In the pre-digital era, data was a subject for mathematicians and scientists. Now, one way or another, we can’t escape it.

Our constant use of online services not only relies on data, we are also a continuous source of data, generating information about all aspects of our lives.

Whether it’s data about the human body—thanks to the rise of wearables—our energy consumption at home, or data tied to our personal finance: we’re creating mountains of data, and now we need to find ways to make sense of it.

The rise of personalized data is poised to be a hot topic as companies seek to deliver real benefits from the information gathered on consumers. The challenge for designers lies in finding a way to reduce the complexity posed by such vast amounts of data and give data a human shape.

Data has to be accessible to the average person. It also has to provide the user with actionable insight in a way that is meaningful and accessible. This is where the true power of design can really make a difference: by using visualizations to help people navigate the confusing world of data we can improve lives.

Data visualization has come a long way since its formative days as the basic pie chart invented over 200 years ago. Now, thanks to the huge upsurge we’ve seen in data and the discourse around its usage, a new design language is emerging that is elegantly simplifying the big data mess into beautiful and meaningful visualizations.

So regardless of whether you’re bringing shape to data on health and wellbeing, shopping habits, or in editorial, Fjord has identified five core principals to follow when embarking on a data visualization challenge:

1. Understand the Source

Make sure you know the data you’re working with. This is the crucial first step in making sense of data. You need to understand the bigger picture: Why has it been collected? What value does the organization put on it? Who is the user? How can it be used to greatest effect? This insight will lay important foundations for the creation of a visualization that’s both meaningful and human.

2. Identify the Narrative

Good data visualization is so much more than a beautiful picture, it tells a story that can be understood by anyone. It’s therefore essential that you identify the story you want to tell first and then use the data as way of bringing that story to life.

For example, we recently helped mobile operator 3 Sweden become more customer-centric and transparent by re-designing the often-confusing monthly phone bill. Instead of continuing to present their customers with an incomprehensible list of numbers, they wanted to create something much more helpful and transparent.

Good data visualization tells a story that can be understood by anyone.

The result is My 3, an app that lets customers see their usage data in real-time so they know exactly where they are with respect to their plan. By using data visualizations we have been able to create a beautiful and innovative way to give customers access to their data that also demonstrates 3’s commitment to customer care.

3.Define the User Experience

Ensure you use data to guide but not dictate the overall experience. Data should act as a backdrop of understanding and learning that allows the user to create his or her own experience. It’s also worth exploring ways to deepen the insight you can share with the visualization and give people the flexibility to interpret the data in the most meaningful way to them. After all, an experience that delights is an experience that people will remember and use again and again.

4. Simplicity Rules

Data visualizations exist to inform the user, they’re not an excuse to overload someone with information they don’t need to know. As a designer, it’s your role to focus on simplicity, taking complex or disparate information and making it tangible, understandable, and, importantly, more human. Remember, where simplicity reigns, the user understands.

5. Avoid Reinventing the Wheel

Try to tap into existing behaviors and understandings around data visualizations. It will make your design more accessible to a broader range of people. There’s a reason why the pie chart is so widely used: people understand what it shows. There’s an innate elegance in designing visualizations this way, as they’ll have a greater impact and be immediately understood.

A Design-Led Approach

Good data visualizations are not only masterpieces of design, they are valuable tools helping us to interpret previously inaccessible content as something much more meaningful and actionable. As increasing numbers of organizations wake up the power and potential in their data, design will play an even bigger role translating something confusing into something that helps people. The trick is to take a design-led approach that puts the user first and focuses on simplicity in order to create an experience that never ceases to delight.

Related Articles

Add new comment


The bar graphs being placed side-by-side do not allow for easy comparison of their lengths (point 3, the bars, not the modified pie chart), and can distort the point. Likewise, it is difficult to judge the relative area of two circles (point 4), due to the area being a complex calculation involving pi (a circle of double the area, literally, appears only marginally larger than the "1x" area circle). A depiction using one dimension, length, is more clear. Point 5 has no labeled axes - what is the price of coffee in Denmark? Over $600? The viewer can't easily tell.

Academic journals have publication standards for the clear and unbiased display of graphical information. But these standards aren't as "pretty" from a design perspective. Thus, typically when designers make lovely graphs, they are often technically inaccurate, distorted, or inefficient.

Furthermore, understanding data slightly more complex than percentages and tallies may require some actual knowledge of quantitative methods and research design, lest a very pretty graph be made that is wholly incorrect.

Hello ,

I heard about a great MOOC (free university online courses, for everybody) entitled "The future of storytelling", and I wanted to share the good news with everybody interested in storytelling.

The MOOC will be about :
• storytelling basics,
• serial formats (on the TV, web and beyond),
• storytelling in role-playing games,
• interactive storytelling in video games,
• transmedia storytelling,
• alternate-reality gaming,
• augmented reality and location-based
• the role of tools,
• interfaces and information architectures in current storytelling.

The course starts on October 25th (this friday!), 2013, so enrol now and don't forget to share the good news with your friends :)

You can follow this URL to to discover the course and/or enrol :

May be I'll meet other readers of "ux mag" in the MOOC?! :)

Here's a free course about "The Future Of Storytelling" powered by @iversity

Well written and allows exploration and expansion of the idea into several related areas. The data need to be be related and relevant and therefore the source and user experience are key factors. Human mind perceives data based on the fact that the "beauty lies in the eyes of the beholder" and therefore it is very uch possible to orient the understanding of the data into different ways - therefore the classic picture of right way of presnting the data to give the message you want to transfer is most critical...Then the whole concept is geared towards proactivity...with out thought and proactivity it is not constructive to follow the five steps to present the data...and so it creates value. Again wel written 5 key points , appreciate the author.

Great post Dominic!!!! though I have a few doubts regarding this line of thoughts:

1. When we say data visualization, are we solely relying on the info-graphics for the visual appeal of the
content we are concerned about?

2. I was a little curious how we can use the same technique for the internal enterprise applications. e.g.
in healthcare systems, how can we reorder the manual form filled by the staff workers into something
more effective when seen on desktop or mobile devices.

3. Further in a few cases we have a requirement to make an error page look a little more useful to deal with.
can the same technique you mentioned be applied there as well with equal magnitude?

Kindly help me understand the above.



I understand your point, and it's more than valid for White Papers and glossy presentations for the general public but for people who deal with numbers every day, line charts, candlestick charts, etc... provide the necessary information. I find the problem comes in having to show several sets of charts and data tables as often times one set of numbers doesn't provide enough information (or would be too difficult to read with additional information) and a second chart is necessary.

I also have to bring in Tufte here -- as designers we have to be careful of using 3D objects (coffee cups for example) to represent changes in data. If sales go up 100% we have to be careful to not make the second cup twice as large as the first.

Awesome post Dominic. I totally agree with what you said.

Marketers should have a better understanding on how to story tell with the use of data. With enough knowledge on data visualization (among many others), they can emphasize to their client's the things that needs to be emphasized on or ignored.

This keeps all the noise at bay and have them focus on the really important things.

"It’s therefore essential that you identify the story you want to tell first and then use the data as way of bringing that story to life."

I have to disagree with this. Wouldn't it be more germane to identify the story the data is already telling you, then translate that narrative in a way that makes sense to the client?

To clarify, scientists have to be very careful to observe first, then formulate a hypothesis. Are you advocating here that we do the exact opposite? That is, decide what we want to say, then look for data to back it up? Wouldn't this lead to erroneous interpretations of data?

Yes. Excellent point.

I think that can be addressed by the first point he makes which is know the data. First you have to explore the data and then from that you share what you find is most interesting. The point he is trying to make is that there should be a clear narrative to data visualization because if there isn't then the audience can get lost very easily, data can be distorted, and essentially lead to confusion or misinformation. So I think he addressed that with his first point but I do understand your concern about the phrasing.

Thanks for sharing your thought Dominic. The data (visualization) you show as example look interesting (for coffee lovers). Is there a place where we can see them in full scope?