12 Essential Tips for Designing Effective Dashboards with Geckoboard
Learn how to design dashboards that communicate data clearly and efficiently. Discover tips on layout, visualization, and user engagement to enhance decision-making.
File
12 Dashboard design tips for better data visualization
Added on 09/25/2024
Speakers
add Add new speaker

Speaker 1: Hello, I'm Dale from GekkoBoard and today I'm going to give you 12 tips that will help you design the perfect dashboard. Data, data, data. Businesses today are surrounded by data, which is nice, but it's not that helpful if you need to make quick decisions. This is where dashboards can help. Dashboards take your most important data and present it in a way where you can easily tell what's going on at a glance. Because dashboards need to get across information quickly and efficiently, the design and layout of the dashboard becomes all that more important. By the end of this video, you're going to have the knowledge to design quality dashboards that let people know exactly what's going on. So the most important piece of advice we have is to be clear about what it is you're trying to achieve with your dashboard. Who is this dashboard for? How will it help with the decision making? Are they trying to monitor some kind of system or process, or perhaps they're trying to track the progress towards a goal? So the clearer you are on the purpose of your dashboard, the easier it's going to be to know what data to include and how to present it. Think of data communication like a game of catch. The better you throw the ball, the easier it is for the other person to catch it. The most common mistake designers make is overloading the user with information. In other words, they throw too many balls at once. Think carefully about the audience of this dashboard. Do they need to see a detailed breakdown of each customer account every time they look at the dashboard? Or would a high-level number be sufficient? The key here is working with your audience. Sure, you need to understand what they do need, but you also need to understand what they don't need. And then you need to be bold in deciding what to leave out. Another way to improve your dashboard design is by reducing the amount of visual noise. Now, the American statistician Edward Tuft actually coined a term for this concept. Data-ink ratio. Data ink is quite literally the ink that communicates your data, whereas non-data ink is the stuff that doesn't. These days, you're unlikely to be printing out your visualizations with ink, but the principle still applies. Well-designed dashboards aim to use as little non-data ink as possible. In other words, they have a good data-ink ratio. In practice, this means removing unnecessary gridlines, icons, colour, labels, or anything else that doesn't actually communicate data. What you will find is that looking at dashboards with a bad data-ink ratio is incredibly hard work for the user, especially if they're looking at this dashboard several times a day. Decorative elements may grab your attention the first time you look at a dashboard, but the novelty very quickly wears off, and over time these things just start to get in the way. Overly precise numbers can actually slow down our comprehension. Ideally, you should round your numbers to a point where the differences between figures represent a material change. As you can see here, abbreviations can also make it easier to understand what's going on.

Speaker 2: Okay, so tell me, which is bigger, red or blue? Need more time? Okay, now try.

Speaker 1: Easier, right? When selecting a visualisation, you should always choose the one which most clearly and efficiently communicates the data. That means it should take as little time as possible to understand what's going on. In general, we humans aren't very good at comparing and contrasting spatial area. That means that pie charts and area charts are rarely the best choice of visualisation. I know it can be tempting to choose a different visualisation just because we want to spice things up a little bit, but just don't do it. If choosing the right visualisation means choosing the same visualisation, then that's fine. You're doing your job. Most of the time, you're going to be using a combination of numbers, bars, lines and tables. Okay, so when I was a kid, before eating things like M&Ms, I would often group them according to colour. I was a pretty weird kid. But at least this type of worrying behaviour makes you more likely to be a good dashboard designer in the future. Why? Because good dashboards group related metrics. So if somebody wants to see the metrics related to revenue, don't make them look here and here and here. Instead, group them. Plus, you save on non-data ink, because now you need less labels. If you have a related group of metrics, it's helpful to be consistent in your choice of visualisation. For example, let's take a look at this dashboard. So here we have the number of orders each day, and here we have the value of those orders. Clearly, there's a relationship between these metrics. So, we're going to want to use the same visualisation to help the user understand that relationship. Again, don't worry if this feels like more of a hassle. Again, don't worry if this feels like you're being repetitive. What you're actually doing is helping people think about related metrics in the same way. Not all KPIs are equal, and a good dashboard should reflect that. If there's a metric you want people to pay attention to, make it bigger. Put it in the top left-hand corner. The top left-hand corner is where our eyes are naturally drawn, followed by the other edges of the dashboard. Numbers on their own don't actually tell us that much. Let's say yesterday's revenue is $5,000. Is that good? Bad? Normal? Unusual? How does it compare to last week? Has it gone up? Has it gone down? How does it compare to our target? Giving context to a number in the form of a status indicator or a comparison can quickly tell us the significance of that number. Just beware of the first rule. Too many status indicators can create noise. If your entire dashboard is red and green, then the significance of those colours is greatly reduced. Similarly, a number or a chart isn't very useful if we don't know what it represents. Use labels to clearly state what the metric represents, and also what time span. Your labels should be clear and concise. Avoid notes, explanations and caveats. These things might feel like they're helpful the first time somebody uses a dashboard, but it very quickly becomes visual noise. So far in this video, we've covered 10 rules for designing the perfect dashboard. But the thing about design rules, like all rules, is that they're made to be broken. At the end of the day, dashboard design is about more than just how efficiently you can lay data out on a 16x9 rectangle. Ultimately, it's about people. It's about to what extent you can get those people to engage with the information. And sometimes, the most engaging dashboards do break the rules. For example, at Geckoboard, one of our company dashboards includes a list of recent subscribers. In a strict design sense, it's probably not as important as some of the other elements on this dashboard. But we include it because it sparks conversations and increases engagements with the dashboard overall. So whether it's a Twitter feed, new deals or a daily cat gif, remember that it's okay to break the rules if it means that people engage with your dashboard. Just don't overdo it. The last thing to say is that you should keep tweaking and evolving your dashboard.

Speaker 2: The best way to tell if you've designed a great dashboard is by using it.

Speaker 1: Does it help people do what it's supposed to do? Make sure you continue to listen to feedback, because when it comes to dashboards, you can always improve. So there we have it. 12 tips for designing the perfect dashboard. Now, all the lessons we've covered today are applicable to all of you. Now, all the lessons we've covered today are applicable to any form of dashboard design. But you may have noticed that many of the examples have been using our very own dashboard tool, Geckoboard. Geckoboard makes it easy to create professional KPI dashboards. It connects with over 80 data sources, and it's just much easier to use than many of the complicated BI tools out there. And one of the best parts is that Geckoboard is built with all of these design practices in mind. So it makes it very, very hard to go wrong. So if you're looking to build dashboards in a matter of minutes, then head over to geckoboard.com and sign up for a free trial. You'll be making dashboards within no time at all. Otherwise, I hope you enjoyed this video. If you did, please like and subscribe to the channel. And if you've got any questions about dashboard design, about Geckoboard, just leave them in comments. I will get back to you. Thank you once again, and have a great day. Transcribed by https://otter.ai

ai AI Insights
Summary

Generate a brief summary highlighting the main points of the transcript.

Generate
Title

Generate a concise and relevant title for the transcript based on the main themes and content discussed.

Generate
Keywords

Identify and highlight the key words or phrases most relevant to the content of the transcript.

Generate
Enter your query
Sentiments

Analyze the emotional tone of the transcript to determine whether the sentiment is positive, negative, or neutral.

Generate
Quizzes

Create interactive quizzes based on the content of the transcript to test comprehension or engage users.

Generate
{{ secondsToHumanTime(time) }}
Back
Forward
{{ Math.round(speed * 100) / 100 }}x
{{ secondsToHumanTime(duration) }}
close
New speaker
Add speaker
close
Edit speaker
Save changes
close
Share Transcript