How data is presented is an important for managers that need a bird's eye-view of their information in their Business Intelligence platforms. In particular, it's important to choose the right kind of graphical form or representation for each information provided.
For example, to get an idea of the differences in population density in a region, it's more intuitive to use a map with coloured areas that increase in colour intensity. On the other hand, if you are a consumer company that wants to highlight which products generate the greatest turnover for a company, it's best to use a pie chart. (Want more in-depth information on Pentaho? Download our free guide Pentaho from A-Z now!)
The same data can be displayed differently depending on the characteristics to be analyzed such as performance of a given value over time, subdivision of a total value, placement of information within a certain location and so on. For each aspect to be analyzed, there are specific visualization techniques that will allow you to have a more immediate understanding of your data. Here are the main ways to analyze a dataset and which graph or visualization is best used for it based on Pentaho data visualization:
It's how a certain piece of data develops or changes over time. In this case, we would be interested in the value reached at a given moment, the before and after comparison of the values, and a bird's eye view on how the figures change over time.
Using a line chart is the best way to present this information.
2. Positioning (ranking)
In this case, it's quite interesting to compare a set of values that have been sorted by order from biggest to smallest or vice versa.
The best way to reflect this information is through the bar chart with ranked values. This allows you to quickly locate the largest value up to the smallest and compare each value with others.
3. Subdivision of a whole into its parts
Here, we have a total value obtained as the sum of partial values.What's interesting here is to see what extent certain values contribute to the total. Percentages are usually used to represent the partial values.
The most commonly used way to present this information is through a pie charts and bar charts. Bar charts, or specifically stacked bar charts are used to represent this information to present the variation of the total and the contribution of its parts over time.
Each bar represents the total value at a specific time and is subdivided into its components that form part of its total.
A set of data is compared with a reference value (target value). It quickly shows the difference between each of the actual data and the target value and compares it with other deviations of other data to identify in which areas are they reaching or deviating from their targets the most.
In the case where the target value is the same for all data, the most suitable way to present it is through a bar chart. And if the target value changes for each of the data, then you can use a graphical bullet graph.
5. Frequency distribution
It's a counter for data that falls into certain categories or ranges. The most widely used type of graph is the histogram which is a variant of the bar graph. Each element represents a category or range of values instead of just a single value.
It details two aspects of a set of data. Here, it's interesting to see if and how they change in relation to the variation of the others. The graph most appropriate for this is the correlation diagram (can also be scatter diagram or scatter plot) where data is represented as points with cartesian coordinates that correspond to the two values taken into consideration.
A line that tends to have the lowest average distance from the points shows the degree of correlation.
It's often necessary to represent more series of the data to be able to compare individual values or a general trend. In these cases, it's best to use line graphs or grouped bar charts. The lines or bars represent the values of the different series of data and are differentiated by colours or patterns (solid, dashed, etc.)
In the case of data being referred to specific points within a geographical location, it's possible to present them on a map and reference them to their locations exactly. You can also reflect the values' differences in relation to each other either through differentiated colours or shape sizes. The most appropriate type of graph for this information is called a cartogram.
Business Intelligence and Data Analytics solutions provide more advanced tools to allow you to build custom dashboard that have different kinds of charts that we previously described.
One of the most comprehensive solutions for managing Business Intelligence processes is the Open Source Business Intelligence platform Pentaho Business Analytics Suite. Among its many components, you can find tools to show information in different graphs such as Pentaho Data Visualization.
Pentaho Data Visualization with Pentaho CDE
The main tool for the graphical display of data in Pentaho's community version is the CDE (Community Dashboard Editor) which allows you to make highly customizable graphs and pie charts. Graphs and charts are made through the CCC (Community Chart Components) library while most of the pie charts are run by components which incorporate 3rd-party mapping tools (Google Maps, Open Street Map, etc.). These tools offer the ability to customize them for data representation.
Extra has developed Business Intelligence solutions using Pentaho Data Visualization especially with Pentaho CDE for many customers in different industries in both the public and private sector. Our extensive experience has given us the ability to fully exploit the potential offered by Pentaho, choosing from time to time the best components for optimal data visualization.
Case studies of Graphical Presentation of Information with Pentaho Data Visualization
By using the Open Source Business Intelligence platform Pentaho Business Analytics Suite, you can have the best solution that suits your organization's needs. Below are some of the experiences of our customers.
For the government of the Autonomous Region of Sardinia in Italy, we developed a dashboard for monitoring the tourism trends that combine some of the previously described visualization techniques. The presence of tourists in the area were reflected through a series of pie charts that allowed you to cascade through their distribution in the various provinces, communes (geolocation by area) and even individual establishments (exact geolocation).
You can further analyze the presence of tourists by their origin, sex, composition of the family unit, time of the year and other factors that are important in understanding the kind of users of each tourism establishment and their habits. The information is particularly helpful for entrepreneurs and industry managers in directing their promotional campaigns and investments for seasons ahead.
For our customer, Morgan Tecnica, a producer of machinery for the packaging of clothing, we developed a dashboard that can track the performance of its customer installations. The parameters monitored included the productivity of machines vs. the optimal reference target (using a bullet graph) as well as the relationship between the work and break times of the machines, specifically highlighting critical cases where the machine operated for less than 35% of the operating time. For this we used a series of stacked bar charts combined with a threshold value that was the same for all machines.
As you can see, Pentaho is the perfect on-demand tool to access and analyze information regardless of industry and is the perfect partner for the daily struggle against time to surpass your competition. You just seen some of the many ways to better present data with Pentaho. To find out more, download our ebook Pentaho from A-Z by clicking the button below!