Reporting API

warning

The Reporting API instructions are applicable only to legacy Tableau-based analytics.

Reporting API represents operational metrics of Control Tower, Workspace, AutoML, RPA, Bot Execution, and Infrastructure. The API enables the following:

• Building custom dashboards
• Using a BI tool against the data warehouse provided by WorkFusion
• Integrating data seamlessly into your data warehouse from WorkFusion's one

The following schema represents all data available for analysis via Reporting API. These include the operational metrics of Control Tower, Workspace, AutoML, RPA, Bot Execution, and Infrastructure. You can leverage the automation metrics without any additional calculation efforts.

note

For a comprehensive overview of the WorkFusion Analytics capabilities, we recommend taking the Analytics for Business Users course in Automation Academy.

Connect to Reporting API

To start using Reporting API, connect it to WorkFusion Analytics Desktop. Mind the following prerequisites:

• The connection to Analytics Desktop requires a license. For details, read Legacy analytics components and licensing.
• The connection employs Microsoft SQL Server database. The default port is 1433. If you use a non-default one, specify it in the following format: <server name>, <port number>.
• Default database name: workfusion.
• Default schema name: dm.
• Default username for SQL credentials: rapi.

To connect Reporting API to Analytics Desktop, follow the instructions below:

1. Start WorkFusion Analytics Desktop. In the Connect section, click More... and select Microsoft SQL Server.

   

2. In the Server field, specify the Analytics server you want to connect to.

3. In the Database field, enter the default database name.

   

4. Enter your Username and Password. Select Require SSL if needed.

5. (Optional) Select Initial SQL to specify an SQL command to run at the beginning of every connection. For example, this can be a command to refresh an extract upon opening a workbook, sign in to WorkFusion Analytics Server, or publish to WorkFusion Analytics Server.

6. Click Sign In.

Select dataset

After signing in, you are on the Data Source page. Perform the following steps:

1. Select one or more Tables to match the goals of your analysis, relying on the Reporting API reference. Use the search box to browse for specific names, if needed.

   

2. Drag the selected Table to the canvas on the right.

   

   Alternatively, set up custom SQL queries. For details, refer to the Tableau official documentation.

   

3. Click the Sheet tab to proceed with building a report.

On the Data Source page, you can also accomplish the following:

• Change or add an initial SQL.

  

• Connect an additional Data Source.

  

Explore Reporting API reference

AutoML

Table	Order	Field	Description	Value
auto_ml_document	1	Id	Primary key for the table.	Auto-generated integer number
auto_ml_document	2	Process Execution Id	Foreign key to the process execution table.
auto_ml_document	3	Execution type	Shows the document processing details: Bot assisted: the model extracted not all fields. Fully automated: only a model processed the document. Fully manual: the document was handled manually without a model. Expired: the Manual Task for information extraction is expired. So, the related documents are to be excluded from calculations.
auto_ml_document	4	Is Automated	Shows whether it was only model that extracted all fields in the document.	0=no, 1=yes
auto_ml_document	5	Gold	Contains the number of fields to be extracted—FP and TN summed up.	Integer
auto_ml_document	6	Extracted	Contains the number of extracted fields—TP and FP summed up.	Integer
auto_ml_document	7	ML Processing Time	Shows the processing time of the automation-extract plugin (post to the ML service + ML extract, so on). Note: the automation-extract plugin is deprecated.	Milliseconds
auto_ml_document	8	Manual processing time	Shows the time to process the document manually.	Milliseconds
auto_ml_field	1	Id	Primary key for the table.	Auto-generated integer number
auto_ml_field	2	Document Id	Foreign key to the document table.
auto_ml_field	3	Group Name	Contains several fields—the equivalent of group answers.	Can be null.
auto_ml_field	4	Name	Field name—an element in the document to be processed by the AutoML model.
auto_ml_field	6	Extracted Value	Means the correct value for the field.	Can be null.
auto_ml_field	7	TP	True Positive (correct). The model should have extracted the value and has done so.	0 or 1
auto_ml_field	8	TN	True Negative (correct). The model should have not extracted the value and has done so.	0 or 1
auto_ml_field	9	FP	False Positive (mistake). No value is supposed to be extracted, but the model has extracted something.	0 or 1
auto_ml_field	10	FN	False Negative (mistake). The model should have extracted the value but has not done so.	0 or 1
auto_ml_field	11	Confidence	A dependent variable representing the estimated accuracy for the given threshold returned by the model.	0 or 1
auto_ml_field	12	Score	Normalized score, meaning an independent variable returned by the model. The variable represents the threshold model confidence score for a given field rounded to the nth decimal point.	0 or 1
auto_ml_field	13	Result type	Stores the field extraction result calculated based on the following conditions: If the FP+FN amount is greater than 0, the stored value is Extracted with Errors. If the total(FP)>0, the result is Extracted with Error. If the total(FN)>0, it is Failed to Extract. If the total(TP)>0, it is Correctly Extracted. Otherwise, the result type is Should not be Extracted.	Extracted with Errors, Failed to Extract, Correctly Extracted, Should not be Extracted
auto_ml_field	14	Report Date	The field is populated when the record containing the data for AutoML processing went through the Analytics process step. Format: YYYY/mm/dd hh:mm:ss.
auto_ml_field	2	Process Id	Foreign key to the process table.
process_execution_ml	2	Baseline hours	Shows the total expected time to complete a task as if all fields were extracted manually.	Hours
process_execution_ml	3	Saved hours	Shows the difference between the expected time and actual manual effort.	Hours
process_execution_ml	4	Manual effort	Shows the actual time spent on completing a task: how long it took to manually extract those fields that the model failed to extract.	Hours
process_execution_ml	5	Manual work reduction	Shows how much time AutoML models save relative to the expected task completion time.	From 0 to 100%
process_execution_ml	6	STP Rate	Shows the rate of documents processed by bots only.	From 0 to 100%
process_execution_ml		Process execution Id	Foreign key to the process_execution table.

Bot Execution Platform

Table	Order	Field	Description	Value
bep_agent	1	Id	Primary key for the table.	Auto-generated integer number
bep_agent	2	Client Appid	This field specifies the client application using the BEP Worker in the format: <Control Tower host>, AutoML model service.
bep_agent	3	Agent host name	IP address of the Agent.
bep_agent	4	Worker Id	BEP worker ID.	Text
bep_agent	5	Worker profile		Control Tower, AutoML execution, AutoML training
bep_agent	6	CPU consumed	Each BEP Worker reserves CPU for execution.
bep_agent	7	Memory consumed	Each BEP Worker reserves Memory for execution.	Bytes
bep_agent	8	Report date	The field is populated automatically every minute to store the latest value for the BEP Worker metrics. Format: YYYY/mm/dd hh:mm:ss.
bep_metric	1	Id		Auto-generated integer number
bep_metric	2	BEP task Id	Primary key for the table.	Auto-generated integer number. Can be null if a BEP task comes from AutoML.
bep_metric	3	Task Id	Foreign key to the task table.
bep_metric	4	Client application		Control Tower, AutoML
bep_metric	5	Input queue date	Timestamp when the task message got into RabbitMQ.
bep_metric	6	Worker queue date	Timestamp when the task message was moved from RabbitMQ to the Worker queue.
bep_metric	7	Worker processing start date	Timestamp when the Worker started processing the task message. It is the same moment when the task message leaves the Worker queue.
bep_metric	8	Worker processing end date	Timestamp when the Worker completed processing the task message. It is the same moment when the task message gets into the result queue.
bep_metric	9	Result received date	Timestamp when the client application processed the task message result.
bep_metric	10	Input queue wait time msec	Time spent in RabbitMQ.
bep_metric	11	Worker queue wait time msec	Time spent in the Worker queue.
bep_metric	12	Worker processing time msec	Time the Worker spent on processing the task message.

RPA

Table	Order	Field	Description	Value
bot_unit	1	Id	Primary key for the table.	Auto-generated integer number
bot_unit	2	Bot Relay Port	RPA bot relays orchestrate work. Bot relay names are in the following format: <RPA Instance name>:<Port name>.
bot_unit	3	Bot port	RPA Bot port.
bot_unit	4	Bot Agent port	RPA Bot Agent port.
bot_unit	6	Host	RPA Bot Relay hostname.
bot_unit	7	Fleet	A group of Bot Relays. Required for proper Bot Task distribution.
bot_unit	5	Worker port	RPA Worker port
bot_unit_session	1	Id	Primary key for the table.	Auto-generated integer number
bot_unit_session	2	Task Id	Foreign key for the task table.
bot_unit_session	3	Bot unit Id	Foreign key for the bot_unit table.
bot_unit_session	4	Start date	The field is populated as the first step is completed in a process run by a user or a schedule. Format: YYYY/mm/dd hh:mm:ss.
bot_unit_session	5	End date	The field is populated when a process is successfully completed or stopped. The null value is valid in case a process is still in progress or paused. Format: YYYY/mm/dd hh:mm:ss.
bot_unit_session	6	Status	Shows if a process is available for running or deleted.	Active, deleted
bot_unit_session	7	Driver type	Shows the driver in use, such as Chrome, Firefox, desktop, universal, and so on.
bot_unit_session	8	Error message	Explains why the session failed. Otherwise, it is null.
bot_unit_session	9	Wait time msec	Shows how long a session waits to be processed by a bot.	ms
bot_unit_session	10	Processing time msec	Shows how long it takes for a bot to process a session.	ms
bot_unit_session	11	Total time msec	Calculated as the sum of the processing and wait times.	ms
bot_unit_status	1	Id	Primary key for the table.	Auto-generated integer number
bot_unit_status	2	Bot unit Id	Foreign key to the bot_unit table.
bot_unit_status	3	Bot unit status	The field is populated when the bot status changes.	Active, idle, unavailable
bot_unit_status	4	Date from	The field is populated when the bot status changes. Format: YYYY/mm/dd hh:mm:ss.
bot_unit_status	5	Date to	The field is populated when the bot status changes. Format: YYYY/mm/dd hh:mm:ss.
bot_unit_status	6	Status duration	The field is calculated based on the status start and end dates.	Seconds
process_execution_bot_utilization	1	Process execution Id	Foreign key to the process_execution table.
process_execution_bot_utilization	2	Report date	The field is populated automatically on an hourly basis. Format: YYYY/mm/dd hh:mm:ss.
process_execution_bot_utilization	3	Utilization rate	Shows the percentage of the bot active time relative to the available bot time. Bots availability is calculated on the 24/7 basis.	From 0 to 100%
process_execution_bot_utilization	4	Active time	Shows the period when bots have active sessions for a given process execution.	Seconds
process_execution_bot_utilization	5	Bot relay port	RPA Bot Relays orchestrate work. Bot Relay names have the following format: <RPA Instance name>:<Port name>.

Infrastructure

Table	Order	Field	Description	Value
component	1	Id	Primary key for the table.	Auto-generated integer number
component	2	Name	Work.AI component names.	OCR, Master, Agent, Bot Relay
component	3	Work type	Defines what type of work a component supports. The Control Tower work type includes all machine tasks that are not related to RPA or OCR.	RPA, OCR, Control Tower, BEP
component	4	Description	Additional information about a component. For example, the Application Server is Control Tower for App.
component_metric	1	Id	Primary key for the table.	Auto-generated integer number
component_metric	2	Component Id	Foreign key to the component table.
component_metric	3	Host	Component hostname
component_metric	4	CPU	Shows the processor load across all cores at a given moment (every minute). The value is a percentage indicating whether any changes are required in the system to avoid capacity deficiency.	From 0 to 100%
component_metric	5	Cores	Shows the number of cores for Work.AI components.
component_metric	6	Disk	Percentage of time the disk is processing read or write requests at a given moment (every minute).
component_metric	7	Memory available	Memory available at a given moment (every minute).	GB
component_metric	8	Memory total	Total memory available at a given moment (every minute).	GB
component_metric	9	Report date	The field is populated automatically every minute to save the latest value for the component metrics. Format: YYYY/mm/dd hh:mm:ss.

Control Tower

Table	Order	Field	Description	Value
process	1	Id	Primary key for the table.	Auto-generated integer number
process	2	Name	Process name, also known as a Business Process name. Defines the flow of Manual and Bot Tasks.
process	3	Status	Shows if the process is available for running or deleted.	Active, deleted
process	4	Execution type	Shows if the process is a combination of tasks or it is just a single task from a process.	Single, composite
process	5	Creation date	Shows when the process was created. Format: YYYY/mm/dd hh:mm:ss.
process	6	Last modified date	Shows when the process was modified. Updated every time after clicking the Save button. Format: YYYY/mm/dd hh:mm:ss.
process_execution	1	Id	Primary key for the table.	Auto-generated integer number
process_execution	2	Process Id	Foreign key to the process table.
process_execution	3	UUID	Universal unique ID for the process execution in the following format: https://<Control Tower instance name>/workfusion/secure/business-process/edit/<UUID>. Helps to find a given process execution from the Control Tower user interface.
process_execution	4	Start date	The field is populated when the first step is started in a process run by a user or a schedule. Format: YYYY/mm/dd hh:mm:ss.
process_execution	5	End Date	The field is populated when the process was successfully completed or stopped. The null value is valid in case the process is still in progress or paused. Format: YYYY/mm/dd hh:mm:ss.
process_execution	6	Status	The field is updated as the process execution status changes. Deleted process executions are not stored in the table. For all statuses, except for completed, the corresponding process execution end date is null.	Completed, paused, processing
process_execution	7	Tracking activated	Shows that transactions are tracked, which helps to identify business entities at a certain step in a Business Process.	Activated, Not activated
process_execution	8	SLA Type	Type of the Service Level Agreement (SLA) applied at the process execution level.	Time, volume
process_execution	9	SLA value	Defines the SLA value.	Seconds or the number of transactions
process_execution	10	Issues	Shows if the process execution has issues, which means the maximal number of retries is exceeded at least for one record and the record failed.	With issues, without issues
task	1	Id	Primary key for the table.	Auto-generated integer number
task	2	Name	Task name, a step in a process.
task	3	UUID	Universal unique ID for the task execution in the following format: https://<Control Tower instance name>/workfusion/secure/business-process/edit/<UUID>?subTaskUuid=<taskUUID>. Helps to find a given task execution from the Contol Tower UI.
task	4	Process execution Id	Foreign key to the process_execution table.
task	5	Task position	Shows the task position in a process.	Integer
task	6	Task type	Shows if the task is in a process or out of it. For instance, it can be a single Manual Task.	Single task, task in process
task	7	Start date	The field is populated when at least one record reaches the task in a process. Format: YYYY/mm/dd hh:mm:ss.
task	8	End date	The field is populated when the task is successfully completed or the process is stopped. The null value is valid in case the task is still in progress or paused. Format: YYYY/mm/dd hh:mm:ss.
task	9	Status	The field is updated as the task status changes. Deleted and draft tasks are not stored in this table. For all statuses, except for completed, the task end date is null.	Completed, bot processing, paused, processing
task	10	In/out of tracking	Shows if the task is subject to transaction tracking.	In tracking, out of tracking
task	11	Excluded from STP	Shows if it is required to exclude the task from STP calculations. Typically, used for mandatory Manual Tasks.	0=no, 1=yes
task	12	Execution type	Shows the task type.	Bot, manual
task	13	Stateless execution	Shows if stateless execution is enabled. In this case, no operational data for the task is stored in the database. Data is stored only for the first marked task and for the Final Result.	0=no, 1=yes.
task	14	Has processing issues	Shows if the task execution has issues, meaning the maximal number of retries is exceeded for at least one record and the record failed.	0=no, 1=yes
task		UUID	Process execution UUID
transaction_e2e	1	Id	Primary key for the table.	Auto-generated integer number
transaction_e2e	4	Creation date	The field is populated in two cases: 1) when transaction tracking is enabled, the field shows when the tracking started; 2) when transaction tracking is disabled, the field shows when the transaction defined as an input file record reaches the first task. Format: YYYY/mm/dd hh:mm:ss.
transaction_e2e	5	Completion date	The field is populated when the transaction is successfully completed or the process is stopped. The null value is valid in case the transaction is still in progress. Format: YYYY/mm/dd hh:mm:ss.
transaction_e2e	6	Status	The field is calculated based on the start and end dates. When the end date is null, the status is defined as processing.	Completed or processing
transaction_e2e	7	SLA result	Shows if the transaction is within SLA or violates it.	Within SLA, SLA violation
transaction_e2e	8	STP	Shows if the transaction goes straight through the process without manual handling.	0=no, 1=yes
transaction_e2e	9	Retries	Shows the number of plugin attempts to process the transaction.	Integer
transaction_e2e	10	Processing time	Shows the transaction processing time in the course of the process execution end to end.	Seconds
transaction_e2e	11	Wait time	Shows the transaction overhead time in the course of the process execution end to end.	Seconds
transaction_e2e	12	Transaction cycle time	Shows the end-to-end processing time—the sum of the processing and wait times.	Seconds
transaction_e2e	13	Activated tracking	Shows that transactions are tracked, which helps identify business entities at a specific Business Process step.	Activated, Not activated
transaction_item	1	Id	Primary key for the table.	Auto-generated integer number
transaction_item	2	Transaction E2E Id	Foreign key to the transaction table.
transaction_item	3	Task Id	Foreign key to the task table.
transaction_item	4	Creation date	The field is populated when the transaction item is created in the process. Format: YYYY/mm/dd hh:mm:ss.
transaction_item	5	Start date	The field is populated when the transaction item reaches the next task in the process. Format: YYYY/mm/dd hh:mm:ss.
transaction_item	6	End date	The field is populated when the transaction item is successfully completed or the process is stopped. The null value is valid in case the processing of the transaction item is still in progress. Format: YYYY/mm/dd hh:mm:ss.
transaction_item	7	Completion date	The field is populated when the transaction processing result is received. Format: YYYY/mm/dd hh:mm:ss.
transaction_item	8	Wait time	Shows how long the transaction item waits to be processed.	Seconds
transaction_item	9	Processing time	Shows how long it takes to process the transaction item.	Seconds
transaction_item	10	Total time	Calculated as the sum of the processing and wait times.	Seconds
transaction_item	11	Work type	Defines the work type applied to the transaction item. The Tower work type includes all machine tasks that are not related to RPA, OCR (if the OCR plugin is used), or AutoML.	AutoML, RPA, OCR (if the OCR plugin is used), Manual, Tower
transaction_item	12	Work type count	Shows how many work types were used for processing the transaction.
transaction_item	13	Worker Id	Foreign key to the worker table. Populated in case the transaction item is processed manually. Otherwise, the value is null.
transaction_item	14	Retries	Shows the number of the plugin attempts to process the transaction item.	Integer

Manual

Table	Order	Field	Description	Value
worker	1	Id	Primary key for the table.	Auto-generated integer number
worker	2	UUID	Universal unique worker ID, also known as the native ID.
worker	3	Name	First and last names of the worker.
worker	4	Type	Shows the worker type.	Bot, Manual
worker	5	Nickname	Nickname, used instead of the worker name when the latter is not specified.
worker	6	Email	Contains the worker's email.	Can be null.
worker	7	Country	Contains the name of the worker's native country.	Can be null.

Build report

To generate a report based on the selected dataset, drag-and-drop rows, columns, and values on the pivot table. Select the most appropriate visualization.

For detailed instructions on building reports, refer to the Tableau video tutorials (compatible with Work.AI 10.x versions).

Publish report

For instructions on publishing the resulting report to the Analytics Server, refer to the guide.

Check out demo

For an example, refer to the Deploy custom dashboards (legacy) guide.