Privacy is guaranteed; even Captova cannot see your documents or data.

Accuracy is around 95% with real-time error detection.
Speed of our hyper automation clocks over 100 pages per second.

Security is as good as your on-prem environment - no cloud connection.

Captova's IDP Engine showcases remarkable efficiency, allowing a single user to process over 100 pages per second from a large pipeline of pre-processed and optimized PDF documents, such as invoices or tax forms. This equates to approximately 6,000 pages per minute or an astounding 360,000 pages per hour. 

Surprisingly, this small company has outperformed three tech giants with a combined market valuation of over $7 trillion in the field of Intelligent Document Processing (IDP). Industry experts consistently rank Captova’s performance above similar solutions offered by Microsoft, Google, and Amazon.

What sets Captova apart is its unique architecture, which diverges significantly from traditional methods employed by its competitors. The company claims its IDP Engine excels in the PASS criteria—Privacy, Accuracy, Speed, and Security.  An irresistible solution for heavy enterprise workloads. 

Below is a recent review of Captova by an IDP expert Tony McKinley.  

I recently enjoyed a virtual meeting with Mohamed Talib, Founder and CEO of Captova, to review the latest version of the Captova IDP Engine. We first spoke when I published a series of four LinkedIn articles on the Cloud Capture services provided by Amazon Textract, Microsoft Azure Form Recognizer and Google Document AI. Captova IDP is comparable to these three in core functionality, but offers dramatically different options in processing speed, absolute security assurance, pricing, and integration of PDF/A-3b.

The similarities all these systems share is the ability to process scanned digital images or original electronic files such as PDF documents. The systems are all designed to recognize and export Key-Value pairs in a variety of generic formats, typically JSON, CSV, or XML.

The first feature that is unique among these four options is that Captova IDP requires no cloud connectivity, which will be very attractive where absolute security is required. While cloud services are highly secure, even Amazon S3 buckets are vulnerable to the eternal bugaboo of User Error. For example, according to the cyber security firm Check Point Software, the three top S3 vulnerability issues are configuration mistakes, lack of visibility and malicious uploads. Because Captova IDP can run on a bare-metal Linux server, vulnerabilities due to shared cloud services or third-party storage are avoided.

This level of security offers a heightened assurance for customers who are concerned with privacy in industries that have suffered widely publicized breaches, such as financial services and healthcare. In other cases, such as in law enforcement and other government agencies, absolute security is required by legislation. In the most demanding environments, classified documents can only be viewed, handled, and processed in Sensitive Compartmented Information Facilities, where cloud connectivity is specifically forbidden.

The second feature that separates Captova IDP from cloud capture offerings from AMZN, MSFT and GOOG is that there are no per/page processing fees. For example, in addition to the basic Detect Document Text API, Textract has separate pricing for the Analyze Document API’s such as Queries, Tables and Forms, or the specialized Lending, Expense and ID API’s. S3 Storage is another cost element. Both Google Document AI and Microsoft Azure Form Recognizer offer similar click-charge pricing models. Captova pricing is per-server based on the size of Enterprise, with standard Maintenance and Services fees.

The third outstanding feature of Captova IDP is startlingly fast processing speed in extracting data, potentially providing game-changer advantages in critical enterprise applications. During our online demo, I watched batches of documents being loaded and then processed at unprecedented rates. At these speeds, with data from 80-114 documents being captured per second (on a modest machine with AMD Ryzen-5 5600 x12 CPU and NVIDIA RTX-4060 GPU), the only reliable way to observe actual performance is to examine output creation metadata. Many cloud capture applications are performed in traditional asynchronous batch processing, so the value of this hyper-speed synchronous performance will vary accordingly by business application.

Another value-add is the choice of pre-processing stages prior to data capture in the Captova IDP methodology. Scanned documents are subject to the complete array of image enhancements as expected, including the basic de-skew, de-speckle, and image optimization. The differentiating feature is that critical files are then rendered as PDF/A-3b, which is the archival version specifically designed to meet the need to combine multiple documents for records retention requirements. All PDF/A files are designed to be reliably accessible over time and platforms, as it says in the name – PDF/Archival, for not only as long as possible but also as free of dependencies on any specific software product or vendor. The unique enhancement of PDF/A-3 is the function that allows any type of content to be included in the PDF document as an embedded file or attachment. Further, the distinction of PDF/A-3b, compared to the other two PDF/A-3 flavors of 3a and 3u, is the focus on visual integrity to assure viewing fidelity of the archival document. This is valuable in business applications like invoice processing where original supporting documents can be attached to the invoices, such as purchase orders or contracts, for reliable future reference.

Further details on the Captova AI Suite, including IDP Intelligent Document Processing, ETL Extract – Transform – Load, and GPT Generative Pre-trained Transformer, as well as contact info is available at https://www.captova.com/.

November 15th 2024

LinkedIn Article at https://bit.ly/3AUrBOv

The landscape of Intelligent Document Processing (IDP) solutions is rapidly evolving, with numerous vendors touting high-accuracy solutions that promise automated classification and various AI capabilities. However, a closer examination often reveals a high degree of similarity among these systems, with many features becoming baseline expectations or "table stakes." The real differentiators emerge in implementation requirements, infrastructure compatibility, integration capabilities, overall performance, and cost-effectiveness.

Like most IDP systems, Captova incorporates a document pre-processing and optimization stage. In high-volume enterprise environments, it first pre-processes and organizes documents into a large pipeline of optimized files. When the end user uploads these  optimized files to Captova IDP, data extraction occurs in milliseconds.

A standout feature of Captova is its innovative approach to document model training. The system claims it requires only a single high-quality sample of a document to train its AI model, whereas most IDP solutions typically need 5 to 20 samples. This efficiency in training presents a notable advantage.

Captova also sets itself apart by eliminating the need for any coding or programmatic effort on the customer’s side. Designed as an out-of-the-box, no-code solution, it can be tailored to meet customer-specific requirements. Users simply upload optimized documents, and Captova handles the rest, processing documents in milliseconds. Its fast indexing algorithms swiftly match documents with the correct AI models, enabling near-instant data extraction—an especially valuable feature that reduces the need to hire developers to adjust IDP applications.

When it comes to performance, Captova is unique, in many cases processing over 100 pages per second. It manages complex documents at a remarkable speed, processing at nearly 10 milliseconds per page, while generating structured data in multiple formats (JSON, CSV, XML) and flagging errors in real-time. I have never seen anything like this before. When processing invoices, it automatically identifies the country of origin and creates a Tax Table that includes a tax breakdown of VAT, GST, or Sales Tax. This extensive feature set combined with unprecedented performance, puts Captova in a position few IDP solutions, if any, can match.

Captova's real-time processing speed and accuracy are transformative for organizations with high-volume document processing needs, where every millisecond counts. The ability to produce instant results without wait times is a powerful labor-saving feature, contributing to Captova’s cost-effectiveness and making it a compelling choice.

Beyond speed and accuracy, Captova prioritizes privacy and security, offering true on-premises deployment to ensure data privacy—a critical consideration for government agencies and highly regulated organizations handling sensitive information.

Captova IDP is a groundbreaking solution with unparalleled speed, accuracy, and a strong focus on privacy, making it an essential option for any organization looking to optimize document processing workflows.