Primp Your Slide Vol. 2: Stacked Bar Chart with Survey Data [Video]

What’s up? Lea Pica here. Random fast fact – I can usually identify a movie’s composer within the first few minutes of the opening. Nerd! And this is Present Beyond Measure Episode 008.<announcer> Welcome to the Present Beyond Measure Show, where you’ll learn the best tips, tools and techniques for creating and delivering data visualizations and presentations that inspire data driven decisions, change hearts and enlighten minds. If you’re ready to get your insights noticed, remembered and acted upon, you’re in the right place. Now your host, Lea Pica!

Hey guys! Welcome to the eighth episode of the Present Beyond Measure Show. The only podcast at the intersection of analytics, data visualization and presentation awesomeness. This is actually the second volume in a video series that I'm calling “Primp Your Slide” – free online makeovers of your crappy charts and stinky slides!

Now, before we jump in I would just love to clue you in on some of the stuff that's been going on with me lately. As I'm recording this I just returned from the fabulous mystical land down under that would be Australia, where I presented in Sydney and Melbourne for wonderful analytics consultancy called Loves Data and it was their annual analytics conference. Whoo! That’s a mouthful. I got to share the stage with just amazing luminaries like Jim Sterne, Krista Seiden, Tim Wilson, Simo Ohava, Carey Wilkins and other amazing thought leaders in the space. It was a truly surreal experience to be halfway around the world and talking to some of the most warmhearted and friendly, kind people. Although I am a little depressed I wasn't allowed to hug a koala. Apparently hugging them could kill them because they need to sleep all day and have to digest their food. So needless to say my three-year-old son was extremely disappointed in me when I came home, but we’re working on that.

So that was incredible and I'm also kicking off a national tour of PowerPoint and Excel data visualization workshops for the Digital Analytics Association. This is an immensely exciting project and it's all about how to get the most out of using PowerPoint and Excel for your data visuals, when you present your information and how to save time doing it which is one of the number one complaints that I get from my audience. There are so many little tips and tricks that we’re not aware of when we're just re-creating those charts and slides over and over and this is going to be a gentle firehose of tips that will really help you make a foundational transformation to how you design. We’ve got locations coming in Philly, New York, Seattle, Atlanta and Miami starting in 2016, with a lot more on the way. My first one is in Philly and it's just around the corner on October 29th. It is open to the public but if you are a member of the DAA you get to save an extra 15% so definitely check that out. Get signed up. I would absolutely love to meet you.

So for those of you listening in your car there is a video that accompanies this episode because it is a video makeover. You will still get a lot out of it just by listening but if you'd like to see the makeover in action, just hop on over to the show notes page at leapica.com/008. Let's do it!

<music>

Hey guys! Lea Pica here. Welcome to Volume Two of the Primp Your Slide video makeover edition of the Present Beyond Measure Show and today is the first submission sent in by you, the listeners. So exciting! Today’s submission comes from Donal Phipps. So Donal, thank you for your bravery, I mean enthusiastic participation in this little exercise. Now Donal submitted a variety of slides so why don't we go check out what he sent in and Primp His Slide! <music>

Today Donal writes, “I work for a subsidiary of an industrial B2B company selling instrumentation and consumables. This is an early attempt to classify visitor behavior on our website. We want to help marketing stakeholders and understand the needs of customers visiting the site so they can plan resources and create web content which meets those needs, improving quality and relevance of the online information we provide to customers. And please don't shoot me for the pie chart. I've subsequently learned the error of my ways.” No worries Donal. We all make mistakes, so you are forgiven this time. Let us never speak of pie again.

So thanks for that Donal. You had some really interesting data here but there is one I'd like to focus on in particular and it's this. It's titled “What needs are we failing to meet?” and it shows answers to a voice of customer survey question. Now I love working with the voice of customer data because it's a true glimpse into what your most important stakeholder really wants. This is a 100% stacked bar graph to show successful versus unsuccessful visits by task and it's using a dual access to encode visit volume.

How do I know that? Because I have access to the raw data and that's actually a little issue that we’re going to be talking about in just a few minutes. Now if you watched Episode 004 you’re already familiar with my approach for creating simple and effective data visuals which I call uncreatively the PICA methodology.

PICA stands for Purpose, Insight, Context, and Aesthetics. It happened to fit the methodology well, and secretly I really wanted to give my married name a meaning other than a small, cold climate rodent with short limbs and spherical body that sometimes eats dead birds. Mmmm…when's dinner?! So if this doesn't sound familiar to you and you haven't watched Primp Your Slide Volume 1, I highly recommend pausing this and going back through it at leapica.com/004. If you don't have the time, no problem. I'll just quickly recap as I go. So, let's whip this guy into shape PICA style.

We’re gonna start with P which is P for Purpose. Remember when presenting data we always want to ask ourselves, “Why does this slide or chart exist? What burning question does it answer?” And this slide's purpose as Donal said is to identify key pain points with an experience. But at first glance, the answer doesn't exactly jump out at you. Remember, an effective visual creates the least amount of work for the viewer to interpret and internalize the information so they can pay attention to you. So the real question this graph should instantly answer is “What tasks had the highest percentage of unsuccessful visits?”

To answer that we’re gonna go to the second step of the PICA methodology and that is I for Insight. If you remember, we define insight as “the capacity to gain an accurate and deep intuitive understanding of a person or thing.” But when we look back at this, I would say it's a little difficult to glean the exact insight here because of the way that this is set up. The first thing I’m going to do here is I’m going to take away some of the noise, so I’m going take away the take away box. (Like what I did there?) I will get back to that in just a second. And then we’re also going to get rid of this source data box. That is important, but not quite positioned that way so we really just want to focus on this graph, so… blow this guy up here because I like to get really up close and personal with my graphs.

Now the first thing I want to do here is remove the dual axis visit count for the moment because I believe this could actually be interfering with the comprehension of this graph. In my agency days, little back story, I used dual axis graphs like it was my job. Dashboard real estate was at a premium so I crammed as many metrics as I could into little tiny spaces as possible. The problem is dual axis charts I learned later especially with unrelated metrics can imply that there's a correlative relationship between two metrics when there really is none.

Even when the metrics are somewhat related, if the lines cross over, it gives the false impression that it’s somehow meaningful, and at last year's eMetrics Boston, I had the pleasure of attending a presentation by someone named Tyler Vigen, author of a popular website and a book called Spurious Correlations. Spurious is defined as false or fake, and the whole basis of this website is about how plotting data on dual axes can create very misleading assumptions about the correlation of data, like per capita cheese consumption correlating with the number of people who died by becoming tangled in their bedsheets. This site is really funny, especially for data geeks. I highly recommend checking it out and I'm definitely going to be getting his book. But I digress. So if you want to check that out it's TylerVigen.com.

So now that we have visits gone, this a little bit bigger, I took this step to sort this chart by most unsuccessful visits. Sorting bar and column charts might be the most important step you can do to help your audience quickly understand your data. We can now clearly see that account information holds the top two places. Uh oh. Might be time to go back and assess that experience, and what's great about this is there was very little brain work needed to pick that out once I sorted it.

Well, isn't that an aesthetic choice you ask? There’s a gray area. Sometimes I think I should put I for Insight after aesthetics in the methodology but then I'd have to call it PiCAI. That doesn't really work for me and you know, and this is flexible. There are really no hard lines so I like to be flexible with this.

So now the title of this could go the extra mile to answer the question instead of asking it. So, I changed it to a key statement. A statement about my key message. And you guessed it, this is a McKinsey Title which is stating the insight as the title of the slide instead of just an observation about what it is and you have what the slide actually is in a little subtitle underneath. And I still find this is the fastest way to send my key message into the audience's brain, and it removes the need for that separate take away box which the only thing it's really taking away is attention from your actual graph.

So, the question we have now is “Is this really telling the whole story?” That's the next up in the methodology which is C for Context. So you remember how we got rid of the volume of visits? It turns out that's a pretty key piece of information here and I rarely like to present performance metrics without volume because you might actually optimize for something that won't move the volume needle for you. So the question we need to ask ourselves here is “how does visit volume impact our decision?” Well, it definitely impacts it I can tell you, but without a dual axis how do we show this?

My solution to this is a side-by-side bar chart that lets you keep a consistent view with your performance bar chart with an instant understanding of the volume context. You can see how I added one here and the story definitely changes a bit. Before we might've been alarmed that the top two worst areas were account information, but you can see that order status had the fewest visits. They were really small. So improving this area is not going to create the largest impact for you. But now take a look at product information specifications. Here it's the third most unsuccessful task and it looks like it has the highest volume of visits. This should be the first place we look to optimize.

So let's bring this home with our last step in the methodology which is A for Aesthetics. Aesthetics isn't just about making something look pretty. It's about using design principles to help quickly and accurately support our key message. So even though we have a real insight here and good context, and the slide is answering its purpose, it kinda looks like it got run over by a reject Crayola crayon truck. Am I right? And there's also a mess of cognitive loading junk on here that's gobbling up precious brainpower to wade through and it's slowing down your audience’s comprehension. So we’re gonna do a little chart detox to let our insight shine through and I love me some detox. So first I increased my data to pixel ratio by removing items like gridlines and axes, adding data labels and increasing the bar gap with. Those bars were looking a little underfed. And you can already see how much clearer this information is coming to you.

We’re not quite finished. Now it's time for a color purge. I love color. I really do. You have to be so careful about how you use color. Red and Green could work for unsuccessful versus successful, but it's a little jarring on the eyes so I’d rather go baseline with two shades of gray. Notice I chose a dark gray for unsuccessful because that's the story I’m highlighting here and I chose a midpoint gray for visits to help with that context. Now if you're thinking this looks kinda like a stormy bag of Hitchcock, fear not! There is a reason behind this. Now to highlight my key insight I'm using a colorblind safe standout blue tone.

You see how much faster my story pops because of the connective tissue I’ve created between the title insight and the data? That's data design baby! And one last thing, make sure to source the data and cite it at the bottom of your slide in small text at the bottom left. That removes the need for that big box we saw in the beginning, but you always want people to believe you actually got the data from somewhere.

So let's take a look at the slide before and where we are after applying the PICA methodology. I hope you'll agree this will work a lot better to communicate this information and answer that business question. I'd love to hear what you think about it, so please leave a comment and let me know.

So to recap the methodology we have P for Purpose. Make sure it has one. I for Insight – are you asking a question or you help answering it? C for Context. Make sure you're telling the whole data story. And A for Aesthetics, making sure data design principles are helping your data story shine through.

So, thank you again so much Donal for your willingness and bravery to share and have your work critiqued. I salute you. And I hope you and my listeners found this useful. And now it's your turn. I would love to see your naughty charts or slides and help you figure out how to tune them up and make them hum. It's completely free and your question or challenge may help others in the measure and marketing community too.

To submit your own work for a tuneup, please visit leapica.com/makeover upload your files with the little form there, and if you just want to submit a general question that's totally fine too.I’m going to be doing regular Q&A episodes as well as address what you need help with the most. So please don't be shy. Make sure to comment on these episodes of any thoughts or approaches you would've had, sharing our challenges and coming up solutions together are what's gonna make all of us learn and grow. Right? Nice.

<music>

Welcome to the upgrade segment. My power tip of the day for doing your work better faster and stronger and instead of doing my typical little tip here or there, from time to time I’m going to be bringing you new different software programs and websites and apps that help me do my job a lot faster and still get amazing results. So the first one I want to bring to you today is called KeyRocket. I have to tell you about this program. It's changed my life in terms of productivity using Windows and Excel and PowerPoint. And it was created by a company called Veodin and I've actually met the folks there and they're really lovely people. What KeyRocket does is as going through your work in Windows or Office or Excel, let's say you use your mouse to like cut and paste from one slide to the next or in your windows you try to create a new folder and you used your mouse to do it. KeyRocket will pop up a little window that says, “Wait guys! You could've done it faster with this shortcut.”

So it's like a little sports coach for you that will jump in and tell you the fastest way to do something. You’re going to start to see your more common tasks pop up over and over and it starts to count how many times you’ve used the shortcut correctly and cheers you on when you eventually mastered the shortcut. It's really great and something else that's great is if you do something a lot of times that take a lot of keystrokes, like for example I often repeatedly send an object to the back of the slide. The keystroke for that is like eight strokes long, so KeyRocket will actually suggest to you a shorter shortcut that you can custom program into your computer and that one shortcut alone has saved me I don't even know how many hours because all those little minutes add up as you’re stopping what you're doing with your keyboard and moving your hand onto the mouse.

And one of the coolest things it does is it tallies all of your productive gains over time and watches your overall progress, so it tells you how many clicks you've wasted, and how many you've remembered and mastered. It's really fun to watch especially as an analyst. Geeking over that is really fun. I'm not exactly in love with the pie chart, but that's a story for another day.

Now if you go straight to their website and download an evaluation copy of their software, you won't actually get to experience the full power of what the program can do, so I’ve partnered with them and they've given me a personal key that unlocks the full version of the program for 14 days. So if you'd like to get that key, head on over to the show notes page at leapica.com/008 and there's a little yellow download box that you can click on and you'll also be signed up for my newsletter, for all of my free stuff, and you’ll get the code. I'd love to know what you think of this program. It's been a game changer for me so definitely let me know.

Alright. That's all I've got for you today. Thank you so much for watching or listening to this special edition of the Present Beyond Measure Show. If you like what you’ve heard hop on over to iTunes to subscribe, leave a rating or review. Ratings and reviews are extremely appreciated because they affect the rankings of the show and I'll be reading out my favorite ones on future episodes. And if you want to continue the conversation and see all of the show notes for this page, please visit leapica.com/008 to review the notes, get a transcript, and find all of the resources that I've mentioned.

And you can also sign up for my newsletter which has tons of exclusive tips, tools and techniques for presenting your data. You'll also never miss an episode of the Present Beyond Measure Show when you sign up. And I would love if you could leave me a comment or any suggestions because I want to hear about the challenges you face when presenting information, doing data visualizations quickly, anything you'd like for me to talk about here, or you can tweet me a question for the show by including my twitter handle which is @leapica and including the #PBM as in Present Beyond Measure.

And I'll leave you with today's bit of presentation inspiration and that comes from Al Shalloway, and that is “Visualizations act as a campfire around which we gather to tell stories.” Let’s all make our presentations a cozy campfire for telling our data story. Til next time…Namaste.