Legal Fee Audits, Document Retention Programs, Tired Solutions, and Blockbusting for Lawyers: Part I in a Continuing Series About Creativity in Problem-Solving

Posted by Jeff Lipshaw

The Money and Investing section of the Weekend Edition of the Wall Street Journal (Saturday/Sunday, Nov. 4-5) has a piece on corporate audits of the bills that outside law firms submit for services rendered.  This is but one example of the kind of fee-based offer of service you get as the general counsel to a big company.  Document retention (euphemistic for document destruction) programs were another (a quick Google search shows you can buy one from Hunton & Williams, as an example and the ABA has materials out on them as well.)  We would regularly get high-powered sales pitches from vendors who specialized in flyspecking bills (as to the audits) and the law firms themselves (as to the document retention).  It was always a classic case of data against intuition, because it always seemed to me a real question whether there was any payoff, in either case, to what was sure to be a time and cost intensive activity.   

This is too big a topic to handle in one blog post, but I want to lay some groundwork first, and then in later posts come back to the relationship between business law, business lawyers, and the theory and pedagogy of creative solutions to complex problems.  Not to hide where I stand (because it may be a while before I get back to it), there's nothing wrong, per se, with legal audits or document retention programs.  I merely think that they are tired and uncreative solutions to problems.*  I have been thinking about the theoretical issues for a long time; nascent ideas about pedagogy are the result of a conversation I had yesterday about the courses you might build into a cutting edge business law curriculum, one of which might be an advanced seminar in creative problem solving techniques.   

Let's start with some basics in organization and management learning and come back in subsequent posts to the professional (and I will argue, theoretical and pedagogical) issues for lawyers.   This segment is devoted to what is known over in the business schools as "lean production" or "lean enterprise," a subset of which is the exercise in reduction of process variability (and therefore defect reduction) called "six sigma."  These terms are, in no small part, the upshot of a global revolution in thinking and action that dwarfs relatively "easy" problems like law school curriculum reform.  Over the last thirty years or so, an entire world (one in which most lawyers do not traffic) looked at the accepted and traditional ways of doing things in human organizations, and realized they were outdated.  Talk about paradigm shift.  If one has been in that world (and law firms generally don't count), then reading an article about auditing law firm bills as the solution to a problem is like having someone's fingernails run across a blackboard.  (The closest I am aware of any legal scholars tapping into this area is the Virginia Law Review article by Margaret Blair (Vanderbilt, left) and Lynn Stout (UCLA, right):  A Team Production Theory of Corporate Law.)

The first lesson, a description of the lean organization, and a couple relatively simple examples of the techniques, comes below the fold.

The point of the lean organization (or lean production) was to do far more with far less, and to do it better, and often to do it without sophisticated technology.  The opposite of lean production is mass production of the kind developed in the U.S. auto industry by Henry Ford and Alfred Sloan.  Mass production accumulated huge amounts of factory floor space, on-site inventory, human effort and tooling investment.  In contrast, lean production looks at everything with a fresh eye and says "what in that process is actually unnecessary to the end product?  How can we get from start to finish faster?"  It is the power of the mind applied to the world.  And not just the mind of  an omnipotent Planner (think The Truman Show), but the collective minds of everyone involved in the process.

We need to focus here on two concepts:  "batch" versus "flow" production, and the power of any worker to stop the assembly process.  This is oversimplified, but pretty accurate.  Imagine that you are making a widget that requires four steps of machining.  First, the metal stock has to be cut.  Second, it has to be trimmed on another machine.  Third, a screw thread must be cut on a screw machine.  Fourth, it then needs to be de-burred and polished.

In mass production, the factory would have had four large rooms.  Each room would have been dedicated to a step of the process.  Huge quantities of steel rods would come into the first cutting room.  The cutters would cut, and amass a huge batch of cut pieces.  Someone would transport bins of the cut pieces in the trimming room, where the batch would be trimmed.  Then someone would transport the bins of trimmed metal stock to the screw machine room.  And so on, until it got to the finished inventory room.

There are at least two problems presented by this system, one of which relates to quality, and one of which relates to speed.   First, if you run a large batch of metal cutting, but there is a consistent defect (the machine settings were off just a couple microns), you certainly don't discover it until you have finished (and probably wasted) a whole batch.  Now it's possible you might rework them, but that is costly, and may not be effective.  And then you've wasted time as well.

Second, it's going to be a long time before you see a finished widget.  Work-in-progress just sits in bins waiting to be processed.  Imagine that it takes a day to get a big enough batch made to make it worth while to push the bins to the next room.  The "cycle time" to make a widget is as much as four days - that's how long it takes a piece of metal to go into the factory on one end and to come out on the other end as a widget.

The solution to the problem begins with what is called "flow" manufacturing.  Rather than have four rooms, each dedicated to a particular part of the process, and holding twenty of the respective cutting, trimming, screw, and de-burring machines, you set up twenty "flow cells" in which a piece goes immediately from cutting to trimming to screw threading to de-burring.  (This means, by the way, that the work is not as monotonous, because the people in these independent cells are now accountable for the complete production of a widget from start to finish.)  You know immediately if there has been a problem cutting, because the cut piece won't fit on the trimming machine (this is called "poka-yoke" in Japanese; it is also referred to as "idiot-proofing" or "fool-proofing" the process).  And because you are empowered to stop the production line if there is a defect, you immediately find the cause, correct it, and avoid making a batch of bad stuff.  And to top it off, the "cycle time" from starting with a piece of metal to a finished widget goes from four days to about thirty minutes. 

For all you lawyers in the audience, think about a huge printing, collating, and stapling project.  The same principles apply.  If you were assembling a brief that needed to be filed in court in twenty minutes, with copies to be distributed to thirty parties, would you use a batch method or a flow method?  Or think about the process of dictating a memo.  (Does anybody still do that?   When I started out as a lawyer, the cutting edge word processing was just moving from the IBM Selectric typewriter to what were known as MAG cards.)  If you do your own typing on a computer, you eliminate the process of having a secretary interpret your dictating, having him make mistakes in the typing, reviewing the first draft, making more changes, etc. etc. etc.  And the cycle time from start to finish is faster.  And, uh oh, you don't need a secretary!  (The ultimate in this was a really old school fellow who would write the memo or brief out longhand on a legal pad, then read it to his secretary, who would then take shorthand of the dictation, and go back to her typewriter to type the draft.)

Finally, and we begin to come back to the point about auditing legal bills, we begin to recognize that inspection of the finished products is far more important in batch processing than flow processing.  Why?  Because in flow processing, we have (at least in theory) already been addressing the root cause of the problems that make the defects, rather than letting the defects occur, and then trying to inspect our way to finding them.

See the analogy to auditing a legal bill yet?  More to follow.  In the meantime, the classic work on this subject is The Machine That Changed the World, by James Womack, Daniel Jones, and Daniel Roos, the story of a five year MIT study in the 1980s of the global automotive industry, and the tectonic change from mass production to lean production.

*Let me make it clear that my skepticism about expensive document retention programs does not indicate a lack of seriousness about potential evidence spoliation.  Indeed, one of my problems was I could never figure out how to undertake the program without a risk of spoliation of evidence somewhere in some case.  Most big companies have some cases going on somewhere.  On the theory that it's always the cover-up that is worse than whatever happened to create the litigation (meritorious or not), I decided that I did not want our innocent decision to put a program into place subject to second-guessing by a judge on an ex post standard, in the context of an adversarial proceeding, and without an ironclad safe harbor.  I will talk more specifically about these programs in later posts.