When evaluating the ‘value’ of data, it is often hard to place an exact monetary amount. Given the variety of data that is now available, the value of one dataset might – and often is – be far greater for one company than it is for another.
This is because data is only as valuable as one’s understanding of it and the insights that can be gained from it.
For this reason, data visualisation has emerged as a valuable tool to help businesses and individuals extract value from data.
Data visualisation is essentially exactly what it sounds like. It is the process of taking raw data and transforming it into graphs, charts, images and videos. These visualisations can be used to ‘paint a picture’ with the data and highlight certain trends that would otherwise be missed by other methods. There’s also an emotional element when it comes to data visualisation. Visual prompts are the strongest form of communication and can serve a purpose when attempting to communicate a complex fact in an obvious way.
Data visualisation is all about making the right decision with how you represent data. A bar plot and a linear plot can both be used for the same dataset – so how do you know which one is right for you.
The PLOS Computational Biology journal published the 10 ‘rules’ of data visualisation in 2014. These include knowing your audience, identifying your message and not misleading the reader. The rules also suggest that the message a visualisation tells will always be more important than the overall aesthetic – or ‘beauty’ – of a graph and also reminds people to use the right tools when creating visualisations.
Another golden rule with any visualisation is to avoid confusing correlation and causation. A reader can be easily misled by a visualisation if correlations are displayed, rather than causations. Although two datasets may visually look similar on the same graphic, it is impossible to legitimately show a cause-and-effect relationship between two variables based simply off a correlation.
By following data visualisation best practices, data scientists around the world have been able to tell major stories that would not have otherwise been conveyed as effectively.
There is perhaps no better example of this than the 2016 Oscars, when Bloomberg published an interactive data visualisation showing the race, gender, age and hair colour of the previous 87 years of Academy Awards Best Actor and Actress winners. The data visualisation came as the public started to criticise the Academy for ‘white washing’ when it came to selecting winners. For many, this visualisation – which showed just eight non-white winners – was a powerful tool in depicting the #OscarsSoWhite movement, which has grown in prominence in recent years.
BBC Sport also recently managed to create a simple but effective data visualisation, which told a story words could not. With England hosting the ICC Cricket World Cup in 2019, BBC wanted to dispel the misconception that all cricket grounds are the same size. Using just Google Maps data, the broadcaster was able to tell the story of each ground and highlight the nuances in size and shape which give each ground a sense of character.
Businesses should endeavour to use data visualisation with this same purpose – to tell stories that would have been difficult to convey otherwise.
As the cricket ground and Oscars case studies show, sometimes simple datasets make the best visualisations. At smrtr, we can help you clean and simplify your data so that it is in shape the create these sorts of powerful visualisations.
By Boris Guennewig, Co Founder & CTO at smrtr