How to add a target line to the chart?
Target options typically encompass a static value, like median, average (mean), maximum, minimum, and range. They can be added to the following Visualization types:
- Column charts
- Bar charts
- Line charts
- Area charts
- Scatterplot charts
To add them, go to the Edit menu within the Visualizations section and select at least 1 dimension and 1 measure. Choose one of the charts available for Targets. Then choose the Y section in the Edit menu and click on the “Add reference line” button.
Then specify the details of the Y axis. Choose the type, value and label. You can also adjust the position and choose the color of the line.
How to add a trend line?
Trend lines show the prevailing direction of the data points moving over time. Not all charts can show trends, but Looker helps you identify those that can. If your chart supports a trend line, a button labeled "Add Trend Line" will appear in the Y-axis menu options.
Adding trend lines is in fact very similar to adding target lines. First, we need to start with the Edit menu within the Visualizations section. Then select at least 1 dimension and 1 measure. Choose one of the charts available for Targets. Edit the chat and go to the Y tab. Click on the “Add Trend Line” button.
Continue editing. Choose the trend type, label type and position, value, and the color. Your trend line is ready!
Important: Be sure to use the proper type of chart as trend lines are not applicable to all of them. Trend lines are not supported for the following charts: with a Stacked or Stacked Percentage series positioning type.
How to add a forecast?
ChurnIQ Studio goes beyond just showing past data. It lets broadcasters add forecasts directly to their visualizations. This helps them predict future trends and track how specific data points might change over time. These forecasts can then be integrated into dashboards and saved for easy reference, making them a valuable tool for informed decision-making.
Important: The displayed data of a forecast include only the results from Explore’s query.
Generating results
ARIMA (AutoRegressive Integrated Moving Average) is the algorithm responsible for generating the forecast results based on the data in the Explore data table.
These forecasts can easily integrate with the existing data and can be represented as their continuation whether is a graph or a table.
The visualization formats that can be used to display forecasts include:
- Column
- Bar
- Scatterplot
- Area
- Line
- Table
- Table legacy
- Single record
- Wordcloud
Important: To create a forecast, Explore must meet specific requirements, including having a timeframe dimension and at least one measure or custom measure sorted by the time frame dimension in descending order.
Select a time dimension and at least 1 measure.
Click on the Run tab.
Select Forecast tab. Here, you can choose:
- field to be forecasted (max 5),
- length of predicted “future (if months selected, and “12” applied, then 12 next months will be forecasted)
Choose Prediction Interval. This setting allows you to define the confidence level for the prediction given by the algorithm.
Set Seasonality to ensure the forecast takes into account known cycles and recurring patterns within your data.
Example: You are a sport broadcaster offering seasonal subscriptions for football games. You want to create a forecast for the revenue for the next three months (during the planned football season).
Start by pressing the Forecast button in the Visualizations menu.
Then select the Revenue measure and enter “3 (weeks)” in the Length tab.
Choose the Prediction Interval for 99% to get a more precise result.
Add seasonality, as in the case of your offer you may expect visible subscriber fluctuations.
Press the Run button to get the results.
You can prolong the time period, eg. into two years to get the big picture of the seasonal cycles of your subscribers affecting your revenue.
Find more details on how to utilize forecasting in your reports.