The ContractStandards Style Guide
A reference (or market standard) is an analysis of a large number of sample agreements. It captures the “digital signature” of the sample and can be used to compare any other agreement to the standard. A market standard is built for each agreement type. First, automated processes analyze a sample set of agreements. This process determines what clauses are found in the agreement, how the clauses are organized, what is the standard language for each clause, and all deal-specific variations. Second, editorial processes review conforms the organization of the market standard outline in a consistent manner with all other agreements of a similar nature.
The model form is a standard agreement in the form of a document that can be used as the starting point for a new negotiation. It contains all the clauses typically found in the agreement type and each clause is drafted in the most conforming language. The model form is crafted by subject matter experts, based on the market standard analysis. The process of creating a model follows three steps. First, the model is conformed to a consistent organization structure. Second, the language is conformed to a standard drafting style. Third, the terms are reviewed by subject matter experts.
Document benchmarking compares an agreement to the selected reference standard. It is a “one-to-many” comparison, displayed as an outline. The analysis shows which clauses in the benechmarked agreement match the standard (and the degree of match), which clauses differ from the standard, and which clauses are potentially missing.
Clause comparison compares a clause to the language of the model clause. It is a “one-to-one” comparison, displayed as a “redline” markup. The markup shows which terms are matched, which terms are included in the compared clause (but not found in the model clause), and which terms are in the standard clauses (but missing in the compared clause).
The benchmarked document is compared to the market standard (described above). A perfect document with a 100% score will contain all the standard clauses, drafted in the most conforming language (not necessary in any particular clause or word sequence). If the analysis determines that a clause is missing from the standard, the score is reduced by a value commensurate with the commonality of the clause in the sample set. If a clause is matched but drafted in non-standard language, the score is reduced by a value measured by the difference in words between the benchmarked clause and the norm.
The clause comparison score is calculated by comparing the differences between the compared clause and the model clause. The difference (or distance) is the number of deletions, insertions, or substitutions required to transform the compared clause into the model clause.