Derecount statistics scenarios

1. Overview

1.1 Application scenarios

The count of a dimension field after the duplicate data is removed under certain conditions.

For example, express delivery has different modes of transportation. Please count the number of shipmethod per month. As shown in the figure below:

1.2 Solutions

This article provides four realization ideas and corresponding advantages and disadvantages of de-counting statistics, as shown in the following table:

Note: Performance is optimal because self-service datasets use processed data directly for dashboard presentation. Because the total rows of statistics are summed by default, the calculation is simple and most performance problems are avoided. By default, functions and dimensions are still automatically transferred, which may cause performance problems such as failure to load data.

2. Solution 1: Realize by self-service dataset

Count the number of shipmethods per month based on "ORDERS" table.

2.1 Create a self-service dataset

In the self-service dataset processing data, to achieve the recount statistics.

1) Under "Package" of "Data Preparation", click "Add Table" > "Self-service Dataset", as shown in the figure below:

2.2 Select fields

Name the new self-service dataset "Distinct count". Select "SIGNDATE" and "SHIPMETHOD" fields in the "ORDERS" table under "Package", as shown below:

2.3 Group summary

Use the "Group Summary", group "SHIPMETHOD" by "SIGNDATE", and select "Distinct Count" as the summary to obtain the number of shipmethods per month.

1) After the fields are added, click "+" in the left operation process bar and select "Group Summary". As shown in the figure below:

2) Drag the "SIGNDATE" field into the group column and group it according to "Year Month". Drag the "SHIPMETHOD" into the summary column and select "Distinct Count" from the drop-down list. After setting, click "Save", as shown in the figure below:

3) After making the self-service dataset, click "Update Data". As shown in the figure below:

2.4 Make a group table

Use group table to present the analysis results of the self-service dataset.

1) After data update, click "Create component" to create a component according to the newly created self-service dataset, as shown in the figure below:

2) Select "group table" to display data. Drag "SIGNDATE" into the dimension field, "SHIPMETHOD" into the indicator field, which shows the number of shipmethods per month, and "Enter the dashboard" to complete the component production. As shown in the figure below:

3. Solution 2: Realize by "Counter"

Count the number of shipmethods per month based on "ORDERS" table.

3.1 Create a dashboard

In the dashboard, you can de-count statistics by "Statistics number dependency" for "Counter". For clarity, the example sets the chart type to a group table.

Enter the corresponding information in "New Dashboard" in "Dashboard" and click "OK", as shown in the picture below:

3.2 Add component

Click "Add component", as shown below:

Make the component using "ORDERS" under "Package", as shown below:

3.3 Make a group table

After a new dashboard or component is created, the "Counter" field is displayed under indicator area on the left. Drag the "SIGNDATE" and "Counter" fields into the dimensions and indicators in the analysis area, as shown in the figure below:

Note: The list does not support the function of recording the number.

3.4 Statistics number dependency

Add the statistics dependency "SHIPMETHOD" to "Counter". Then the number of records will be counted according to the number of shipmethods in the same date dimension and the number of shipmethods per month will be calculated. Click "Enter the dashboard" when finished. As shown in the figure below:

4. Solution 3: Realize by aggregation function

Take the "ORDERS" under the "Package" as an example to count the number of shipmethods per month and use the COUNTD_AGG function in the dashboard.

4.1 Create a dashboard

Enter the corresponding information in "New Dashboard" in "Dashboard" and click "OK", as shown in the picture below:

4.2 Add component

Add the component using "ORDERS" in the same steps as above, see section 3.2 of this article for details.

4.3 Add calculation indicator

Use the "COUNTD_AGG" function to implement the "SHIPMETHOD" field de-count statistics.

1) Click "+" in the indicator area to add calculation indicators, as shown in the figure below:

2) Input the formula COUNTD_AGG(SHIPMETHOD) to get the statistic value of "SHIPMETHOD" in the corresponding dimension. After setting the indicator name, click "OK", as shown in the picture below:

COUNTD_AGG(array) : Dynamically returns the deduplicated count of a field based on the current analysis dimension. The result is a dynamic data column. The number of rows is the same as that of the current analysis dimension. For details, please refer to: Aggregation functions.

4.4 Make a group table

Drag "SIGNDATE" and the newly added "Shipmethod Counter" fields into the dimensions and indicators in the analysis area. The newly added "Shipmethod Counter" field displays the de-re-counting statistics of the number of shipmethods per month. "Enter the dashboard" when finished. As shown in the figure below:

5. Solution 4: Realize by converting dimension to indicator

Take "ORDERS" under "Package" as an example to count the number of shipmethods per month.

5.1 Create a dashboard

Enter the corresponding information in "New Dashboard" in "Dashboard" and click "OK", as shown in the picture below:

5.2 Add component

Add the component using "ORDERS" in the same steps as above, see section 3.2 of this article for details.

5.3 Convert dimension to indicator

The "SHIPMETHOD" field is "Convert to indicator" from a dimension to realize the de-counting. As shown in the figure below:

5.4 Make a group table

Use group table to display data. Drag the corresponding field into the dimension indicator bar of the analysis area. The "SHIPMETHOD" field shows the number of shipmethods per month. "Enter the dashboard" when finished. As shown in the figure below:

6. Preview effect

6.1 PC

See section 1.1 of this article for more details.

6.2 Mobile

The effects of APP and H5 are shown in the figure below: