S is for Search.png

S Is For Contract Search

Putting aside the “Software” piece which plants the contract search & legal AI company firmly in the IT space, why did the founders choose the name Seal for their venture? While it is tempting to make up some narrative around a love of semi-aquatic marine animals or a mid-90s soul singer, the reality is somewhat more prosaic but also rather more far-reaching. Each letter of Seal is significant in that it identifies a core part of the platform that rolls up to the category of Contract Discovery and Analytics. Over the next four weeks, we are going to take a look at each of these, beginning with “S” for search, an often under-appreciated component of the contract analytics process.

As a marketing professional, I have spent portions of my career thinking about brand: brand ingredients, brand identifiers, brand manifestos and ultimately, brand value. One of the core elements is how a product, solution, and company are named. That can help the interested party orientate themselves quickly about the entity and what might be promising.  Similarly, the identifiers like the logo, color palettes, fonts used, and so on parlay the brand promise. Sometimes all these identifiers are intuitive and obvious, other times, not so. For example, think about when Google created Alphabet through their corporate restructuring a few years ago, it needed some explaining as to what that holding company was all about.  When Orange and T-Mobile merged and became EE (Everything Everywhere) the name meant something but again needed explanation. When Verizon was formed out of numerous older telecoms companies, they chose to use a portmanteau of veritas and horizon. Who knew? Again, those names all had meaning and a rationale for their selection.

In Seal’s terms, contract search has two components; first, contract discovery, finding the contracts to bring into scope, and second, once under management, an optimized engine to search contracts for surfacing contract data that is most important to the client.

When we first engage with clients, one of the key scoping questions seeks to determine where the client’s legal contracts are at that point in time. Usually, it falls into one of three camps. Camp number one, there are those who know exactly where all their contracts are, generally because they have implemented some form of Contract Lifecycle Management (CLM) solution and use it as a system of record, or source of truth. While the use of a CLM is common, it is very uncommon for all contracts to be housed in the repository. Camp number two, are those clients who can point to where the majority of their contracts are, usually spread across multiple CLMs, cloud storage devices, file shares or in other enterprise applications like DocuSign and Salesforce. These clients are not entirely convinced they know where all their contracts are located but can account for the majority and the most important ones. Then finally there is camp three, these are the organizations that have a weak handle on where their corpus of legal documents resides, with many still in physical paper format. Seal handles all three camp scenarios[1], but the most common, the second, is where Seal excels. Point Seal at the directory links, provide it the appropriate access credentials and then let it go to work to search – contracts. Seal can scale to hundreds of thousands of legal documents and associated documents that make up the entire corpus of contracts.

Seal will determine what documents, images or other files are legal contracts. This process includes reviewing PDF, TIF, GIF, JPEG, Word, and other digital file formats. Seal uses proprietary legal AI software to identify whether the document is a contract, is not a contract or whether it needs a manual contract review to determine its status. This first step of contract discovery is of significant value to clients who do not have all their contracts in a centralized place and otherwise would have to open each document and perform a manual review.

Seal then copies the contract document files and stores them into a centralized repository. The Seal repository is based on a Postgres database and allows the creation of unlimited metadata (more on this in a later post). However, each contract needs to be readable in order for the power of the legal AI engine to take effect, and so if the contract is not machine-readable (i.e., it is an image file), Seal will use an optical character recognition (OCR) engine to create a machine-readable version. Any files that Seal cannot render machine-readable due to low quality or other factors are tagged for manual contract review and placed in a queue. In parallel with the document examination and rendering, Seal will de-duplicate the contract collection as oftentimes, multiple copies of exactly the same document have been created.

Again, Seal contract search is adding value by automating a process that is often ignored or skipped. To manually search contracts and determine whether thousands and thousands of documents are legal contracts or not and then determining which are duplicates and OCRing them, so they are searchable, would require a huge resource and time drain. It also opens the process to errors by not accurately identifying exact duplicate contracts or not being sure that you have the entire collection. Automating the contract search, collection and de-deduplication processes not only saves time and resources, it is consistent. The results of this are a single point of access for all contracts which simplifies searching and readies the contract data as a source of minable business data.

The second aspect of contract search is to allow some core logic to be run against the corpus. As we shall see in part two, sophisticated data extraction processes help in identifying core topics, sub-topics, clauses, sentences, and words, but even without that, the ability to search for specific words across all contracts is inherent as a result of part one. These can be keyword, wildcard, proximity and other Boolean searches. These data extraction searches are easy to construct, and novice users can get up to speed in a matter of minutes.

So is for contract search, although it could also stand for sophisticated, surefooted or even special!

Next week we’ll take a look at E.

To learn more about Contract Discovery and Analytics and how Seal Software improves business performance with contract search, attend our on-demand webinar.


[1] Seal does not handle physical paper directly.