Marketers, news agencies or entrepreneurs like Amazon founder Jeff Bezos have been talking about it for a long time and doing it — they avoid PowerPoint, bullet points and dusty spreadsheets and tell stories instead.
Anyone who analyses data has a purpose. You want to draw insights from this, see the context and the big picture, and derive strategic steps from this. If you visualize your data instead of looking at it as a table like an anthill, you can create new connections on the screen, highlight the most exciting KPI's and tell something — because good visualizations tell a story and these have always fascinated people, are much more memorable and understandable than “naked” data and can answer many questions. After all, our brains can capture images 60,000 times faster than text.
But which visualization fits which question?
The classic bar chart is probably one of the most frequently used visualizations and is particularly suitable for comparing data from different categories and presenting their similarities and differences. Bar charts also provide a quick overview of highs and lows over time.
Whether it's the dynamics of share prices or access to the company site — the line chart connects specific data points and shows changes over a certain period of time.
Although proportions can be clearly presented in a pie chart with the partial values of a sum — this alone usually offers only a few new insights and is much slower for our brain to understand than bars, for example. However, if there are only two or three partial values, it can definitely make sense as a supplementary stylistic device on a dashboard.
If you want to visualize data that is linked to geographical information, the classic map is an ideal format. Geocoded data is quick and easy to display — whether buyers by zip code, companies by country, or other connections between locations and your data. Additional filter options make analysis efficient and accurate.
If a relatively large amount of data falls on a small location, density maps can be particularly beneficial. They show data concentrations and patterns that may be obscured by other visualizations due to overlaps. With the help of this type of visualization, geographical locations can be easily compared based on their data density.
Between which variables can a connection be identified, which change independently? With a scatterchart, the ratios of key figures, such as the health status of women compared to men or the buying behavior of students and retirees, can be efficiently analyzed.
Where is the project? Which steps have already been completed? Where is there a bottleneck of resources? Time series data is very easy to illustrate in a Gantt chart — it shows the progress of projects and other processes. This also makes it easy to see which goals have already been achieved and which have not yet been achieved.
Strictly speaking, bubbles are not a standalone type of visualization. However, they are particularly suitable for clarifying the relationship between data concentrations of several (3+) key figures, for example as a supplementary element in maps or scattercharts.
If you want to relate different parts of your data to the big picture, tree maps can be very helpful. By arranging rectangles of different sizes and subordinate branches, categories can be easily compared with each other, depending on the proportion of their data points.
Viz Credits: Tableau Public
If it hasn't already been tackled, it's often at the top of the to-do list — using data and creating added value. At the same time, new data is constantly being generated back and forth.
Read more
What are we actually talking about here?
Read more
