Product News: November 2019

We’ve upgraded our platform to give you even more control over your data capture processes while increasing the default capacity of our AI Core Engine. You may also notice that a couple of the app’s operations now run faster. Keep reading to find out about the new features we released in the past month.

General capabilities

Queue management

Queue management is a huge advance in the customization options of your Rossum organization account. You can now create Workspaces and Queues directly within the app. You’ll especially appreciate the ability to choose a default Extraction Schema from US/UK/EU invoice templates, which simplifies and accelerates the process of creating and setting up a queue. You can also rename Workspaces and Queues, assign users to a queue, and update queue settings.

To access this feature, click on the gear icon in the top left corner of the User Interface.

Performance optimization

Through our ongoing efforts to optimize the app’s performance, we’ve doubled the speed of switching between fields and editing fields in documents with long tables.

Table extraction accuracy

We’ve improved table extraction accuracy by as much as 3% in some cases. We mainly improved the accuracy of data captured from cells that contain text from adjacent cells.

Long Document Support

The AI Core Engine has been updated to handle documents of up to 32 pages by default, compared to the previous 16-page limit. Individual users may also request an ad hoc limit increase to up to 1,000 pages subject to a suitable subscription plan.


New elisctl version

We released a new version of our command line tool elisctl (2.8.0) for setting up your organization account over the API. You can now manage your Webhooks objects with the tool. Also, you can change inbox properties.

Retrieving object data in a single API call

You can now retrieve all the data contained in an object (e.g. an annotation, its content, queue details, and modifier user) in a single API call. This means you no longer have to download each additional object by an extra query. To achieve this, you retrieve a list filtered by id and use the sideloading feature to retrieve other associated objects simultaneously as needed.

Automate data extraction from your documents with Artificial Intelligence.
Free trial