Cognitive data capture uses artificial intelligence (AI) to mimic the way the human mind reads structured documents. This approach has two key features:

1. AI learns to recognize information through exposure to examples rather than manual configuration by experts,
2. AI can recognize a lot of information in documents with layouts it has never seen before.

Unlike manual data entry or traditional OCR, cognitive data capture requires no extra manpower. It also saves you the hassle, time, and cost of setting up endless rules and templates.  You can read detailed comparisons of effort in our TCO analysis series.

Rossum’s cognitive data capture AI uses deep neural networks to recognize patterns in documents, just as a human mind would. This enables the platform to understand the underlying general structure of business documents such as invoices. Rossum’s unique neural network architecture allows it to comprehend a wide range of layouts. It also ensures highly accurate data extraction.

Read our founders’ blog series on cognitive data capture for an in-depth look at the technical differences between legacy OCR solutions and cognitive data capture. You can also learn how Rossum’s technology works.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

To ensure up-to-date and widely scalable security, maintenance and regular updates, we do not offer implementing Rossum on-premises.

Cloud is the most reliable medium through which we can provide broadly scalable service with the highest level of security. It is worth noting that cloud-based solutions deliver security benefits comparable to those of on-premises solutions.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

Rossum runs on AWS out of Ireland. Enterprise customers can have Rossum deployed on AWS cloud data storage in a different country.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

You can implement Rossum in less than 1 day.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

Rossum can immediately recognize most languages that use a variation of the Latin script. The languages that Rossum recognizes are: Czech, Danish, Dutch, English, Finnish, French, German, Hungarian, Italian, Lithuanian, Norwegian, Polish, Portuguese, Brazilian Portuguese, Romanian, Slovak, Slovenian, Spanish, Catalan, and Swedish.

Rossum also supports Japanese and Chinese (beta).

Don’t see the answer you’re looking for? Visit our FAQ section for more.

Before we can give you a price estimate for Rossum’s annual subscription, we need to understand your requirements.

In addition to the estimated annual document volume, we also need to know what data fields you want to extract and what customization and/or training requirements you may have.

If you’d like to receive a quote from us, please fill out this form. One of our experts will then contact you immediately.

Don’t see the answer you were looking for? Visit our FAQ section for more.

Since our main focus is on data extraction, we do not provide integration services directly. We have extensive API documentation that allows smooth integration with most business systems. Read more about the API.

You can also check out this guide, which includes examples of three different types of Rossum integration, including manual integration. We have also published an example of UiPath integration.

For minor changes you need to make to ensure your integration works properly, such as adjusting the output file format, we can develop a custom connector for you.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

The Rossum data capture engine can handle any date or decimal format and normalize it according to your preferred standardized representation.

For truly ambiguous cases, there is a special “locale” setting that allows you to adjust the platform to handle individual document queues based on their region of origin. Date formats are very flexible, and we can customize your UI to display exported data in the format you want.

You can find some of the supported formats here, under the “Date format” section; the tokens mentioned there are available at this link.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

Rossum is fully committed to ensuring GDPR compliance. We process customer-supplied documents for the primary purpose of data capture, as instructed by our customers. The secondary purpose is further research and development of data extraction technology.

We have every reason to believe that this processing fully complies with the GDPR. This is based on the nature of the data and the GDPR balance test, especially since it does not involve invoices from third-party customers or invoices that contain sensitive personal data.

Please see our Terms and Conditions for more information.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

We follow the OWASP Secure Coding Practices and rely on the extensive experience of our senior team members.

In the event of a code change, we perform design reviews, code reviews, and security reviews. At least one other software engineer inspects and reviews each commit. We use thorough automated testing, including unit tests and integration tests, as well as manual testing to ensure code quality and security.

We also use third-party automated tools for static source code checks and vulnerability scanning. Our platform undergoes regular penetration testing by an independent third party.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

When transferring data in and out of the cluster, we always use encryption. For data at rest, we use AES 256 keys managed in the AWS Key Management Service. For all data in transit using HTTPS (including HSTS), we use TLS v1.2.

When in motion, all external communication is strictly encrypted, typically via HTTPS for regular production operations. We use SSH encryption to encrypt external communication for some service and maintenance purposes.

Communication with the database is always encrypted. We use an audit log for all operations executed in the application.

Don’t see the answer you’re looking for? Visit our FAQ section for more.

To improve Rossum’s automation processes, we have partnered with RPA giants UiPath and Blue Prism. Take a look at our invoice extraction use case with UiPath and our invoice processing automation use case with Blue Prism. 

Don’t see the answer you’re looking for? Visit our FAQ section for more.

Rossum can extract data from semi-structured documents other than invoices, such as receipts, purchase orders, and shipping documents. We can also train a custom AI model to capture data from specific document types, so you can use Rossum to process other documents.

Our custom training add-on can capture data from any document that is at least partially defined by its layout. Invoices, bills, and receipts are examples of such documents. Rossum can also process documents with similar formats, such as purchase orders, delivery notes, confirmations, statements, and forms.

For Rossum to successfully capture data from any document, two conditions apply:

1. content must be in Latin characters,
2. tables must be in a grid format and their columns must have uniform meaning.

Don’t see the answer you’re looking for? Visit our FAQ section for more.