Previously published in Practice Innovations, January 2012
The bad economy has given new life to knowledge management, as clients pay more attention to law firm efficiency. Document automation is one method of improving document drafting speed and consistency. Several new products have entered the legal market in recent times, affording opportunities that were not previously available. In this article, the author details his firm's experience with "kiiac", a linguistic analysis tool that the firm is using to help generate precedents.
The recent down economy is making knowledge management (KM) more popular than it has been for a while, due to the need for increased efficiency. As a result there has been a renewed focus on the use of document automation tools. Several new products have entered the legal market in recent times, affording opportunities that were not previously available. My firm is in the process of embracing one of these new tools as a means of providing improved client service.
By way of background, my firm, Cassels, Brock & Blackwell LLP, is a single-office firm in Toronto with about 200 lawyers and 45 law clerks, which offers the full range of large firm corporate and litigation services. We have a much larger percentage of lateral hires than most firms, which brings with it many advantages but also carries the risk that there may be many different approaches to the right way of doing things. One key sentiment that is widely shared, however, is acknowledgment of the need to create a comprehensive collection of authoritative and up-to-date precedents.
Creating a precedent bank and then keeping it up to date has been one of the cornerstones of traditional KM. Unhappily, though, this initiative commonly has not been as effective as it could have been. Indeed, precedent projects often fail, mainly due to the huge time investment in creating them, the often overly cumbersome documents that are produced, and the failure to keep them "green."
There is usually a disproportionate amount of time devoted to producing what ultimately appears to be of little benefit. This is because the authors (or owners) of the precedent tend to want the document to be as perfect as possible. Therefore, the resulting precedents often contain large numbers of possibly unnecessary clauses, requiring users to strike large sections of unnecessary text in order to prune the document back to essentials. The risk of this approach, though, is that users may end up throwing the baby out with the bathwater as they cut vast swaths of what appears to be unnecessary language.
More significantly, the resulting precedent might simply reflect the personal predilections of the drafters on the precedent team. It may be a pretty good starting point, but nonetheless may be missing essential clauses. Finally, the precedents, once developed, tend not to be updated to reflect current experience, since users in the heat of drafting rarely think to incorporate their improvements made to the base documents.
Essentially, the spanner in this approach is exclusive reliance on human effort to generate these documents. However, today's linguistic analysis software tools, which did not exist even five years ago, enable a significant degree of automation not previously possible. Document assembly tools that previously made sense only for very frequently used documents can now be used for regular documents.
It is now possible to leverage computer-based content analysis to:
- Fast-track the process of precedent production and renewal
- Generate good, reliable precedents
- Guide users in including necessary, and excluding unnecessary, clauses
This has essentially re-energized the firm's precedent initiative.
While increased efficiency and consistent branding of the firm's content always have been the primary business goals for a precedents initiative, there are other important ones. From a risk management perspective, it is crucial that all necessary clauses be included, all unnecessary clauses be eliminated, a logical structure be adopted for the final document, and there be consistency of language across similar documents.
These goals are now more easily attained when using computer-based linguistic analysis tools. From the point of view of training, good precedents also constitute a just-in-time means of delivering professional education. Further, there is increased satisfaction of professional staff, particularly associates, who have immediate access to top-quality content. Although impossible to quantify, this could have a positive effect on associate retention.
In the last year, we have begun using kiiac, the linguistic analysis tool developed by Kingsley
Martin, to help focus our precedent activities. We identify a particular document type and then, using our enterprise search engine, gather a representative sample (ideally at least 20, preferably more on the order of 40 or 50). These documents are then fed into kiiac, which identifies commonly recurring clauses and the consistency of drafting within clause types. Through what is essentially a computer-based elucidation of the wisdom of crowds, we can identify both a logical structure to the particular document type and all the necessary clauses, as well as the most commonly encountered drafting of each particular clause.
There is, however, a concomitant need for lawyer review, not just computer analysis. Kingsley's clear conclusion, after reviewing both publicly available documents on EDGAR and those submitted from many different firms, is that similar substantive clauses are often jumbled together with others. The process, therefore, involves teasing out the separate substantive issues so that each can be addressed on its own. There is accordingly a consultative effort involving Kingsley and the appropriate KM lawyer to arrive at the optimum form of the template document.
The resulting draft is then put out for use, and suggestions are invited. Realistically, however, drafting improvement suggestions are rarely made during the heat of document-drafting sessions. Accordingly, once a year we will identify and collect documents created from each particular precedent, run them through kiiac, identify substantive improvements that have been made to the underlying document, and update the precedent accordingly.
There are at least two advantages of using this computer-based tool in precedent drafting. First, the structure of the template document enables markers to be embedded within it (to identify things such as the names of the parties, the jurisdiction of the governing law clause, etc.), which can then be used to generate a document assembly checklist. A lawyer can, for example, review the opinion checklist to determine the relevant assumptions, opinions, and qualifications needed for the particular matter.
Instead of having to strike out unnecessary clauses from an omnibus precedent opinion, the drafter can focus on the substantive issues dealt with in the opinion. The resulting document accordingly should include only the necessary clauses. There are clear risk management advantages to having the drafter review the document assembly outline and think carefully about each potential issue. As well, the list can be given to a more junior member of the team, who will be obliged to think about the various substantive issues, thereby providing that lawyer with an excellent professional development opportunity.
Even five years ago, document assembly tools such as HotDocs" were available, but at a significant cost and intended for documents that were used repeatedly. With the checklist tools embedded in the precedent software, document assembly is now available even for documents that are used only occasionally.
The second advantage of newer solutions such as kiiac is the benchmarking capability. A firm's precedent document will only be useful half the time, namely when the firm creates the first draft. The rest of the time, the firm's role is to review the first draft prepared by opposing counsel. One cannot, however, blackline the latter document against the firm's precedent because its structure is completely different. The kiiac tool, on the other hand, because it analyzes the document's substance, provides a benchmarking capability to quickly identify what is missing, what is added, and what the drafting variances are.
While we are still at a relatively early stage of this process, there are some words of advice that I can offer.
- First, as with any similar initiative, it almost goes without saying that it is best to start small and to target the low-hanging fruit first.
- Second, it is important to recognize that this initiative strikes at the heart of the lawyers' drafting prowess. Generally speaking (and this is a conclusion that Kingsley has seen across the range of documents he has analyzed), there exists a much greater diversity in drafting styles than one would expect. While each lawyer recognizes the need for greater consistency in the drafting of firm documents, many do not recognize that this will require each of them to adapt (and indeed change) his own drafting practices.
As these documents become more common and are seen to be updated according to the firm's experience, and as lawyers use the document assembly checklists to generate first-draft documents, firms should see more consistent and substantively reliable documents being produced, thus advancing the business requirements that the precedents were intended to serve.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.