You’re putting together a Jira report to present at a meeting. The presentation needs to be full of all kinds of reporting on Jira visualizations – timelines, percentages, and totals. Now, you could just hand over the raw data for your audience to sift through, but the chances are they don’t want that, so you’re not going to do that. You’re hoping to present at one of those rare non-terrible meetings. So instead, you take some of the key data and you turn them into various Jira charts, all designed to support the point you’re making, and hoping they’ll connect the remaining dots themselves.
Why, though? What is it about charts that makes them so effective? Would your presentation lose any punch by not including them? Answer these questions, and you’re on your way to making better charts, avoiding a myriad of potential pitfalls along the way.
Essentially, charts are pictures (which allegedly speak a thousand words). That’s exactly what charts do: they’re a kind of visual encoding that translates potentially massive amounts of information into a rapidly digestible format. For example, you can use Jira visualization to see in an instant something that could otherwise take much longer to identify. You might even spot a trend or anomaly that would go unnoticed were you to look purely at the raw data. Of course, there may be times when you have to crunch the numbers as they are, but when you’re looking to communicate information quickly and effectively, charts are the way to go.
It’s important that your charts are accurate and not overly complicated. Likewise, choosing the right kind of chart is crucial.
A poorly constructed or chosen chart, rather than making data more accessible, can have the opposite effect, causing confusion and the possible misinterpretation of data.
To illustrate this, let’s take a look at the pie chart below. Can you identify the third largest portion of the pie?
Not easy, is it? Of course, it helps when there are only three segments, and you can toggle on the key, but still… For at a glance Jira Visualization, let’s try the same data, in a different type of chart.
Much easier this time, right? That’s because pie charts are best suited to showing the relationship between all the parts that make up a whole, but elsewhere they tend to fall down. In the words of statistician Edward Tufte, “the only design worse than a pie chart is several of them”. A bit harsh, perhaps, but it’s certainly true that pie charts are frequently used in situations they’re just not suited to.
It’s an easy trap to fall into: pie charts are visually appealing and comfortingly familiar. When you want to show small differences between data points, though, they’re practically useless, unless you include annotations that state the values of each portion – in which case you might as well just present the raw data.
If you want to make better charts with clearer Visualization of your Jira data, you have to do more than just avoid pie charts. First of all, you have to be sure that what you’re aiming to present is actually worth creating a chart for at all. Sure, you could build a bar chart for two or three units of data, but it’s probably easier and more informative to simply present that data in a table.
Once you’ve decided that a chart would be genuinely useful, you should consider the pros and cons of different types of charts and choose whatever is truly right for your needs… For example, bar charts, as demonstrated above, are much better than pie charts when it comes to displaying small differences in values. However, they’re not without their own shortcomings: they don’t work well for small sample sizes or for representing continuous data1, such as temperature or time. Other chart types, meanwhile, such as scatter plots and histograms, are much better ways to show things like sample size and distribution of data.
Why Do Some Jira Visualisations Make Better Charts Than Others?
Humans vision didn’t evolve so we could read charts in PowerPoint presentations. Our eyes, our visual cortex and so on, they were made to help us survive in nature – to spot predators, to identify things that are safe to eat and to help us find our way in the physical world. It seems plausible that our strengths and weaknesses when it comes to interpreting data in charts stem from such adaptations.
Human perception is built so we can interpret nature, rather than data in charts. Understanding its strengths and weaknesses is a vital part of making better charts.
In any case, research has shown that humans are much more effective at understanding certain types of visual encoding than others. The work of statisticians William Cleveland and Robert McGill2 has been particularly influential in how we understand this phenomenon. They ranked all the different forms of visual representation used in charts, finding that people are generally better at judging the length of lines or bars, while they tend to struggle more with differences in things like angle, area, color, volume, direction, and shading. This is why bars generally make better charts than pies. It’s also why adding 3D elements can, as the Financial Times3 puts it, “obfuscate the information you want to communicate.”
Ultimately, though, the type of chart you use to represent your data is often what is most important.
How To Make Better Charts In Jira Reports
Statisticians and scientists spend a lot of time thinking and writing about the effectiveness of charts because data visualization4 plays a key role in their jobs. The right charts can help scientists better understand the work of their peers, thereby informing their own research and aiding progress. They may also be instrumental in convincing funding bodies that a project is worth paying a grant for.
It’s the same in the business world. Get the right chart in front of the right person, and it could be exactly the thing you need to push a deal over the line. Stick reams and reams of spreadsheets under people’s noses, and they might simply walk away – into the arms of a company that knows how to present data more clearly.
The trick is to choose a chart type that can relay whatever message you’re trying to send. Some kinds of charts, like pie charts, don’t say a lot about their underlying data, while a scatter plot can tell you a whole lot more. But that doesn’t mean you should default to the most complex chart type all the time, because it may not be appropriate, and you could end up confusing your intended audience. In spite of what Edward Tufte says, sometimes a pie chart is exactly what you need.
Despite how much thought scientists5 put into chart types, however, there is no definitive methodology for data visualization, but there are some straightforward practices you can follow to help you make better charts in general:
- Graphs should present data that is otherwise too numerous or complex to describe in text form, and they should take up less space. If your chart doesn’t do that, it’s not doing its job.
- Choose your colors carefully. If your chart relies on the reader being able to distinguish between different colors, they need to offer enough contrast for that to be comfortable. Also, think about the connotations that may be attached to different colors, otherwise, you could be sending a message that you are entirely unaware of. People have been shown to have very real physical and emotional reactions to certain colors. Colors also give you the opportunity to match your branding.
- Don’t default to pie charts. It’s tempting to use pie charts for everything, but they’re really only useful in a handful of situations. If you’re not trying to show how a series of parts make up a whole, then choose a different type of chart.
- Don’t use a chart at all if you don’t need to. Sometimes, it is better to simply display the raw data, perhaps in a basic table. This is especially true with simple data sets and small sample sizes.
- Use better tools. The right software can be enormously helpful when it comes to creating great charts. Custom Charts for Jira, for example, offers much more than the standard Jira visualization reporting feature set, giving the user much more control over how their data is presented. By default, Jira’s data visualization features are rather more limited, which can lead to the very problems described in this article.
- Make sure your charts are easy to read. That means avoiding confusing background images or colours and not laying out text in a way that makes your audience tilt their heads to read it.
- Make better charts because you want to. If you’re including charts in your reports and presentations just for the sake of it, you’re not likely to be doing yourself any justice.
- Consider using more than one chart for the same data set. If you want to present a lot of information about your data, two simple charts may be better than a single complicated one.
- Never sacrifice legibility for fancy styling. When it comes down to it, charts are there to communicate information, and if they fail to do that, no one will care about how pretty they look.
Can Better Charts Really Benefit Your Business?
The more numerous and complex your data is, the more likely you would benefit from using charts. They make something accessible which would otherwise be overwhelming. Scott Berinato, writing for the Harvard Business Review6 illustrates this succinctly with an aeronautical example:
“At Boeing the managers of the Osprey program need to improve the efficiency of the aircraft’s takeoffs and landings. But each time the Osprey gets off the ground or touches back down, its sensors create a terabyte of data. Ten takeoffs and landings produce as much data as is held in the Library of Congress. Without visualization, detecting the inefficiencies hidden in the patterns and anomalies of that data would be an impossible slog.”
Similarly, if you’re in a software development company, your department might encounter hundreds or even thousands of issues a day. When it comes time to assess the performance of your team for, say, the last quarter, you’ll want to look at the response times of your staff members – the aim being to identify who’s doing what, which tasks are taking the longest and so on. It would be impractical and self-defeating to go through all the data one by one, but a good chart? Well, that’s a different story. That will enable you to see the bigger picture in a matter of moments, so you can make better decisions, faster.
It’s this kind of thinking, of course, that underpins all business intelligence solutions, and it naturally translates into improved business performance.
Power-Up Your Jira Visualization Reporting
With everything you’ve learned from this blog post, you can begin to make better charts in Jira, but if you need more control, then check out Custom Charts for Jira visualization, from Old Street Solutions. It enables you to quickly change between different chart types, to customise colours, to filter data, and more. But it’s also accessible, meaning more of your people will be able to use it – so your whole organisation can benefit.
Below, you’ll find a comparison between standard Jira reporting and Custom Charts for Jira Reports.
For more information, you can try Custom Charts for Jira Reports here or on the demo playground.