With grouping of the results you can change the look of the final report.
Applicable when at least one metric and one dimension are selected e.g. app and CTR.
The final report will be divided into several tables, according to the number of variants of values in the dimension. The dimension itself however will be excluded from table columns.
A total sum (Total) would be displayed at the bottom of the table for all metrics, which can be added up (e.g. Downloads) and a final sum at the very bottom of the table.
For instance, if the structure of the report is as follows: App, Date, Clicks, Downloads and the report is made for 3 days for 2 apps, then it should look like this:
| Date | Clicks | Installs |
|---|---|---|
| Application X | ||
| 2024-11-18 | 326 | 0 |
| 2024-11-19 | 142 | 83 |
| 2024-11-20 | 532 | 4 |
| Total | ||
| 1000 | 87 | |
| Application Y | ||
| 2024-11-18 | 704 | 94 |
| 2024-11-19 | 985 | 62 |
| 2024-11-20 | 732 | 33 |
| Total | ||
| 2421 | 189 | |
| TOTAL | ||
| 3421 | 276 | |
Grouping can also be with transposition.
If no grouping is selected, the report would look like a single table where every column would represent a dimension. Column order is identical to the selected metric order in report structure.
For instance, if the structure of the report is as follows: App, Date, Clicks, Downloads and the report is made for 3 days for 2 apps, then it should look like this:
| Application | Date | Clicks | Install |
|---|---|---|---|
| Application X | 2024-11-18 | 278 | 65 |
| Application X | 2024-11-19 | 829 | 7 |
| Application X | 2024-11-20 | 445 | 67 |
| Application Y | 2024-11-18 | 928 | 26 |
| Application Y | 2024-11-19 | 155 | 34 |
| Application Y | 2024-11-20 | 749 | 74 |
| Total | |||
| 3384 | 273 | ||
Therefore, the example demonstrates that simple grouping (i.e. no grouping) is useful only in cases when the report is not divided into several cohorts (e.g. when the report is built for only 1 app and broke down just date).
