Best practices for choosing visualization types
A good visualization clearly presents relationships among many values, and lets you analyze data at a glance. Qlik Cloud Analytics offers a range of Qlik Sense visualizations, charts, and analyses. Each chart and analysis excels at visualizing data in different ways for different purposes. You should select your charts and analyses by deciding what you want to see from the data in your charts.
If you are unsure of what visualizations to use, Qlik Sense can offer recommendations. You can use analyses to start with the type of analysis you want to make rather than starting with a chart. For more information, see Creating and editing analyses. Insight Advisor can also help you create visualizations. For more information, see Creating visualizations with Insight Advisor and Creating visualizations using Insight Advisor chart suggestions.
The following lists the purposes for viewing data and the chart and analysis type recommended to achieve that purpose:
Viewing comparisons
Comparison charts are used to compare values against each other. They show the differences in values, such as the difference between categories, or how values are changing over time.
Questions that might be answered by comparison charts include:
- What product has the highest total sales this year?
- How have product sales risen or fallen over the last 24 months?
| Chart type | Common purpose |
|---|---|
| Bar chart | Comparing categories against the same measure or measures. |
| Line chart | Comparing trends over time. |
| Combo chart | Comparing measures that are different in scale. |
The following analyses are useful for creating charts for viewing comparisons:
| Analysis type | Common purpose |
|---|---|
| Comparison analysis | Show multiple measures for a dimension. |
| Ranking analysis | Show dimensions in the order of their contribution to a measure. |
| Ranking (grouped) analysis | Show hierarchical dimensions in order of their contribution to a measure. |
| Trend over time analysis | Show the performance of a measure over time, optionally broken down by a dimension. |
If you want to compare different values of the same dimension, you can use alternate states. For more information, see Using alternate states for comparative analysis.
Viewing relationships
Relationship charts are used to explore how values relate to each other. A relationship chart allows you to find correlations, outliers, and clusters of data.
Questions that might be answered by relationship charts include:
- Is there a correlation between advertising spending and sales for our products?
- How do expenses and income vary per region?
| Chart type | Common purpose |
|---|---|
| Scatter plot | Viewing the relationship between two or three measures for a dimension. |
The following analyses are useful for creating charts for viewing relationships:
| Analysis type | Common purpose |
|---|---|
| Clustering (k-means) analysis | Show clusters of measures associated with a dimension using a statistical algorithm. |
| Correlation analysis | Show the strength of the relationship between two fields. |
| Mutual information analysis | Detect and show dependencies between the source and driver fields. |
| Relative importance analysis | Show the relative importance of dimensions contributing to a total. |
Viewing compositions
Composition charts take a total value and discover what component values make up that total. Composition charts can be static, showing the current composition of a total value, or they can show how the composition of a total value changes over time. Composition charts can display compositions either by percentage of the total value or the fixed values in the total value.
Questions that might be answered by composition charts include:
- What percentages of our total sales come from which regions?
- What is each department's allotment of our total quarterly budget over the past year?
| Chart type | Common purpose |
|---|---|
| Bar chart | Viewing the changing composition of a value over a short period of time. |
| Line chart | Viewing the changing composition of a value over a long period of time. |
| Pie chart and donut chart | Viewing the static composition of a value. |
| Waterfall chart | Viewing the static composition of a value with accumulation or subtraction to the total. |
| Treemap | Viewing the static composition of a value's accumulation to the total. |
The following analyses are useful for creating charts for viewing compositions:
| Analysis type | Common purpose |
|---|---|
| Breakdown analysis | Show multiple dimensions in order of contribution. |
| Time series decomposition analysis | Decompose a time series into trend, seasonal, and residual components. |
| Trend over time analysis | Show the performance of a measure over time, optionally broken down by a dimension. |
| Trend with forecast analysis | Show measures along with forecast over the current and future time periods. |
| Year to date analysis | Show a comparison of dimensions for the same period in previous years. |
Viewing distributions
Distribution charts are used to explore how the values within data are grouped. Distribution charts show you the shape of your data, the range of its values, and possible outliers.
Questions that might be answered by distribution charts include:
- What is the number of customers per age group?
- What cities have the highest use of our services?
| Chart type | Common purpose |
|---|---|
| Histogram | Viewing the how data is distributed over intervals. |
| Scatter plot | Viewing the distribution of two measures. |
| Distribution plot | Viewing the distribution of measure values in a dimension. |
| Box plot | Viewing the range and distribution of numerical data. |
The following analyses are useful for creating charts for viewing distributions:
| Analysis type | Common purpose |
|---|---|
| Overview analysis | Show the distribution of measures for one or more dimensions. |
Viewing performances
Performance charts provide a quick view of a performance measure. Looking at a performance chart, a user can quickly identify the measure value and whether the results are as expected or not.
Questions that might be answered by performance charts include:
- What are the current total sales for this quarter?
- Are current total sales for this quarter meeting the projected sales for the quarter?
- How is performance for this product line compared with other product lines?
| Chart type | Common purpose |
|---|---|
| Bullet chart | Comparing performance of a measure for several dimensions. |
| Gauge | Viewing a performance value to understand performance immediately. |
| KPI | Viewing one or two performance measures. |
| Text & image | Viewing text or several measures with an image. |
The following analyses are useful for creating charts for viewing performances:
| Analysis type | Common purpose |
|---|---|
| Anomaly (spike) analysis | Detect and show large data variations including spikes and dips in a time series. |
| Anomaly (trend) analysis | Detect and show abrupt data variations including change points between time series segments. |
| Clustering (k-means) analysis | Show clusters of measures associated with a dimension using a statistical algorithm. |
| Process control (mean) analysis | Show measures over a time period compared with the overall mean of expected values. |
| Process control (rolling mean) analysis | Show the performance of a measure over time between two calculated control limits. |
Viewing data
Data charts present detailed data rather than a visualization of the data. Data charts are useful when you need to view precise values, and when you want to compare individual values.
Questions that might be answered by data charts include:
- What are the records for each transaction for this month?
- What are the quantity and sales for each item in each product group for each of our customers?
| Chart type | Common purpose |
|---|---|
| Table | Viewing precise values from your data without trends or patterns. |
| Pivot table | Viewing precise value for several dimensions and measures. |
Viewing geography
Geographical charts let you visualize your data by geography, displaying your data on a map either as points or areas.
Common questions that might be answered by geographical charts include:
- What cities have the highest use of our services?
- Which countries have the most customers?
| Chart type | Common purpose |
|---|---|
| Map chart | Viewing data represented geographically by point or area. |
The following analyses are useful for creating charts for viewing geography:
| Analysis type | Common purpose |
|---|---|
| Breakdown (geospatial) analysis | Show geographical contributions to a measure. |
What if no standard chart suits my purpose?
You can use controls or objects from a bundle supplied by Qlik:
You can also create custom visualization objects if none of the standard charts provided fit your requirements for visualizing your data.
For more information, see Creating a visualization using a custom object.