3 Ways Data Visualisation Improves Analytics Impact



Salesforce, a cloud computing service as a software (SaaS) company recently acquired Tableau, a data visualisation company for US$15.7Bn. That is a huge sum for a software company bringing together the world's #1 CRM company with the world's #1 analytics platform. We are in an inherently visual world, where images speak louder than words.


Data visualisation is the graphical representation of information and data. There are many ways to visualise data such as using charts, tables and maps. Data visualization tools and technologies are essential to analyze massive amounts of information and make data-driven decisions, especially in the world of big data.


"The greatest value of a picture is when it forces us to notice what we never expected to see" - John W. Tukey

Data visualisation increases the value obtained from data science by increasing the effectiveness of communicating the overall picture to stakeholders. It is definitely an important part of data science.


 

1) Makes big and small data easier for the human brain to understand, therefore helps you begin to ask the right questions


Data visualisation, in large part, helps create a compelling narrative. You’ll easily be able to look at a large amount of information at once and hopefully see trends (internal and potentially combining with external) that can then be applied to improve the success of your business.


Below is an example of data table vs graph to communicate insights:


Data analytics, big data, data graph, data visualisation, data visualization
Data Visualisation - Table vs Graph (Source: Forbes)

With the graph, immediate patterns can be seen and much easier to understand at a glance. We can deduce that the higher the obtained qualification, there is a lower unemployment rate and a higher median usual weekly earnings.

Key decision-makers aren’t always experts in every activity of the company, but with the right data visualisation comparisons, leaders are able to make quicker, more informed decisions.


It allows the right questions to be asked. It identifies relationships and patterns within digital assets, and increases curiousity on what happened. It helps challenge the status-quo to bring expansion or improvements.


An example will be to understand why sales has declined in a strong market share area. Discerning trends in the data can help us solve problems, and lead the company to perform better.


 

2) Identifies outliers quickly


Detecting outliers or anomalies is one of the core problems in data mining. Data visualization quickly reveals the outliers in data. You can get a sense of the overall distribution and identify what doesn’t belong.


Data visualisation, data visualization, data outlier, data analytics
Example of Outlier in a Frequency Chart (Source: Brendan Gregg)

Example above shows outliers present in the graph. The value is too high and improbable. However, it can also be noted that there are better ways to represent the outliers such as using a density plot so that outliers can be identified more clearly.


The next step is to understand why outliers are there. This step is important as it may tell another important underlying problem. Sometimes, you may need to rectify the cause of the outlier, such as a faulty recording device. This step can help improve your data processes and prevent outliers from happening in future to save effort on data cleaning.


It is important to always deal with outliers during analysis. Otherwise, it can skew results significantly, resulting in the wrong decision being made.


 

3) Provides immediate direction and makes the data more memorable for stakeholders


Visualisation makes data more impactful and memorable. A study by World Bank found that visualisation can be quickly taken in and understood (and with little eye movement). It also appears to be easiest to recall later on.


Additionally, the clearer, more succinct the message, the more memorable the results will be.


It therefore amplifies your messaging to internal (and external) stakeholders, more easily uniting everyone towards a direction. In other words, it helps achieve overarching business goals.


"It’s not what you look at that matters, it’s what you see" - Henry David Thoreau

However, it should be noted that it is important to choose the right charts and visualisation. Otherwise, choosing the wrong data visualisation might overwhelm or confuse your audience – achieving the opposite of your intended result.


 

With data visualisation, we can get clarity of answer to the problem very quickly. It allows discovery and makes the message from data analysis more impactful. But it is important to note that improper data visualisation taints decision-making and muddles message.


It’s hard to think of a professional industry that doesn’t benefit from data visualisation. It is now a must-have for industries. Big data is especially the primary force behind this change.



What are challenges that your company faces when communicating data? Share with us by leaving a comment. Start using more accurate data visualisation to have a better impact. If you require assistance on data visualisation, contact us. Subscribe to our newsletter for regular feeds.


Did you find this blog post helpful? Share the post! Have feedback or other ideas? We'd love to hear from you.


 

References


Forbes, Data Visualization - How to Tell a Story with Data, https://www.forbes.com/sites/nicolemartin1/2018/11/01/data-visualization-how-to-tell-a-story-with-data/#176621f64368, published 1 November 2018

World Bank, What Makes Data Visualisation Memorable, https://blogs.worldbank.org/opendata/what-makes-data-visualization-memorable, published 5 November 2015


27 views0 comments