Unlocking patent power: How Patently streamlines collaboration with vector search and NLP

Speed, size, and accuracy were three components that Patently soon realized were essential for managing its enormous patent collection as the company grew. Due to the intricacy of patent search and legal procedures, Patently needed a reliable system that would be adaptable enough to satisfy customer needs and to manage big data sets.

"We wanted it to be swift, and we knew we had to ingest a fairly large data set," said Andrew Crothers, creative director at Patently. “That was our main concern. For our clients, the speed at which we could search and retrieve what they were looking for was a cornerstone business requirement, and it still is."

To achieve these goals, Patently incorporated advanced search techniques, AI, and real-time data processing to reduce administrative bottlenecks. This enables users to focus on higher-level tasks leveraging AI-enabled features designed to simplify the complexities of vast patent-related data.

Understanding patent information involves managing vast amounts of structured and unstructured data dispersed across several systems. Patently wanted to build a solution that could expand and manage heavy data requirements while also providing a smooth search experience to users. Accurately identifying pertinent patents is essential for patent attorneys, mainly when different phrasing is used across geographies. Patently wanted to help users locate relevant and contextually meaningful patents, even in cases where an exact match was not found. 

The team chose Elastic’s Search AI Platform to build a versatile system that could handle contextual searches and exact matches, making it simple for users to find relevant information. With Elastic Search AI and vector search capabilities built into the Patently platform, it could interpret and create more intricate data relationships. "With Elastic, it’s like having a patent attorney with decades of experience guiding every search," Crothers said.

Additionally, the Patently platform supports patent litigation processes by providing accurate results quickly. Using sophisticated Search AI features has dramatically increased the efficiency and accuracy of its legal procedures.

The process of developing a modernized AI-powered solution was challenging. One of the main challenges was making sure the solution could accommodate each client's unique needs for patent searches, identify critical relationships between patents and patent filings, and seamlessly integrate Patently's custom search tools to accelerate data retrieval.

"We had some initial challenges when attempting to integrate our custom fields, but we collaborated closely with Elastic and their innovation team," Crothers said. “Their openness to listen and to work with us was really beneficial.”

Elastic Search AI is now incorporated into Patently's solution to improve search results and give users relevant information. As a result of this transition, users can now collaborate more efficiently based on how they think and work, leading to quicker and more informed decision-making.

The Patently team sought a platform capable of handling vast amounts of data, streamlining patent searches, and delivering real-time insights to empower better decision-making. They required a solution that could accommodate present data requirements and be ready for future expansion. The platform was built cloud-first, guaranteeing a smooth transition with Patently's other products and laying the groundwork for future innovation.

“With the implementation of Elastic Cloud, we've been able to significantly expand our product offerings by introducing Vector AI. This powerful addition has positioned Patently as one of the most innovative platforms for semantic patent search and is core to our technology stack. Elastic is used to deliver the new Standard Essential Patent (SEP) tool, Patently License. It's even used by our AI patent drafting assistant, Onardo, to search for prior art whilst preparing a patent specification,” said Jerome Spaargaren, founder and director.

Consider the magnitude: with over 82 million patent families comprising 135 million individual patents, each associated with 226 field mappings, the complexity of searching and retrieving patent-related information increases exponentially. This immense volume underscores the need to align patent management with business goals to streamline operations and safeguard innovations effectively in a rapidly evolving legal landscape.

Patently chose Elastic to build a platform that could manage increasing data volumes while providing easy-to-use fast search results to access information from global patent data sets quickly. Real-time data access and AI-powered vector search became essential components to speed up and scale the patent discovery process.

"During the transition, operational stability was a major concern," Crothers stated. "Our ability to scale seamlessly while keeping everything up and running is critical — we can’t afford any downtime."

Leveraging a scalable cloud architecture allowed Patently to maintain operational stability and concentrate on providing value to its clients. The ability to track in real time meant the system could adjust and scale as needed to keep everything up and running.

As it refined and deployed the platform to customers, Patently concentrated on enhancing system efficiency. Features for real-time monitoring and alerting became crucial for preserving the system's integrity. Through proactive monitoring, the team can fix any problems before they affect users.

One significant advancement involves the transition from a monthly batch processing methodology to real-time data intake. This makes it possible to get the most recent patent data right away. Having access to the most recent data allows users to make better decisions and give better advice more efficiently.

Once the new approach was implemented, Patently's search speed and accuracy significantly improved. Thanks to the platform's vector search capabilities, users can now locate what they’re seeking more easily because search results are prioritized by relevance. This has greatly enhanced patent searches and overall system usability.

AI-powered patent search tools are transforming how we identify and protect intellectual property. As Laurence Brown emphasized in a recent post, "AI-related patents are crucial to securing ownership over discoveries and inventions.” For example, using Patently Vector AI, Brown inputs “In-ear headphones with noise isolating tips” and sets a priority date filter before 2000. The tool returned 300 relevant results, which he further filtered to Sony applications. He was able to quickly identify the desired patents in less than five minutes, thanks to relevance-based sorting powered by Elastic Search AI features in the Patently platform.

Collaboration has also become considerably easier. Patently has established itself as a leader in the patent technology market by providing clients with cutting-edge solutions to assist them in navigating the ever-complex world of patents — all made possible through the use of machine learning and AI-driven insights. 

"The mechanism operates quickly. When working with millions of patents, it’s imperative that we obtain findings as rapidly as possible,” said Crothers.

The story of Patently demonstrates how progressive businesses can revolutionize their operations and create new product offerings with AI and cutting-edge search technology. Patently has revolutionized access to patent information by utilizing AI-powered search, real-time observability, and seamless data integration, setting a new benchmark for the industry.

Speed, scalability, and accuracy are critical success factors for organizations looking to remain competitive in the rapidly evolving field of patent information. Patently is at the forefront.