Frequently Asked Questions
Basic questions that you are likely to have while using Task Mining technology:
What applications are required to work with Task Mining?
Task Mining technology uses desktop user interaction data to define tasks and create their digital representation for further analysis and process drill down. The full workflow includes several applications:
Records user actions in different applications and web browsers.
- Recording Service
Stores recorded user actions, allows to apply different settings to logs such as data obfuscation, and helps to easily upload logs to Timeline for further analysis.
Allows to define tasks from uploaded logs and proceed with task analysis using built-in tools.
What user actions are recorded?
When Recorder is on, it logs all performed user actions in desktop applications and browsers: applications and tabs opening and closing, switching between windows, entering information in text fields, clicking buttons, typing text on the keyboard, copy and paste operations, etc.
What are the best practices for recording user actions?
It is generally recommended to record only applications and actions related to business processes and tasks. Any extra and unnecessary actions will clog up the logs, slowing down task definition and hampering analysis.
Follow these tips when recording user activity:
- Encourage users to focus on the tasks that are related to the business processes you are interested in.
- Encourage users to avoid multitasking and perform tasks one by one.
- Encourage users not to use applications that are unrelated to tasks.
- Log same-type tasks performed several times.
Where are recorded logs stored?
Recording Service collects and stores logs captured by Recorder instances installed in the Server managed mode.
In some cases, Recorder can be installed in the Standalone mode. Then recorded logs are stored on the users' computers in the %appdata%\ABBYY\Recorder\Recorded folder.
For details, see Data Upload.
How to improve log quality?
Using Recording Service, you can configure lists of applications that must be monitored or should be completely ignored. Such INCLUDED or EXCLUDED lists of applications can significantly reduce the amount of unwanted data in the project and make tasks and process analysis easier.
For details, see Recording Service User's Guide > Excluding and Including Applications.
Recorded logs include sensitive data. How to protect this information?
Use these Recording Service features to hide sensitive information in logs:
- Obfuscate user data
This option allows encrypting sensitive information in logs and blurring text on recorded screenshots when uploading logs to the Task Mining project.
It allows removing sensitive data that has been captured by Recorder as event names or casting the names of several similar events to a common name.
For example, if it is recorded in the log that a user has opened a document called PO-BankOfAmerica.pdf or BillGatesSalary.xls, it is best to replace these with redacted strings like opening PO-XXX.pdf or XXXsalary.xls respectively. Configure such transformations and apply them to selected logs.
Note. Transformations cannot be undone.
For details, see:
Recording Service User's Guide > Adding and Applying Transformations
How to upload logs to a Task Mining project?
You can upload logs directly from Recoding Service or manually upload ZIP files with logs to the project.
For details, see Data Upload.
What is a task?
A task is a sequence of user actions to achieve some results. For example, a user books tickets for employees. Booking a single ticket is a task instance. A pool of all task instances is a task.
What is a task definition?
This is a template for the program to find task instances in logs. It shows Timeline which user actions are the completed part of the business process. Such a template consists of the rules list, which includes, for example, what forms can start or end a task instance, if there are any required forms, etc.
For details, see Task Definition.
How to select start and end forms?
To select the start and end points of the task, begin with a log review. Decide what is the most common way to begin or finish a task.
For example, a user creates invoices. The task can start with receiving an email, opening a Microsoft Excel document, or creating a new file. Select the best option to begin with. It is significant that all task instances should include this step so the program can consider them as a part of the same task.
For end forms, follow the same workflow, or choose one of the task properties. For example, a task can finish with any form that goes before the next start point.
For details, see Task Properties.
Why do I need to cut logs again to apply task definition changes?
When you make changes in task definitions, the program needs to check all logs and apply new task definitions to them. To renew the whole task statistics and update information in the project according to modifications, you need to rerun the log cutting process.
Can Timeline define tasks automatically?
The Discover task allows you to choose the forms you want to include and instantly create a task with them. You can experiment with multiple tasks creation and compare how the presence or absence of forms or applications in definitions influences the tasks.
For details, see Discover Task.
How does cutting mode affect the task definition?
A log cutting mode sets the boundaries for a task with multiple identical start and end forms.
For details, see Cutting Mode.
What is a subtask?
A subtask is a sequence of actions in task instances that can be defined as a separate task. For example, while creating an invoice a user often checks if items are in stock. You can identify these actions as a subtask, to estimate how it influences the whole task.
For more information about subtasks, see:
Task Schema Customization Options
What is the use of the task schema?
A task schema is a graph that represents created tasks on different levels. Task visualization makes task definition and analysis easier and more effective. Task schema also allows preparing and generating a Process Definition Document (PDD) that collects recommendations for tasks automation in one file.
For details, see Task Schema.
What can I find out about tasks after cutting logs?
After the program cuts logs into task instances, Timeline will update the Task view page. Reviewing this information, you can see which tasks are the most complicated. You can estimate the gain from task automation, the time that will be saved if you automate processes, and the money saved if you configure tasks cost.
For details, see Task View.
How to analyze tasks?
After you upload logs to the project, create task definitions, and cut logs into task instances, the program will treat each task instance as a timeline. Then you can use built-in analysis tools to explore tasks.
Another way to dive deeper into processes is to create an association of a task with an event from a Process Analysis project. This will expand the abilities of process analysis. You will have a detailed description for each event: what a user does to perform a task that results in the event.
For more details, see:
What are complexity and gain of the tasks?
The complexity is determined by the number of task variations, the number of forms, and the applications involved. The greater its value, the more complicated the task is to perform. The complexity of the task is calculated automatically after defining the task. You can see this value in the Automation Candidates table and the Candidates Distribution chart.
The task gain is determined by the time required to complete a task. The greater the gain for a task, the more you will benefit from automating it. This value is shown on the Candidates Distribution chart axis.
For details, see:
Candidates Distribution Chart