someone couldn't see it. But really what a caption should do is drive the point of that chart home.

What is it that you're trying to demonstrate by including that chart? And then provide plain language summaries alongside those complex dashboards. Use parallel structures. So same words, same sequence, for consistency across multiple visuals.

We remember this as children, when we learned how to read. And we remember this possibly with either helping our own children or nieces, nephews, and cousins and our family. When we're teaching children how to read, we start with what's called predictable text. So books that utilize the same phrases over and over and over, with maybe just a word at the end being the only variable.

We're doing the same thing here with data, because it makes it a lot more understandable a lot more quickly. Thinking about reader burden. So how many steps does someone have to take to acquire meaning? And we don't want it to be that many steps.

We want them to be able to acquire meaning as quickly as possible.

So now it's time to build our narrative arc. We want to set the stage. Why does this dataset matter? And what context frames it?

This is a great cheat sheet for how to create a narrative arc through your data storytelling. So first you set the stage. Why does it matter? What's the context around this?

Why do we care about this? The second area is when you're going to show the shift. So what changed and what patterns emerged? And then thirdly, you're gonna have a call to action.

What decision should we be making after this? Or what action needs to follow after this? And be sure you come with some solutions in mind. Accessibility ensures that each step is available to every audience.

Not just those who primarily acquire information through vision. So we are all different types of learners. Some of us are visual learners. Some of us are auditory learners.

Some of us learn by doing. Okay? So when you're finding different ways to introduce those modalities through your data storytelling, that's where you're gonna make the greatest impact.

So this is a more visual way to see that narrative arc. So, for example, I might start my story by demonstrating that for quarter 1, we only had an engagement rate of about 45%. But we shifted our engagement by doing the following things. And then I would explain what it was that we did, and how that allowed us to have a shift within our organization.

And then after I demonstrate how we shifted our thought process, or our processes or our... Whatever it was that we did, to make things even better, to really increase that engagement, that's where I kind of give the call to action. So now that our engagement is at a really impressive 87%, what are we going to do from here? How are we going to fine tune what we've been doing well, and what are the next steps going to be?

So now I haven't told you the numbers. Right? I've just told you a story. But you're gonna go home later and you're gonna remember that we more than doubled our engagement through the story that I told you.

This is the power of narrative storytelling through data.

The next thing you want to try to rely on when you're telling your story is that accessibility is fundamentally about empathy that is integrated into process. Okay? This is where you're making people care. This is where you're giving them something that they can take intrinsically to think about why this would mean something to that person.

Okay? So before publishing your data, ask yourself: Would I still get this story if I couldn't see the chart? So even if I couldn't see those charts at all, would I still understand the story that's being told to me? And even if you couldn't see that engagement chart, if I told you that we doubled our engagement through X, Y, Z initiatives, you would still understand the story even without seeing the chart.

And that's the goal. So then you could ask yourself: Would I still understand this, even if I couldn't process the numbers quickly? I didn't ask you to process 40, 61, 87. I told you we more than doubled our engagement.

No numbers there. Right? There's no numbers. So even if understanding numbers and percentages and things like that isn't your strong suit, by telling that story, that's where you're gonna grasp the whole point of what it is that we're trying to achieve.

So empathy transforms accessibility from something we have to do, because it's the law, to something that we want to do, because we've been utilizing more compassionate communication. So there are some pitfalls with this whole process. Right? So without accessibility, sometimes you're going to have a lot of what's called visual noise.

Too many compelling elements hiding the real story. So just as I'm sure all of us have been in a really noisy restaurant, where we can't even hear the person in front of us, or understand what they're saying, because there's just too much noise, that happens visually as well. So if anyone has ever been on a website where you're squinting and trying to read the website and all of a sudden there's all these different elements, the one example I can give is online recipes.

Has anyone ever tried to look up an online recipe? And then it's like playing whack-a-mole with the 15 different videos and pop-up ads and whatever, and you're looking for the "jump to recipe" button. Where is it? Where is it?

I just want the recipe. That is a perfect example of what visual noise is. Jargon and acronyms create silos instead of shared understanding. If I stand up here and I'm talking at you and it's a whole bunch of alphabet soup and you're not familiar with the industry terms, because maybe we're working with different organizations, you know, and maybe Navy Federal and ArcTouch are trying to collaborate on something, and all of a sudden Navy Federal is using all of the military jargon that we use at Navy Federal, but ArcTouch is gonna have no idea what we're talking about.

So when you use jargon, never assume that people understand the terminology that you're centered in, within your organization. Just call it what it is. That's why I never call WCAG WCAG when I start talking. Even if it's to a group of accessibility professionals.

I always say it first. Because you never know who is in your audience. Reliance on a single format excludes others. So if you have data charts and dashboards that are only interactive, someone who does not have the ability to follow continuously interactive elements might really struggle with that.

So you want to offer both static and interactive options. If you have the chance to do so. And then accessible storytelling really does reduce friction instead of adding it. So even though my chart didn't tell you what we were trying to double the engagement for, I was still able to tell you that story and convey that narrative to you, regardless of what we were actually talking about.

So that's where you want to ensure that you're creating those areas and pathways of understanding as quickly as possible.

So if you're interested in how to design for accessibility anything from data metrics, program management, you know, so many different things within design, I really encourage you to check out this book. Like I said, chapter 2 is really particularly fantastic. I'm kidding. The whole book is phenomenal.

This was such a great opportunity to be able to co-author this publication, at the invitation of Dale and Denis and all of my co-authors working with us all together -- was just phenomenal. It is a practical guide to digital accessibility, UX, and inclusive web and app design. And it's available on Bookshop.org, or wherever you prefer to buy your books.

So... Let's talk some tools. You're ready to start tomorrow morning with creating data storytelling, and you're ready to tell a story about something going on at your organization right now. How can you do that tomorrow?

The first thing you can do is check out Power BI, which is Microsoft's flagship BI tool, with built-in accessibility features, including keyboard navigation, screen reader support, and alt text for visuals. You can also check out Tableau. It's a powerful visualization tool. The accessibility features are improving, but it does require disciplined design for compliance.

Excel. Look -- it can be boring. Right? A lot.

A lot boring. But it's really overlooked. And it is unfortunately for how many people... How many times people are looking for something really flashy and fun... But it really is one of the most screen reader-friendly tools that you can have.

It also lets you build accessible charts and tables with the accessibility kind of baked right in, which is nice. And then there's an app called Flourish, which is a web-based visualization platform that supports alt text and responsive embeds for inclusive online charts. Color Oracle is a free colorblindness simulator for checking designs against multiple vision types. Now, I would like to pause and add a caveat here.

Disability simulation is not something that is encouraged by the disabled community. However, what I like about this particular tool is that if you are not really familiar with colorblindness, or how it can create barriers in understanding information, it really does do a nice job of showing you exactly what your specific presentation would look like to somebody with the different types of colorblindness. So it's really unique in that you can customize it like that.

And then Stark is an accessibility plugin for Figma, Sketch, and XD, for my design people. It checks color contrast and generates alt text guidance. If you're sitting there and you're writing alt text for 20 images, sometimes it can be difficult to come up with something creative. Stark does a really nice job of taking that pressure off.

Some resources. There's a great book. Storytelling with Data by Cole Nussbaumer Knaflic. It's a classic guide to clarity, simplicity, and storytelling with charts.

There's another -- this is a passion/side note/rabbit hole that I've gone down, but I really like. Data Feminism. It examines power, equity, and inclusion in how we use and present data. Accessibility Guidelines for Data Visualization in WCAG 2.2.

Obviously go right to the source. How do we ensure that data is accessible? WCAG is really your bible for that. Nightingale is by the Data Visualization Society.

It's the online journal of the Data Visualization Society, and it has many accessibility oriented articles. Really, really great read. Some people for you to connect with on LinkedIn, aside from myself. Tamara Munzner.

She's the author of Visualization Analysis and Design and thought leader in viz theory. She's phenomenal. Really a great resource. Andy Kirk is a visualization consultant, trainer, and author of Data Visualization: A Handbook for Data Driven Design.

It's another phenomenal print resource. I think it is also available on ebook. And Nadieh Bremer is an award winning data visualization designer who shares really cool creative out of the box innovative approaches. And Stephanie Evergreen is a specialist in data reporting and visualization for clarity and accessibility.

After this presentation, all of these resources and the tools that you saw on the slide before are linked so that you can access at least the main websites for the tools and the people's profiles on LinkedIn, and then as well as the different resources where you can purchase or find those resources.

Okay. Thank you so much. I really appreciate your time tonight. I really appreciate you coming out.

Please connect with me on LinkedIn. I pretty much live on LinkedIn. So if you ever want to talk more about this, or ask some follow-up questions, take some time to process, please reach out. I would be happy to have a chat with you.

Thank you so much, Angela. Awesome presentation. We are now in the question and answer phase. So we're gonna take questions both online and here in person.

We're gonna alternate back and forth. We'll start here in the room with Eric. Eric, I'm gonna hand you the microphone. It's a little tiny one.

Thank you. A little too close there. What was the name of that storytelling book? There was another noise over it.

So I just didn't get it. So... Storytelling with Data. By Cole?

By Cole Nussbaumer Knaflic. And when you receive the presentation on the resources slide, it's the very top link. Okay. Thank you.

You're welcome.

And I think we had just a question online. Which is... The slides being made available. So... They will be made available.

No, no slides for you! Ha-ha. Yes. The slides are gonna be available.

All the resources that are there are linked. The links were working as of yesterday. So that's the furthest I can promise, as far as links being up to date. And then the people you can follow on LinkedIn will link you directly to their profiles as well.

And we will also show that Slack link, which -- we will be posting the slides in that community as well. All right. Anyone else have questions here in the room? Okay.

Not that I design slides, but a lot of times at my company, when I look at slides similar to what you had for market share or bar graphs, wherever they're pulling that from and making the slide, it seems like the labels are always incredibly small. I don't know if that's a standard, and I don't know how the people that design those can fix those. Is it a simple fix? Or is it literally like putting a text box with a big number in there, or even your bars, when you had 40% -- a big number 40 in the middle of that bar would be helpful.

But they probably don't do it at my company because it's manual and the program doesn't allow it. Or... Any thoughts around how to make those better? Sure.

So... It really depends on the platform you're using. For example, with Excel, obviously, you can adjust the font sizes on your charts and your datasets. In Power BI, there are options and settings to the right after you create a chart that lets you add all kinds of different elements to your chart.

So it's literally about just going through the menu and clicking what you want, and not clicking what you don't want. So there are ways to add different elements. You can tell it what size you want. So it does really depend on what platform or app or whatever it is that you're using.

Like I said, there are some that are getting better with accessibility. But not as good. If you want the two most readily available to people who might have accessibility needs, you would be looking at Excel or Power BI. Sure.

All right. From online... Where do most dashboards in reports... Where do they most typically fail when it comes to accessibility?

Okay. So in my opinion, the reason why dashboards will fail for accessibility is because you're trying to show too much in too little of an area. So we want to feel like we can show something impressive to leadership. Right?

But if it's too crowded and you have too much going on, the understanding is immediately going to be lost. So when you want to tell a compelling story through dashboards, make sure that you're utilizing layers, that you have different dashboards that you can reference for different things. Also, don't throw all your data for something into one dashboard. Because, again, you want it to tell a story.

So try to bucket your data by theme or by concept, or by whatever it is. The narrative that you're trying to create. So, for example, going back to our engagement chart, I might have dashboards just for Q1. And then a whole separate set of dashboards just for Q2.

And what I can do is I can go through those dashboards to show that story that I'm trying to tell, while they're not all on the same page. Because then everything gets lost in translation. If it's too busy, the human brain automatically just goes... You know what?

I'm not paying attention to this. And we don't want that to happen when you're trying to tell a compelling story.

Thank you so much. So... Assuming that all the colors and captions and...

And font sizes are all accessible, are there types of charts and visualizations that are more or less accessible than others? There are. That's a great question. So simple is better.

So... When we think about what we have always used with charts, or what the most basic charts are, we're thinking about line graphs, provided that the line is thick enough. Bar graphs. I'm actually not a fan of pie charts.

Because I think you're squishing too much together. I really like line charts and bar charts the best. There's really not a whole lot that you couldn't do through a line chart or a bar chart. So those are definitely the two most accessible.

Because you're able to see the information parsed out. But again, if it looks too busy to you, then it could be virtually incomprehensible to someone who doesn't have the cognitive -- the ability for that cognitive load, or the appropriate amount of vision that they would need to be able to see that chart. So I really -- I try to stick with just line graphs and bar charts. They're not that exciting.

But again, when you're telling that compelling story and you're not forcing everybody to read every element of the bar chart, and you're focusing on the point that you're trying to make, it doesn't really matter anyway what you're using to show the data. Thank you for that question.

More questions in the room?

Do you usually go straight from data to charts? To form your narratives? Or do you ever put your data through any type of models to help form that narrative? It really depends on how complex the narrative is.

For the story I'm trying to tell. But then also if the narrative is becoming too complex, then sometimes I worry that I'm trying to tell too much of the story. And when I focus on what is, you know, essential in this story, then I usually do just go straight to charts. Yes.

And then to building that narrative. If I find that something needs to be put through a model, then I might be trying to do too much in too little of a time span. Oftentimes, management only has a few minutes. So I might just be trying to convey something very quickly, super high level.

And then again, that's where I might follow up with an email that includes attachments for all the other things that -- if they want to do a deeper dive, they at least have access to those materials.

Thank you for your talk. I wanted to ask... Do you have any experiences with clients or brands that have very strict brand guidelines? Or, for example, they only have... You know... One main color?

I wouldn't say it's very accessible to have multiple shades of green in a pie chart. So what would be your solution or suggestion for that? Other than telling them they need to change their brand guidelines, which... I have done in the past... So... This is where whitespace is gonna be your friend.

If it's not perfectly accessible, usually if you make it bigger and you add more whitespace, it at least helps a little bit. If you can't get them to agree to colors that pass the color contrast, then rely on black and white. And making it big. I mean... Sometimes I've done presentations where the client or the organization I was with will say: This doesn't really follow our brand guidelines.

And I'll say: Okay. Pass it over to design and let them work on it. And then sometimes we can have a -- and this is a way to have a constructive conversation with UX. Right?

It might just be the brand guidelines aren't accessible, because they didn't realize. So this is a great teaching moment. This is where you can kind of come in and partner with them and say: Hey, let me show you how we can make this more accessible while still adhering to the brand guidelines. And then you can just tweak how you're using your brand colors and where you're using them.

And again, if all else fails, black text, lots of whitespace, big letters, and simple datasets.

Ben here.

I'll pass it to you, Julia, right afterwards. I have a hack for the brand guideline color palette thing that's worked with a couple of clients. If you show them a leaf in sunlight and part of the leaf in shade, that leaf is the same grown. But if you can show them a step ramp or a tint treatment of their brand color that is slightly darkened or slightly lightened, and just kind of show them that that is a gradient along this -- it's maintaining the brand presence, but in a way that allows that content to reach the broadest possible audience, I've had some success in convincing them to adopt, for certain use cases, that step ramp kind of brand-adjacent color palette for specific use cases, even if it's not the primary brand color for other things.

So there's ways you can kind of hack the color palette as well with the right storytelling. The right narrative, to help them see that... So that's another hack you can try. Anyway... I'll hand to you.

Right there. Right here? Okay. Yeah.

So... You mentioned how data and visual storytelling meet the perceivable, operable, and understandable principles of the WCAG standards, Web Content Accessibility Guidelines, for those of you who aren't familiar. You know... Again, like it says in that... But how does that...

How does that relate to the robust principle? Can you talk about that? Sure. Not a problem.

So... The robust aspect comes from the storytelling itself. So when you're ensuring that everyone can understand the message that you're throwing out there, and everyone can then process your message, and then help you to solution -- to find actionable next steps -- that's where it's robust. So when you're looking at the data chart itself, you're not so concerned at that point what people are understanding.

Because it's not their job to sit there and pick through all of the data in the chart. It's their job to understand what you're trying to convey, and that's where the storytelling is ultimately your responsibility, to do that.

Got a great question here online for... This makes sense. Have you been able to get all teams in an org to use this? They mentioned that they're on a security team.

So as a security professional, they're having a lot of visualizations. They don't generally see accessible security reporting unless they create it themselves. How do you get that going? One of my most favorite soapboxes ever.

If something is not accessible, it is not secure. Okay? So let's think about that for a second. In banking, if you cannot access your dashboard, or if you cannot access your information, or whatever it is that you're trying to understand for your account, and you need someone else to help you do that, that is a security breach, and that is a security risk.

Okay? So what I tell my security team -- I'll give a huge shoutout. His name is Alex. And he is amazing.

He has been designing accessible security trainings for his team. We've been working together all year long on this. And when we first, first had a conversation, and I looked at him, and I said... For our members, accessibility is a security risk.

And then we started having that conversation. His mind was blown. He was like... I didn't even think about it that way.

So if you're on a security team, this is where it is imperative for your organization to ensure that everything they're churning out, everything they're trying to convey through their data storytelling, is accessible, because at that point, you have a security risk, where if people don't understand an element of that, then it could create a real problem. So I love, actually, one of my favorite things to say is... You know... If you need someone else to help you understand something, you're putting things at risk by the extra layers of people that you're adding onto the process.

Right? So that's why it does become a risk for security. Internally, let's say, somebody can't log onto their device. Let's say your device is using reCAPTCHA, which I hate, but that person needs someone to help log in every day, which is a security risk.

Let's say you're trying to convey information in a report and somebody asks somebody to read it to them and it was confidential. That's a security risk. So that's why this becomes really important. Especially within that context.

I'll take one last question in the room, if anyone's got one.

Hi. Thank you for an amazing presentation. So our company will be using the Power BI for designing the data visualization. But one of the challenges that we've been suffering is... The font size.

If we increase the font size, there's no whitespace at all. But if we reduce the font size, it's becoming 8 point, which I can't read. So I was wondering if there was any way to solve this issue or problem or challenge. Sure.

So typically the issue in Power BI with font size is when you're trying to add too much data into your visualization. So this is where it comes down to... You might need separate layers of data visualization to tell your story. So you can almost think of it like story boarding.

So you have all of this data that you want to share, but you're gonna do it in a logical step by step process. And that enables people to focus on one aspect of your data story at a time. Instead of trying to gather all the information at once through one dashboard. So keep it simple.

Layer it. And ensure that you're telling a cohesive story that is chronological or naturally follows from one concept to the next.

I'll add also: Give that feedback to whatever tool you're using. To make the authoring tool better. If you're having issues, give them that feedback. There's probably people working in the company that love to hear people want more accessible designs.

They need that feedback to come in to support the work. This is where networking comes in. So if you're not connected with someone at Microsoft who does design or data or someone who works on the Power BI project, you might want to start digging. Find those people.

Make those connections. Build that rapport. And then you can really have that conversation together. Great.

All right, everyone. Let's say thank you so much to Dr. Angela Young.