Data Visualization: a successful design process — Save 50%
A structured design approach to equip you with the knowledge of how to successfully undertake any data visualization challenge efficiently and effectively.
In this article by Andy Kirk the author of Data Visualization: a successful design process , we look at the broad variety of options for building our solution and the remaining important tasks to undertake before launching.
We will run through a selection of the most common and useful software applications and programming environments to help you select the most appropriate tool to match your design requirements and technical capabilities.
We will look at some of the key considerations around testing, finishing, and launching a design solution as well as the important matter of evaluating the success of your project post-launch.
Finally, we wrap things up with a discussion about the best ways for you to continue to learn, develop, and refine your data visualization design skills as you seek to master this fascinating and rewarding discipline.
(For more resources related to this topic, see here.)
For constructing visualizations, technology matters
The importance of being able to rationalize options has been a central theme of this book. As we reach the final stage of this journey and we are faced with the challenge of building our visualization solution, the keyword is, once again, choice.
The intention of this book has been to focus on offering a handy strategy to help you work through the many design issues and decisions you're faced with.
Up to now discussions about issues relating to technology and technical capability have been kept to a minimum in order to elevate the importance of the preparatory and conceptual stages. You have to work through these challenges regardless of what tools or skills you have.
However, it is fair to say that to truly master data visualization design, it is inevitable that you will need to achieve technical literacy across a number of different applications and environments.
All advanced designers need to be able to rely on a symphony of different tools and capabilities for gathering data, handling, and analyzing it before presenting, and launching the visual design. While we may have great concepts and impressively creative ideas, without the means to convert these into built solutions they will ultimately remain unrealized. The following example, tracking 61 years of tornado activity in the US, demonstrates a project that would have involved a great amount of different analytical and design-based technical skills and would not have been possible without these:
Image from "Tornado tracks"(http://uxblog.idvsolutions.com/2012/05/tornado-tracks.html), created by John Nelson/IDV Solutions.
In contrast to most of the steps that we have covered this far, the choices we make when it comes to producing the final data visualization design are more heavily influenced by capability and access to resources than necessarily the suitability of a given tool. This is something we covered earlier when identifying the key factors that shape what may or may not be possible to achieve.
To many, the technology side of data visualization can be quite an overwhelming prospect—trying to harness and master the many different options available, knowing each one's relative strengths and weaknesses, identifying specific function and purpose, keeping on top of the latest developments and trends, and so on.
Acquiring a broad technical skillset is clearly not easily accomplished. In order to accommodate the absence of technical skills, in particular, you may need to find a way to collaborate with others or possibly scale down the level of your ambition.
Visualization software, applications, and programs
The scope of this book does not lend itself to provide a detailed dissection and evaluation of the many different possible tools and resources available for data visualization design. There are so many to choose from and it is a constantly evolving landscape—it feels like each new month sees an additional resource entering the fray. To help, you can find an up-to-date, curated list of the many technology options in this field by visiting http://www.visualisingdata.com/ index.php/resources/.
Unlike other design disciplines, there is no single killer tool that does everything. To accommodate the agility of different technical solutions required in this field we have to be prepared to develop a portfolio of capabilities.
What follows is a selection of just some of the most common, most useful, and most accessible options for you to consider utilizing and developing experience with. The tools presented have been classified to help you understand their primary purpose or function.
Charting and statistical analysis tools
This category covers some of the main charting productivity tools and the more effective visual analytics or Business Intelligence (BI) applications that offer powerful visualization capabilities.
Microsoft Excel (http://office.microsoft.com/en-gb/excel/) is ubiquitous and has been a staple diet for many of us number crunchers for most of our working lives. Within the data visualization world, Excel's charting capabilities are somewhat derided largely down to the terrible default settings and the range of bad-practice charting functions it enables. (3D cone charts, anyone? No, thank you.)
However, Excel does allow you to do much more than you would expect and, when fully exploited, it can prove to be quite a valuable ally. With experience and know-how, you can control and refine many chart properties and you will find that most of your basic charting requirements are met, certainly those that you might associate more with a pragmatic or analytical tone.
Sample screenshot of Excel's charting capabilities
Excel can also be used to serve up chart images for exporting to other applications (such as Illustrator, see later). Search online for the work of Jorge Camoes ( http://www.excelcharts.com/blog/), Jon Peltier (http://peltiertech.com/), and Chandoo (http://chandoo.org/) and you'll find some excellent visualization examples produced in Excel.
Tableau (http://www.tableausoftware.com/) is a very powerful and rapid visual analytics application that allows you to potentially connect up millions of records from a range of origins and formats. From there you can quickly construct good practice charts and dashboards to visually explore and present your data. It is available as a licensed desktop or server version as well as a free-to-use public version.
Tableau is particularly valuable when it comes to the important stage of data familiarization. When you want to quickly discover the properties, the shapes and quality of your data, Tableau is a great solution. It also enables you to create embeddable interactive visualizations and, like Excel, lets you export charts as images for use in other applications.
Sample screenshot of Tableau's charting capabilities
There are many excellent Tableau practitioners out there whose work you should check out, such as Craig Bloodworth (http://www.theinformationlab.co.uk/ blog/), Jérôme Cukier (http://www.jeromecukier.net/), and Ben Jones (http://dataremixed.com/), among many others.
While the overall landscape of BI is patchy in terms of its visualization quality, you will find some good additional solutions such as QlikView (http://www.qlikview.com/uk), TIBCO Spotfire (http://spotfire.tibco.com/), Grapheur (http://grapheur.com/), and Panopticon (http://www.panopticon.com/).
You will also find that there are many chart production tools available online. Google has created a number of different ways to create visualizations through its Chart Tools (https://developers.google.com/chart/) and Visualization API (https://developers.google.com/chart/interactive/docs/reference) environments. While you can exploit these tools without the need for programming skills, the API platforms do enable developers to enhance the functional and design options themselves.
Additionally, Google Fusion Tables (http://www.google.com/drive/start/ apps.html) offers a convenient method for publishing simple choropleth maps, timelines, and a variety of reasonably interactive charts.
Sample screenshot of Google Fusion Table's charting capabilities
One of the first online offerings was Many Eyes, created by the IBM Visual Communications Lab in 2007, though ongoing support and development has lapsed. Many Eyes introduced many to Wordle (http://www.wordle.net/) a popular tool for visualizing the frequency of words used in text via "word clouds". Note, however, the novelty of this type of display has long since worn off for many people (especially please stop using it as your final PowerPoint slide in presentations!).
The ultimate capability in visualization design is to have complete control over the characteristics and behavior of every mark, property, and user-driven event on a chart or graph. The only way to fundamentally achieve this level of creative control is through the command of one or a range of programming languages.
D3 enables you to take full creative control over your entire visualization design (all data representation and presentation features) to create incredibly smooth, expressive, and immersive interactive visualizations. Mike Bostock, the key creative force behind D3 and who now works at the New York Times, has an incredible portfolio of examples (http://bost.ocks.org/mike/) and you should also take a look at the work and tutorials provided by another D3 "hero", Scott Murray (http://alignedleft.com/).
Sample screenshot of D3.js development environment
D3 and Flash are particularly popular (or have been popular, in the latter's case) because they are suitable for creating interactive projects to work in the browser.
Over the past decade, Processing (http://processing.org/) has reigned as one of the most important solutions for creating powerful, generative, and animated visualizations that sit outside the browser, either as video, a separate application, or an installation. As an open source language it has built a huge following of creative programmers, designers, and artists look to optimize the potential of data representation and expression. There is a large and dynamic community of experts, authors, and tutorial writers that provide wonderful resources for anyone interested in picking up capabilities in this environment.
Moving briefly away from interactive programming environments we turn to R (http://www.r-project.org/), a highly extensible, open source language for statistical analysis and graphical techniques. R has developed into a powerful and versatile method for creating static charts and graphics that transcend the creative limitations of software packages such as Excel. There is a large active online community, which can really help with the challenge of going through the learning process. To demonstrate R's worth, the New York Times use this extensively in their data sketching and static graphic workflows. Check out Mondrian (http://rosuda.org/mondrian/) and Wolfram Mathematica (http://www.wolfram.com/ mathematica/) for other powerful, statistical graphing capabilities.
Quadrigram (http://www.quadrigram.com/) is an innovative visual programming environment designed to enable anyone working with data to create powerful, flexible, and custom visualizations. It is intended to be accessible for people with limited technical and programming experience.
In support of open interactive journalism, the Miso project (http://misoproject.com/) developed by the Guardian and Bocoup, is a fairly recent arrival. It provides open source tools for developers and non-developers alike to facilitate the quick creation of impressive, extensible interactive data visualizations. There also exists an option for developers to get under the hood and extend and expand the computational methods.
Other notable programming tools to mention include Nodebox (http://nodebox.net/), which is a Python-based tool for creating generative, static, animated, or interactive visualizations, and KendoUI (http://www.kendoui.com/) for building interactive HTML5-based data visualizations for both web and mobile applications.
Finally, in recent times, we have witnessed the rise of WebGL (http://www. chromeexperiments.com/webgl/), a new web technology for rendering interactive two-and three-dimensional graphics. The utilization of this standard has so far seen more experimentation than particularly solid visualization work, but it certainly offers new capabilities for pushing the creative boundaries of data representation.
|A structured design approach to equip you with the knowledge of how to successfully undertake any data visualization challenge efficiently and effectively.|
eBook Price: €16.99
Book Price: €27.99
Tools for mapping
The great opportunity that exists these days for plotting geo-spatial data onto maps is matched by the range of tools available to accomplish it.
Powerful options come in different shapes and forms through Arc GIS (http://www.esri.com/software/arcgis), Indiemapper (http://indiemapper.com/), Instant Atlas (http://communities.instantatlas.com/), Geocommons (http://geocommons.com/), and CartoDB (http://cartodb.com/). Across these options you will find the ability to create, rich interactive visualizations of geo-spatial data and full-on mapping applications, typically offering flexible licensing and pricing plans from free/trial through to premium levels depending on your needs.
Sample screenshot of Instant Atlas mapping capabilities
For those designers with developer skills wishing to have greater creative control and freedom over their mapping solutions, there are a number of open source mapping frameworks and libraries such as Polymaps (http://polymaps.org/), Kartograph (http://kartograph.org/), Leaflet (http://leafletjs.com/), and OpenStreetMap (http://www.openstreetmap.org/).
Additionally, TileMill (http://mapbox.com/tilemill/) offers an extremely versatile and accessible application for making elegant data-driven maps whether you are a beginner designer or more-established cartographer.
Other specialist tools
Not all visualizations are interactive, of course, and some of the finest visualization works we see are static pieces. Infographics in particular are typically manually crafted designs, comprising a blend of different visual design elements (such as charts, illustrations, and diagrams). As we have already mentioned, often the chart elements we use for our static work originate from tools such as Excel, Tableau, or R with images imported to help construct a final work.
The vast majority of statics are produced using Adobe Illustrator ( http:// www.adobe.com/uk/products/illustrator.html ), the long-established and all-powerful creative package that has been the graphic and illustration tool for many years. There is now an open source alternative called Inkscape ( http://inkscape.org/) ,which offers an impressive array of features that offer a viable alternative for many peoples' needs.
For many people (perhaps those with limited access to varied resources) PowerPoint (http://office.microsoft.com/en-gb/powerpoint/) or Keynote (http://www.apple.com/uk/iwork/keynote/) provide a perfectly adequate platform for their data presentation needs. Another Adobe package, InDesign (http://www.adobe.com/uk/products/indesign.html) provides a further means for creating and publishing final static works.
If you're looking to create advanced motion graphics, modeling, simulation, and visual effects then Maya 3D (http://usa.autodesk.com/maya/) and Adobe After Effects (http://www.adobe.com/uk/products/aftereffects.html) are incredibly powerful, industry standard production platforms.
Finally, to showcase your static work, once you've created your final designs and want to publish and share image files, sites such as closr.it (http://www.closr.it/) or zoom.it (http://zoom.it/) enable navigable, zoom-able windows to host large, detailed images.
|A structured design approach to equip you with the knowledge of how to successfully undertake any data visualization challenge efficiently and effectively.|
eBook Price: €16.99
Book Price: €27.99
The construction process
So, you've selected the tools you'll need to build your design and you are now well in to the execution stage. We're not far from the finishing line but it's not yet time for you to lower your guard, lose your focus, or cease your momentum.
You see, this is the part of the design process where stresses and strains emerge—the ill-timed bugs, dataset problems, functional failures, unwanted interference. During this stage it is important that you keep your cool and see your tasks through as efficiently as possible.
As you work through the construction process, it is important to focus on getting the functional elements of your solution working first before spending too much time achieving your desired aesthetic or incorporating technical flair. It is always very tempting to spend too much time, too soon on things that shouldn't really be given such priority. Just remind yourself that there is no point running out of time trying to make something look good when it doesn't yet function. Remember, it will be easier to make something that is functional, beautiful, than it is to make something beautiful, functional.
As we mentioned earlier in the book, you will rarely create a worthy project without the need for iteration. While we have to present the sequences of the methodology in this book in linear fashion, there is always going to be movement forward and backwards between stages. This is something that should be accepted but also embraced—it is part and parcel of any creative process. While a methodological approach to this challenge gives you structure and a neat framework of concerns to work through, iteration gives you the creative breathing room to allow different ideas to blossom and influences to take hold. It is something that you should be prepared to do and plan for.
You clearly want to avoid long iterative cycles but smaller ones can really help you explore, clarify, and refine your potential solutions. It may be that you end up following two or three parallel options to quite an advanced stage and then see which emerges as the strongest. Indeed, some clients will state a need for evidence of alternatives before committing themselves. For these client-based projects, you need to maintain open dialog throughout to avoid any inconsistency in interpretation from either party. Do your absolute best to eradicate the possibility of last minute surprises about a solution not matching requirements or expectations. That is a sting in the tail nobody wants!
As you work through your construction stages, in particular, there will be points when you recognize a need to make certain sacrifices. There may be things you intend to include but can't justify them. Trade-offs are a constant necessity caused by time or resource constraints.
Some of the things we find hardest to drop are the most irrational. We often find ourselves in a sense of denial. This may come from a desire to include features that you have slaved over or become overly precious about.
We saw an example of this in the discussion about color in article 4, Conceiving and Reasoning Visualization Design Options. Here we saw an initially conceived title format for an Olympics project that was formed out of thumbnail images of all the historical event posters; this image is shown here:
When it came to incorporating this title into the final piece, it was clear that it drew too much attention away from the rest of the visualization. Despite this being obvious, because of the time and energy spent on making this title image, it was hard to relinquish. Thankfully, a sensible voice determined that we should drop it and find a simpler solution. Simple advice? Take the hit and just get over it!
Image from "Pursuit of Faster" (http://www.visualisingdata.com/index.php/2012/07/newvisualization- design-project-the-pursuit-of-faster/), by Andy Kirk and Andrew Witherley
As we approach the maturing stages of our development work, this idea of getting input from others becomes more important. It can be quite a tough moment to convince yourself that something is ready to be judged (in a prelaunch setting) but it is invaluable to test out people's responses to what you are creating.
You want people who are informed about the context of the work and also about the challenges involved in creating a visualization. You also need to trust them to give you constructive and reasonable feedback, otherwise it may prove a wasted effort.
You should be seeking feedback on a number of dimensions of your design in order to determine if the intention of your solution is consistent with the audience experience:
What is their instinctive reaction? Positive, negative, intrigue, confusion, or just a plain "so what?"
Can they understand how to read the graphic or use the tool? Does it have clear explanations and intuitive design in terms of visual hierarchy and structural arrangement?
Can they derive insight from it? Maybe throw them some test questions to assess the visualization's ability to effectively inform.
Does it work functionally? Can they find any errors, mistakes, programmatic errors, or any other design flaw that undermines the clarity, accuracy, or performance of the solution?
There are plenty of evaluation methodologies and techniques, probably much more sophisticated than this, but these are just some of the most useful prompts for you to gather feedback against before finalizing your work.
Approaching the finishing line
Here is a quote from Antoine de Saint-Exupery:
"You know you've achieved perfection in design, not when you have nothing more to add, but when you have nothing more to take away."
The finishing line is now getting ever closer. However, apart from those projects where there is a clear finite deadline to work to, the judgment of when a design is actually finished is not necessarily always obviously recognizable. A deadline provides this finality, but open-ended projects need their own completion point to be determined. It is natural to keep tweaking, refining, and enhancing your piece but eventually you need to call out something as being completed.
A useful signpost to note your progress was proposed by designer Martin Wattenberg (co-developer on the "Wind Map" project that we saw earlier). Martin describes the subtle but telling change in your role as you shift from debugging a design (programmatically or figuratively) to finding yourself becoming an enthusiastic user, engaging with your own work to unearth insights.
As the quote at the start of this section expresses, another viewpoint is to step back and away from your design and challenge everything that you have included. Justify to yourself (and/or to others) the reason why features or design choices need to stay, but also determine what elements you can eliminate, those that don't add any communicative or functional value. It's not necessarily about striving for minimalism; rather the most elegant and clear form.
As well as challenging all our design choices we also need to switch mindsets more towards the Project Manager or Administrator's perspective and undertake some important checks. Sometimes, when you're close to finishing you would prefer to stick your head in the oven than seek issues that need addressing, but you've got to continue to strive for optimal accuracy and intercept any potential mistakes.
Simple errors can completely undermine quality—an extra zero in a value, the mislabeling of a country, an emboldened font when it wasn't wanted, and so on. It might be the smallest and most innocent of mistakes, but that can be enough to tarnish the rest of your work with doubt in the eyes of your audience.
Try to see this as the final push. Paying attention to the finer details of your work will safeguard the project's integrity (and by extension your own, as the designer). Hopefully, much of the user-testing and evaluation work outlined just before will help in the identification of any problems in accuracy. People with a freshness of perspective can often provide great value on this front.
Whether it is them or you looking for these characteristics, here are a few things you need to watch out for:
Data and statistical accuracy: Scan through a good-sized sample of all your visualized data values to ensure there aren't any erroneous items or incorrect outliers. Check the rigor of all your statistics and calculations.
Visualization accuracy: Make sure that the way you have represented your data is functioning effectively and does not mislead the user or reader. Do all your representation choices accurately portray the data values they're associated with?
Functional accuracy: More concerned with interactive pieces—do all the functions and features on your design perform as you intended?
Visual inference: As we stated before, visual inference should equal data inference. If it looks like data, it should be data. If something looks significant, maybe through its positioning or color choice, then it should be significant. If there is any decorative element or other artifact that appears to be implying something it is not meant to, remove it.
Formatting accuracy: Check the consistency of your typography, in terms of type, style, and size. Make sure your color usage is consistent down to the RGB or CMYK code level.
Annotation accuracy: Read through all your titles, labels, introductory text, credits, captions, and check any units that you have included. It's not just about spelling or grammatical errors but checking to see if things make sense and are succinctly expressed.
The exciting and also probably anxious moment has arrived and your visualization has now been launched in to the wild!
How, where, and what this launch actually looks like clearly covers a very broad range of possibilities—it might be a chart in a report, a presentation to a board meeting, an infographic in a newspaper, or an interactive web-based project.
Regardless of how this piece exists, in an ideal world you would now seek to assess the visualization's effectiveness and impact in a post-launch setting. I say in an ideal world because sometimes you simply don't have sufficient capacity or resources to allocate to the post-launch evaluation.
However, you should still care to seek an assessment of how well your project has served its purpose. Has the reaction and consequence of the work been consistent with its intent and reason for being created, as we determined earlier in the process? It is important to recall the following terms of reference because they frame the type of feedback we seek:
Was there a positive reaction to the piece we created?
Did it deliver the appropriate tone of voice?
Did it reach the intended audience type and volume?
Were users able to effectively consume or discover insights?
Where we had a set idea of the intended consequences of this work, were they experienced?
What problems did people experience, if any?
To obtain feedback of this type and breadth we must consider multiple channels. Each of the following options provides an incremental level of evaluative value but consequently also requires a proportionate increase in the amount of time, effort, and probably cost to obtain:
Metrics and benchmarks: For web-based visualizations there are a number of easily obtainable measures to indicate the reach and popularity of your project. The traditional analytic measures for page views, visits, and visitors can now be easily supplemented with social media metrics such as Tweet counts, Facebook likes, Google+ shares, and so on. These are very simple, cheap, and accessible indicators to help you form a basic understanding of your design's utilization. What you need to think about is: what does success look like? What are the relative benchmarks of performance against these measures that will inform your overall satisfaction?
Client or customer feedback: Of course, the most tangible form of feedback for many projects will come from those who have asked or commissioned you (and hopefully paid you) to create the solution. You'll learn in no uncertain terms whether or not what you created fell short, matched, or exceeded their expectations. Sometimes, you have to judge yourself against the requirements outlined to you and not the resulting performance. After all, you can only respond to the brief you were given.
Peer review: Sometimes the most important and constructive evaluation can come from peers, perhaps expert practitioners or thought leaders. In the visualization field, there are many examples of bloggers who will conduct a review and critique of new work. Getting visitor hits is one thing but receiving a positive review and mention from a peer is worth its weight in gold.
Unstructured feedback: This type of evidence might come via online comments forms, reaction on social media, or through anecdotal channels (e-mails, in-person conversations, perhaps overheard comments) to add a layer of qualitative reaction and evidence of success or failure.
Invite user assessment: Rather than placing value on anecdotal or reactionary and opportune feedback, you could be more proactive by offering simple mechanisms for users to provide more structured qualitative responses, perhaps through small-scale questionnaires.
Formal case studies: Taking things to a more advanced level of evaluation (almost academic in its nature), case studies can take many forms using techniques such as interviews, observations, and controlled experiments, where you might set tasks, manipulate conditions, and record responses. These will often be undertaken by an independent observer to offer that degree of integrity.
No matter through which of these methods you obtain your evaluation feedback, you should be prepared for and accept criticism. Of course, in this digital age everyone is a critic—and too often anonymous—but you should always welcome constructive feedback and use this to fuel your development.
Finally, from your personal point of view, how effective did you think it went? Your own satisfaction is very important because this is what also drives your future decisions and development. Often we'll know best whether something could be considered to be an effective outcome and a satisfying process. Even if the results are very positive, there may be many things you thought could have gone better:
Did you accomplish the outcomes you wanted?
Did you create something you were satisfied with?
Were you satisfied with how you rationalized the choices?
Maybe you hated the project, the client, or the subject matter
Perhaps you spent far too long on the work and you haven't been paid or rewarded sufficiently for the time you invested
Maybe you regret consuming so much caffeine late at night
Try not to weigh yourself down with too many thoughts of regret around "wish I'd not done this" or "wish I'd managed to include that". Instead, put all reflection to best use as a learning experience to inform your development and preparedness for future opportunities.
Developing your capabilities
The project is over. You can take a deep breath and relax. Well, at least for an hour until your next project is lined up!
For you the bigger picture now is to consider your ongoing development in this discipline, learning from each experience, and building up your expertise.
The single most important message that I want to put across in this book is the value of practice, experience, and ongoing self-improvement. Data visualization is such a multidimensioned and rapidly evolving craft that cannot be mastered overnight.
Earlier we looked at the framework of the "eight hats". Through assessing yourself against this collection of capabilities, skills, and attitudes you can self-determine where your strengths and weaknesses may be and then look to address them. There are several strategies to help ensure your development continues.
Practice, practice, practice!
When it comes to developing your practical design skills the major piece of advice is simple—practice, practice, and more practice. There are so many different variables and subtle challenges involved in every project that you can't fail to learn from each project that you undertake.
We've just reinforced that data visualization is a craft. You need to continue to exercise your creative and analytical muscle to stay in good shape.
If time permits, try forcing yourself to stick to a practice agenda: maybe, do small personal projects every week then a bigger project every month. You might never launch the work in public but just testing yourself against the challenges of gathering data, analyzing, and presenting it will help maintain your development.
An especially ideal opportunity for practice exists through the frequent data visualization contests that are held these days, often with added incentive of prizes for the best-judged work. These typically involve a basic design brief, a published dataset, and a timeframe to create a compelling solution. A great value of these contests comes from seeing all the other solutions that are submitted. This lets you learn how others have tackled the same problem but maybe in different ways as compared to your own approach.
I have already stressed the importance of maintaining a written record of how you have tackled your design projects. It is worth repeating because it will really help you identify areas for improvement both in terms of effectiveness and efficiency. It will also be a useful reference guide should you ever need to take on a similar problem or comparable dataset.
Also, keep all your trash! Whether it is sketches on paper or little developments on the computer that you deemed redundant, where possible, keep them because you never know when they might come in useful.
Earlier in this article, we profiled the importance of technology and the potential limitations of your own capabilities. It is up to you to decide how far and the direction you may wish to take your technical skills. You may not always have the time or opportunity but if you are really serious about advancing your visualization design skills, you should try to push yourself beyond your comfort zone. Rather than relying on the same old tools, pushing them to do things that they're not really designed to do; try out new software, applications, and programming environments. Accept that there will be relatively steep learning curves involved but that the rewards could be great.
Evaluating the work of others
One of the most effective ways of sharpening your visualization design "eye" is by evaluating other designers' work. Not necessarily through providing formal feedback, but just testing your reaction and analysis of the designs you see.
Try to take on the dual mindset of a user and of a designer, in order to undertake a forensic assessment of what has been produced and how well it works using the following prompts:
What one word describes what is your immediate, instinctive reaction? Is it positive or negative sentiment?
If it is not necessarily an "instant" piece, does it have the qualities of a "slowburner", seemingly becoming more appealing after a certain duration?
What purpose do you think the designer had in mind? Does the style and function of the end product match the likely intention?
We rarely create these pieces in perfect project conditions, so consider what type of inherent factors might have surrounded and influenced this project? Does a sense of sympathy with the possible influencing factors of the design process effect your impression?
Work through the five design layers and ask yourself how well each has been executed and what improvements could have been made?
Also think about the general design considerations we outlined about creating accessibility through intuitive design, as well as the idea of reward versus effort, and see how these qualities are achieved.
Eventually, with enough practice, you will develop your critical eye and will become much faster, more informed, and fairer in your judgments of other peoples' work. This will be a great way to educate your own design techniques and refine your own style.
Publishing and sharing your output
One of the contemporary ways of developing your capabilities is to publish yourself. A platform such as a blog will create an ideal means of sharing your work and your ideas.
Posting your design work and building up a public portfolio of your projects creates a virtual shop window. You can take advantage of this format and share narratives about your design process, explaining to people how and why you arrived at the various solutions.
Writing articles, publishing critiques of work, and facilitating discussions are also great ways of promoting yourself. It helps you to learn about the subject. As you write about a topic, you are forced into developing a conviction, to structure arguments, and learn about different perspectives. It really does sharpen your views remarkably, even if (initially, at least) the only visitors to your site are your devoted parents, out of duty.
Eventually, through hard work and dedication, you will create an interested audience and this opens up wonderful opportunities for developing connections with other practitioners, creating rich networks around the world, across demographics, and beyond your subject field.
If you don't have the energy, time, or enthusiasm for a commitment such as a blog, then there are plenty of online galleries and communities through which you can share and publicize your work.
Immerse yourself into learning about the field
Over the past few years we have seen a relative explosion in the amount of online content covering the subject of data visualization, infographics, and data-driven journalism. Websites, blogs, designer sites, and online galleries are now bursting at the seams with interesting articles, new tools, latest projects, and endless amount of inspiration. Social media too is a wonderful platform to learn about key opinions and contemporary developments. The visualization field is particularly active on Twitter where you will find a very spirited and positive community.
Immersing yourself in the array of online resources will keep you up-to-date with contemporary developments. The following is a list of just a small selection of some of the best websites that you should visit and keep a track of. They have been loosely organized by their general remit, though many offer a wide variety of value:
Latest projects, trends, articles, announcements, and developments:
Visualising Data (http://www.visualisingdata.com/)
Information Aesthetics (http://infosthetics.com/)
Flowing Data (http://flowingdata.com/)
Discourse around data visualization:
Perceptual Edge (http://www.perceptualedge.com/blog/)
The Functional Art (http://www.thefunctionalart.com/)
Eager Eyes (http://eagereyes.org/)
Fell In Love With Data (http://fellinlovewithdata.com/)
Michael Babwahsingh (http://michaelbabwahsingh.com/)
Design narratives, process, and project critique:
Charts 'n Things (http://chartsnthings.tumblr.com/)
The Why Axis (http://thewhyaxis.info/)
Junk Charts (http://junkcharts.typepad.com/)
Graphic Sociology (http://thesocietypages.org/graphicsociology/)
National Geographic (http://juanvelascoblog.com/)
Design or technical tutorials, advice, and much more!:
Jérôme Cukier (http://www.jeromecukier.net/)
Jim Vallandingham (http://vallandingham.me/)
Gregor Aisch (http://vis4.net/blog/)
Naomi Robbins, Forbes (http://blogs.forbes.com/naomirobbins/)
Visualization communities, designers, design agencies and general inspiration:
Information is Beautiful Awards (http://www.informationisbeautifulawards.com/)
Any New York Times design (http://www.nytimes.com/)
Guardian datablog (http://www.guardian.co.uk/news/datablog)
Pitch Interactive (http://www.pitchinteractive.com/beta/index.php)
Moritz Stefaner (http://well-formed-data.net/)
Santiago Ortiz (http://moebio.com/)
Tulp Interactive (http://tulpinteractive.com/)
It should go without saying that an intimate appreciation of the many books about and around the subject is vital for learning this craft. Of course, you have already shown great wisdom in choosing this book but there are so many fascinating and invaluable titles to choose from. You can find a list of the most influential titles by visiting http://www.visualisingdata.com/index.php/resources/.
It is also important to expose yourself to influences from outside the specific boundaries of this field. You can pick up a great deal of inspiration from reading about graphic design, architecture, product design, typography, move-making, video game design, and journalism—all areas from which we can translate, transport ideas, and learn.
Another critical layer of learning comes from the world of academia and the value of keeping abreast of latest research and studies, from which many tools and best practices naturally emerge. The open access movement is gathering pace and making academic literature much more accessible to those not directly affiliated with academic institutions.
Conferences are also naturally a great way to keep in touch with the very latest developments, hearing from great speakers, and seeing inspirational presentations of case studies and examples. You also get to interact with other similarly passionate practitioners, something that can prove very rewarding.
Beyond these options, clearly a further avenue is through formal training and these days this comes in all shapes and sizes—from online tutorials, video tutorials, and webinars, through to in-person training courses and undergraduate or postgraduate programs. Scour the Internet to find the right solution for you.
Whichever way you go about developing your skills and knowledge, you can be sure there will be plenty of support from across the field. Data visualization is blessed with a wonderful positive and supportive community of incredibly talented and humble people, so you will always be met with a warm welcome.
We have now come to the end of this design journey. Hopefully, you got to your destination smoothly and you didn't experience many frights along the way!
In this article, we have focused on the execution stage of the visualization design process, bringing form to all your preparatory efforts, and transforming your concept into a produced work. We have introduced some of the most useful technologies to give you a flavor of the variety of tools being used for the different stages or types of visualization design.
We have looked at some of the important final steps to take before launching your design, the importance of running final checks across all elements, and conducting testing to get an evaluation of your solution before launch.
Once launched, it is then your prerogative to seek evidence of the impact of your work and we outlined a number of different tactics for undertaking this.
Finally, we suggested some strategies for you to consider pursuing to continue to develop your data visualization skills, knowledge, and experience. This will give you the best chance of taking your capabilities forward and achieving success in this thoroughly exciting field.
Good luck to all of you with the visualization challenges you take on in the future and thank you so much for taking the time to read this article. I hope it helps in any way possible!
Resources for Article :
- Designing Site Layouts in Inkscape [Article]
- Displaying SQL Server Data using a Linq Data Source [Article]
- New features in Domino Designer 8.5 [Article]
About the Author :
Andy Kirk is a freelance data visualization design consultant, training provider, and editor of the popular data visualization blog, visualisingdata.com.
After graduating from Lancaster University with a B.Sc. (Hons) degree in Operational Research, he spent over a decade at a number of the UK's largest organizations in a variety of business analysis and information management roles.
Late 2006 provided Andy with a career-changing "eureka" moment through the serendipitous discovery of data visualization and he has passionately pursued this subject ever since, completing an M.A. (with Distinction) at the University of Leeds along the way.
In February 2010, he launched visualisingdata.com with a mission to provide readers with inspiring insights into the contemporary techniques, resources, applications, and best practices around this increasingly popular field. His design consultancy work and training courses extend this ambition, helping organizations of all shapes, sizes, and industries to enhance the analysis and communication of their data to maximize impact.
This book aims to pass on some of the expertise Andy has built up over these years to provide readers with an informative and helpful guide to succeeding in the challenging but exciting world of data visualization design.