Viewing analysis results

Once you have completed recording variations of your task, it is time to leverage AI.

We recommend 50K actions as a good baseline. The model can support up to 200K actions collected in a project.

Tip:

As a quick way of estimating, an active user typically provides 200-1K actions per hour, depending on the scenario they are recording.

After running the Machine Learning model, the Results tab is available in your Unassisted Task Mining project. The results are determined by collecting all the recorded actions and screenshots and leveraging AI to determine discrete steps and their repeatability.

Once the model has determined the discrete steps that are frequently repeated, it will create tasks to group together those steps into visualizations.

Reviewing the presented tasks, traces and trends will provide additional insights into your organization and process. You can follow our Unassisted Task Mining Analysis Guide after you have run the machine learning model (see steps in the next subsection) to leverage best practices to drive continuous discovery and actionability from your project.

Results

Every time you run an analysis, a new analysis name is available in the Results tab with some details that can help you out in finding the proper analysis to deep dive into.

After the analysis is completed and results are generated, you can select an analysis to find out more details metrics about the captured data.

The Results page displays details of the analysis.

Below is a description of the display columns on the Results tab.

Column	Description
Analysis name	The name of the analysis created in the project.
Tasks identified	The number of repetitive tasks identified by the machine learning model.
Total time	The total time recorded (hh:mm:ss).
Users	The number of users that have recorded their daily activities in this project.
Actions	The number of actions (click, type, etc.) captured in the dataset.
Used apps	The number of used applications.
Discovery period	The date interval when data was recorded.
	Enables you to open a context menu with the following options. Open - displays a detailed analysis overview Remove - deletes the selected analysis Export to .CSV

Analysis overview

Select a specific analysis to view a more detailed analysis overview and the list of tasks that were identified.

Below is a description of the charts in the Analysis overview.

Chart	Description
Tasks	The number of repetitive tasks identified by the machine learning model.
Actions captured	The number of actions (click, type, etc.) captured in the dataset.
Total time captured	The total time recorded (hh:mm).
Users	The number of users that have recorded their daily activities in this project.
Total number of applications used	The number of used applications.
Analysis details	Detailed information on the analysis: The name of the analysis. The date at which the data was uploaded. The date interval when data was recorded.

Tasks tabular view

After selecting an analysis, the Tasks tabular view lists the repetitive tasks and the detailed metrics for each task.

Note: The tasks are ranked based on these tasks' automation potential, considering various factors, including repeatability and complexity. However, we strongly recommend that users review the tasks and prioritize them for automation based on their own criteria.

Below is a description of the columns in the Tasks tabular view.

Column	Description
Task rank	A distinctive rank to help you find the best discovered task opportunity to automate.
Total duration	The total time the users spent performing this task, across all traces attributed to the task. Note: The blue bar indicates the execution time of the task related to the other tasks.
Users	The number of users who have performed this task repetitively.
Traces	The number of times this task has occurred in the dataset. One trace is one iteration of this task done by one user.
Median duration	The median execution time across all traces attributed to the task. It indicates the length of performing the task.
Actions	The number of actions (click, type, etc.) in the most representative trace for this task - indicating the complexity of the task.
Task name	The name of the task. Default is "Task + number". See also Editing a task name.
Bookmark	Enables you to bookmark the task, to easily filter selected tasks.
Status	The status of the task. See also Changing the task status.
	Enables you to open a context menu with the following options. Open. See Viewing task details Export to .CSV

Editing a task name

By default, the task name is set to "Task" + number. For example Task 1. You can edit a task name to give the task a more descriptive name. Follow these steps to edit a task name.

1. On the analysis details page locate the task for which you want to edit the name in the Tasks tabular view list.

2. Hover your mouse over the Task name field and select the icon.

3. Edit the name as desired and press Enter.

Tip:

You can also change the name of a task in the task details page.

Changing the task status

All tasks will start with the status Not started by default. You can manually change the status of a task. Below is an overview of the available statuses you can select for a task.

Status	Indicates that
Not started	the task is not started (default status).
Review in progress	the task is being review to find automation opportunities.
RPA candidate	the review is completed. The task can be moved forward for automation.
Uninformative	the review is completed and did not result in needed actionability.

Follow these steps to set the status for a task.

1. On the analysis details page, locate the task for which you want to set the status in the Tasks tabular view list

2. Select the Status field for the task.

3. Select the new status for the task from the context menu

Tip:

You can also change the status of a task in the task details page.

Export to .CSV

You can export the data from an analysis results to CSV.

Exporting the data for all the tasks in the analysis

Follow these steps to export the data for all the tasks in the analysis to a .CSV file.

1. On the Results page, locate the analysis for which you want to export the data to .CSV.

2. Select the icon.

3. Select Export to .CSV from the context menu.

   

   The Export to .CSV dialog is displayed.

4. Select Include unprocessed data if you want to have all the recorded actions used in the analysis, including unprocessed data, in the exported .CSV file.

5. Select Export.

A notification message is displayed.

Exporting the data from a selected task

Follow these steps to export raw data from a task to .CSV.

1. On the Results page, select the analysis that contains the task for which you want to export the data. The analysis details are displayed.

2. In the Tasks tabular view list locate the task for which you want to export the data to .CSV.

3. Select the icon.

4. Select Export to .CSV from the context menu.

A confirmation message is displayed.

Note:

When you select the Export to .CSV in the upper right corner of the analysis details page, the data for all the tasks in the analysis will be exported.

Viewing the .CSV file

When the export to .CSV is completed, you will receive an email containing a link to download the exported data.

Note:

The link in the email will be available for 30 days. After that period, the exported file will no longer be accessible from the link.

Favorite traces

Follow these steps to mark a trace as favorite.

1. Expand the variant and locate the trace you want to mark as favorite.

2. Select the icon.

The trace is marked as favorite.

Favorites tab

You can select several traces from different variants as favorites. The selected traces will be available for quick access from the Favorites tab.

Removing a trace from Favorites

Follow these steps to remove a trace from the Favorites list.

1. Locate the trace you want to remove.

2. Select the icon.

The trace will no longer be accessible from the Favorites tab.