Configuring A New Visualization¶
All visualizations require you to have data connected to your workspace. As soon as you've connected a dataset, you can start creating visualizations. Every visualization requires you to configure how data is aggregated, grouped, filtered, and displayed, this is where the configuration panel comes in. The configuration panel provides you systematic way of specifying the way you want your data to be displayed.
Selecting Your Key Metrics (Y-Axis)¶
Selecting the column¶
The first step for any new visualization is to select the key metrics that you want to visualize, which you can do under the Y-Axis section on the Configuration Panel.
Y-Axis Settings for Table Views
Even if you plan on visualizing your data in a table, the first step is to set your Y-Axis column. You can think of the name "Y-Axis" as a synonym for "KPI".
Your Y-Axis metrics must be columns in a connected dataset. They can be any type of column (numeric, text, date, etc) but you'll have the most flexibility with numeric columns. With non-numeric columns, all you'll be able to do is aggregate them by counting them. But with numeric columns, you can aggregate them in a variety of different ways.
Advanced metric settings panel¶
When you first select your metrics, they will be displayed in the Configuration Panel under the Y-Axis settings section. Next to the name of your columns, you'll see a white pill with the word "SUM" (this is the default aggregation function that will be used on your data, described below).
When you click on this pill, you'll open up a panel that will let you transform your series data and name. Below you'll find a description of the different settings you can provide in this panel.
Your data will be aggregated based on the aggregation methodology you choose. Data will be aggregated by groups based on what you select for the X-Axis and Grouping fields (defined below). If you don't specify anything for these settings, your data will be aggregated in whole. The following aggregation methods are currently available for selection:
- SUM: This is the default aggregation method, it will sum all of the values in your Y-Axis metric
- COUNT: This method will count the total number of values in the Y-Axis column you select
- AVERAGE: This method will take the average value of your data, missing values will be excluded
- MIN: This will result in the minimum value in your data
- MAX: This will result in the maximum value in your data
- NONE: Selecting this method will perform no aggregations on your data. If you select this, you cannot select a grouping (since, if you select a group, it will result in multiple values for a single point on the X-Axis, and there's no straightforward way to visualize all those values)
Analytics functions allow you to understand how your numerical data is changing over time. You can optionally select to run an analytics function to understand how the values of your data are changing. You can choose from the following analytics functions:
- CUMULATIVE SUM: This function will keep a running total of your key metric and display that running total over each X-Axis or Grouping value. If all of your key metric values are positive, this will result in a graph that goes up and to the right (i.e. a monotonically increasing graph)
- % GROWTH: This function allows you to see how your values are changing from one period of time to the next. If you opt to use the % Growth function, you'll need to specify a column in the X-Axis or Grouping fields that is a date or a time, and you'll also need to select the periodicity that you want to use to calculate your changes over time
You can also optionally transform your columns using simple arithmetic. You can add, subtract, multiply, or divide the values in your numerical columns by a number that you specify.
Metric display name¶
At the bottom of the advanced metric settings panel, there's a spot to show the display name for the data that will be graphed. This display name will be used in the legend and in tooltips. If you leave this blank, a default name will be used based on the aggregation and column name that you select.
Configuring the X-Axis¶
After you configure your key metric, you have the option of specifying how your data should be displayed on the horizontal axis (X-Axis). By default, if you don't specify anything for the X-Axis, you'll only have one data point - the total aggregated value across your data.
Specifying the X-Axis lets you display your data over time as a time series or break down your data into groups whose names will be displayed along the X-Axis in your results.
When you select a column that has date or time values, you'll have the option to open up an advanced settings panel that will allow you to specify how to display your dates. In order to open this panel, you'll need to click on the white pill labeled "Date", which is to the left of the name of the X-Axis column you've selected.
The Pill Will Only Appear For Columns With Date Values
The pill will not appear unless the data in your column has been recognized as a date.
Configuring the date display¶
When you open up the date grouping panel, you'll have the option to display your data aggregated across different granularities of time. The default option is "Date", which will aggregate and display your data by unique date. Similarly, if you choose to aggregate and display your data by month, or by year, your data will be aggregated and displayed using those settings.
Many datasets frequently have a column that's used to track what group a specific row belongs to. For example, a dataset tracking wine quality could have a column to track if a wine is red or white, a dataset tracking company headquarters could have a column to track the state or country, etc.
When you want to break down data by one of these attributes, you use can specify the column in the "Grouping" section. This section acts similarly to the X-Axis section above, but it allows you to specify multiple columns to create deeper levels of grouping.
Whenever a grouping column contains dates, you'll be able to specify the date granularity just like you did with the X-Axis column.
In the Quickstart Guide we showed you an example of how to break down avocado prices by type (organic vs. conventional).
In many situations, you won't want to visualize all of the data in your dataset. Instead, you may want to only use a subset, breaking it down and visualizing it based on the values that exist in your data.
Filters allow you to visualize a subset of your data based on the filter values that you specify. The following is a list of filters that are available on the Apteo platform, along with details on how to specify their associated values.
- Less than: This filter will allow you to graph data where the column's value is less than the value that you specify in the filter value input text box
- Less than or equal to: This will allow you to graph data where the column's value is less than or equal to the value that you specify in the filter value input text box
- Equals: This is the default filter setting. It allows you to select a subset of your data that exactly matches the value you provide. For filtering on text, do not use quotes.
- Does not equal: The converse of "Equals" - this will only use data that does not exactly match the value you provide
- Contains text: This will match any column value that contains the value that you specify. Do not specify regular expression or matching parameters like %.
- Does not contain text: The converse of "Contains text" - this will only use data that does not contain the value you specify
- Greater than or equal to: This will only use data where the value in the column is greater than or equal to the value you specify
- Greater than: This will only use data where the value in the column is greater than the value you specify
- In list: This will allow you specify a comma-separated list of values, one of which must match the column's value, in order for your data to be included in the visualization. For text, do not include quotes, regular expression, or matching parameters
- Not in list: The converse of "In list" - the data used in your visualization will be included only when it does not match a value in the list