Why Data Visualisation Matters
Monday, 18th April 2016
People love pictures. And pictures help us recall information when we need it. The reason is simple enough, the human brain is wired for pattern recognition. But when it comes to interpreting the relationship between data sets, or understanding complex statistical notions, most of us quickly waiver.
For marketers in particular this is a problem. The past decade has seen an explosion of new data sources, typically unstructured such as social data and increasingly owned by other people.
In addition, the companies where marketers ply their trade are getting better at breaking down internal silos and plugging into partner organisations through APIs, broadening the data pool further.
But with all this information about customers and markets pouring into the company like water from a fire hose, there is more noise than ever to obfuscate the signals that actually matter.
That’s why visualisation and the tools around it have become so important. Once the preserve of data and hard core data scientists, software is pushing visualisation capabilities out into the hands of business users, allowing them to at least start the discovery journey themselves before eventually handing off for more technical analysts when the job gets too complex.
Gartner’s Martin Kihn says that even with the best data, marketers often have trouble convincing peers and managers to make decisions.
In Tell a compelling story using data visualisation principles for marketing, Kihn says that marketing analysts can motivate their peers to make more effective decisions by presenting data visually. “Much of the challenge lies in the nature of the data itself. It’s abstract. You can’t understand it without scrutinising it and thinking about it,” he says.
We all know the adage that ‘a picture is worth a thousand words’ – and this holds particularly true in those business contexts where managers need to make a clear point, quickly and with impact. “A visualisation needs to tell a story. Visual interest alone does not make for a compelling presentation of data. Static data, statistics alone – will not persuade anyone,” says Kihn.
Presenting information visually isn’t new – humans have been using pictures to communicate important information literally since the dawn of time.
Visualisation matters because humans are hard-wired to engage with visual stimuli and have strong visual pattern recognition skills, but many of us have proportionally poorer verbal pattern recognition – or more to the point, our short term memories can’t hold enough chunks of verbal information sequentially to process complex verbal patterns without much repetition.
Therefore, if you need to communicate a complex message, or make a point strongly, presenting it visually can be a more effective way to get both audience attention and understanding.
The greatest advantage of visualisation for both executives and data practitioners is that it provides a medium that they both can interpret readily – a common language if you like.
Visualisation helps data practitioners translate the complex code, data structures and rules they work with into something that business leaders can see and evaluate – a tangible output that they can then use to drive business decisions. So in terms of encouraging data-driven decision making, visualisation has been a hugely powerful tool.
IBM’s Ian Wong, partner IBM Interactive Experience says a key factor in the evolution of data visualisation is the use of design thinking to get in the mindset of the end user and fully understand their requirements. “There is much more consideration for the person who is consuming the data and insights, and the context within which the data is being consumed.
This has led to the realisation that increasingly, data is being consumed and visualised in an omni-channel manner, leading to a mobile first approach to data visualisation.”
“We’re now seeing data being consumed in an increasingly mobile sense, where the same person might consume the same set of information multiple times a day across different devices. Visualisations therefore needs to present the data in a responsive and adaptive way, and needs to make the visualisation available in the right way at the right time, leading to drive to simple elegance in the way data is presented,” says Wong.
And all of this of course at a time when we have seen the emergence of a greater variety of sources of data. “Previously, it was typical for organisations to use structured, or internal, data only. Now to have a holistic view, organisations are incorporating external, unstructured data such as social media, which can reveal sentiment and trends, and also things like the weather, which can impact on consumer behaviour and business operations.”
Due to increasing information sources, cognitive and analytics is then needed to understand all of the data. It allows analysts to make correlations and action points for the end user in a much quicker fashion, pairing unstructured and structured data in a meaningful way, he says.
There has definitely been an explosion of interest in visualisation over the past few years, although I think we are a long way yet from seeing a comparable explosion in expertise at delivering great visual stories.
In terms of the evolution in what is possible, there has been considerable expansion in the range of ‘pre-canned’ visualisation options over the past two to three years, which makes visualisation tools more accessible for nontechnical users, says Wong. There is now a much wider array of chart types, increased animation options, and much stronger geospatial mapping capabilities, the latter of which has had a really strong impact on how much traction visualisation gets in many consumer facing sectors.
Because the tools are becoming more ubiquitous and easier to use, there’s a much broader range of users starting to explore data visualisation and this in turn starts to push the boundaries of what tools can do. According to Wong, “In particular we have seen some amazing visualisation from creative users who have an innate sense of visual design and are learning about data (or working in tandem with an analyst).”
“As the visualisation market grows, there are more free or low cost subscription based tools available, and this also opens up access to a wider pool of users, most specifically students, who can then enter the workforce with a stronger blend of design and data skills.”
Walter Adamson, general manager, KINSHIP digital Victoria describes the evolution of the visualisation space over the past three years as dramatic. This has been driven by the trend to put more data visualisation capability directly in the hands of end users he says. “This is directly observable in the marketing messages of key data visualisation vendors who all emphasis that users need no longer need to rely on IT in order to produce high quality charts and reports.
In pursuing this goal vendors have been aided by the whole changing world of IT in the “digital era” where users expect access to flexible cloud-based systems, and where the amount of data being collected and needing analysing is increasing rapidly. In marketing alone the need to track engagement and interactions across web and social and to relate the analytics to sales and “ROI creates a tremendous amount of data which requires analysis by those with domain experience. This is where end-users are clamouring to apply easy-to-use tools which provide easy and flexible reporting of the data, and vendors are obliging.”
In terms of visualisation as a discipline this trend is rapidly changing expectations and demands. According to Adamson, it is enabling an enormous number of non-specialists to do very effective visualisation, while at same time highlighting that many visualisations are more buzz than insight. As people experience the latter they are seeking more training and a better basic understanding of good visualisation principles and practice. The pull for the latter is going through the roof.
“Because companies are becoming much more aware of the value of data because of the rise of the “data analytics companies” such as Amazon and Umber, the impact of data is being felt, or is felt being missed, across many parts of the organisation. The old approach of Business Intelligence as an IT project is incapable of serving the new needs and demands. The demand to put data analytics and visualisation in the hands of users is the story of the last three years.”