As mentioned in my last post, law students often respond to their poor scores on a basic Word assessment by explaining to me that they need not need worry about this tech stuff because “that is what secretaries are for.” I think this is wrong for a number of reasons, a few of which I outline below:
- Legal work still entails some drudgery
- Drudgery is a rite of passage for young lawyers
- Drudgery increasingly falls on young lawyers as law firms reduce support staff
- Doing work themselves also plays into lawyers’ propensity for autonomy and urgency
- Delegation remains of critical importance
- But proper delegation requires proper oversight
- Proper oversight requires some of level of tech competence in order to:
- Delegate the right work
- Delegate to the right people
- Delegate the right way (i.e., appropriate instructions and expectations)
The Law Factory
Grunt work is a rite of passage for young lawyers toiling in a “law factory.” There is even a genre of legal reporting where the lede is the human-interest angle of the author-as-young-attorney laboring through some thankless task. For example, last month’s Atlantic article asking the question “Why Are So Many Law Firms Trapped in 1995?” begins:
After I graduated from law school, my first assignment at a large New York law firm was to assist in the discovery phase of a securities case. For 12 hours a day, I sat in a conference room jammed with bankers’ boxes full of documents, reviewing them one page at a time.
On what it is like to be a young lawyer, I can’t recommend enough the insightful and empathetic writings of The People’s Therapist, an associate turned psychotherapist. He writes of “a sweatshop where the billable hour is all that matters, no one tells anyone anything, and young associates are reduced to a fungible commodity.” (see also, here, here, here). For a lighter take, I suggest this parody video of a young lawyer who identifies a typo amidst an endless hell of due diligence and is celebrated as a conquering hero (until he leaves the office and tries to explain this grand feat to a normal human being). It’s funny because it’s true. It’s sad because it’s true. It’s a core reason that young lawyers are miserable (see here).
The notion that young lawyers are burdened with rote, routine, and repetitive work is not limited to young lawyers. More than 95% of hiring partners believe new attorneys lack key practical skills. The menial tasks young attorneys are consequently given has resulted in more than half of managing partners being able to envision their new hires replaced by Watson-like software in the next 5 to 10 years (much to Ryan’s chagrin). Young lawyers consigned to drudgery is so systemic that Vault rates legal employers on whether or not they give new recruits “substantive work.”
Everyone knows that young lawyers are handling labor-intensive work that can be augmented (if not completely replaced) by a machine. Except when they don’t. Except when someone like me comes along and suggests that lawyers (like staff) should therefore get better at using the basic technology tools of their trade. Then the condescension kicks in, and someone explains to me, “That’s what secretaries are for.” My agitated interlocutors include law students, who, as I explained last post, when presented empirical evidence of their technological shortcomings explain that this tech stuff is unimportant because the labor-intensive work will be handled by someone else. To that, I respond:
In theory, you will get no objection here (well, a few, that I get to below). For me, law is a team sport where staffing, process, and trust in allied professionals are mission critical. But I have a number of questions about how that theory is actually applied in the vast majority of legal settings.
Legal support staff will be heartened to learn that their jobs are secure because lawyers rely on them for all labor-intensive work. But this sentiment will be of little consolation to the thousands who have already lost their livelihood. Lawyers, apparently are using technology instead (see, e.g., here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here, here). In announcing staff layoffs, the firms release public statements like:
“Advances in technology have resulted in a need for fewer support staff and related services in law firms.”
“Technology has changed the legal industry – the way we work today is very different than the way we worked as recently as five years ago.”
“our need for secretaries has been substantially reduced as a result of technology and the work style of our attorneys, who themselves perform a number of functions previously handled by secretaries.”
Law firms seem to want it both ways. They lay off support staff because lawyers are using technology and then dismiss the need for lawyers to get technology training because lawyers rely on support staff.
There is less support staff. The support staff that remains prioritizes the work from partners. Both of these realities (less staff, focus on partners’ work) are all the more conspicuous during the late nights, early mornings, and long work weekends in which so many lawyers take such perverse pride. Lawyers, especially young lawyers, end up doing much of the labor-intensive work themselves.
It has been suggested to me more than once that the post-Recession staff layoffs are evidence of greed. Greed by the partners trying to boost their profits by directing a higher percentage of work to billable resources. Greed by the individual attorneys trying to pad their timesheets. While I can not disprove this hypothesis, avarice is not essential to the explanation. We can take law firms at their word that the lawyers really have changed the way they work due to technology.
Whether or not lawyers are naturally proficient at using computers (the answer is definitely not), the computer-centric model of working is conducive to maintaining independence. The tape from a dictaphone and the handwritten notes on a printed page are intended for someone else. Digitization, however, is meant to empower the user to accomplish tasks themselves. Control is attractive to many lawyers.
In a previous post, I outlined two psychological traits where lawyers most consistently diverge from the general population. High skepticism and low resilience drive a deep statu quo bias and orient lawyers’ issue-spotting prowess towards resisting change. The other traits that separate lawyers from the general population: autonomy, sociability, and urgency.
Autonomy: lawyers prefer to maintain control
Sociability: lawyers prefer not to interact with others
Urgency: lawyers want it done now
Sound familiar? Personal computers make it possible for lawyers to work independently, avoid interaction, and ensure their work gets immediate priority. There are many people, some of whom are lawyers, who would rather take ten minutes to do something themselves rather than wait two hours for someone else to get to it–even if takes that other person one minute to complete.
Combine these predilections with a lack of staff support for junior attorneys, and it is fairly easy to envision lawyers’ work habits changing over time without reference to any perverse financial incentives.
An Empirical Question
My father was an educated, curious man who also liked to imbibe when the opportunity presented itself. He took great pleasure in sitting at a bar stool and holding forth endlessly on topic after topic (yes, as this blog proves, I am very much his son). I frequently joined him. In the few years before his passing, however, I would often ruin the fun. As soon as his idle speculation entered the realm of fact (e.g., did Marco Polo introduce pasta to Italy after visiting China?), I would pull out my smartphone and let Google resolve the issue (No. That myth started as food-industry propaganda). He never much liked my adherence to the dictum that we are all entitled to our own opinions but not our own facts. His personalized factual universe was so much more congenial.
While there are normative questions to be addressed with respect to delegation (more below), these should not distract from the extant empirical reality. Law firms should have a good sense of who is doing what kind of work. They track time meticulously and have document management systems that can tell them who is spending how much time in which applications. There is no need to speculate.
Indeed, a firm that I profoundly respect not just pulled but published their document management statistics to demonstrate their commitment to data-driven change. The firm found that, consistent with the changes described above, their lawyers’ share of the keystrokes in Microsoft Word had increased from 39% to 80% between 2004 and 2014. Over that same period, the top shareholder’s share was unchanged. In 2004, the top shareholders accounted for 0% of the keystrokes in Microsoft Word, and, in 2014, the top shareholders accounted for 0% of the keystrokes in Microsoft Word. If the top shareholders had based training priorities on their own lived experience, the firm would have provided no technology training for attorneys. But these were trial lawyers interested in the evidence. They based their training priorities on the data, and their firm is better for it.
But Delegation Is Important
Yes, yes it is. Staffing is part of my Service Delivery Review because I believe that even technologically-enabled lawyers should take a team-based approach to large-scale projects. Delegation may run counter to the lawyer personality profile, but it is an important aspect of proper client service. That said, tech competence is important not only for lawyer-as-doer but also for lawyer-as-delegator. Here’s why.
What To Delegate
Delegation has costs, including communications overhead, the attendant dangers of miscommunication, waiting for the work to be returned, review of the returned work, and the opportunity costs of occupying the resource to whom the work is delegated. If technology is used properly, it can reduce work that would take hours down to a matter of seconds and thereby mitigate the need to delegate to a human. That is, it frequently requires less time and effort to delegate to the machine than it does to delegate to a human, as I hope I demonstrate in that video I always use:
Delegating to the machine can be a better option than delegating to a person (who is just going to use the machine). But the machine, like the person, requires proper instruction. Technology training is key to knowing what work to give to the machine and how to instruct the machine in completing the work.
To Whom To Delegate
I advocate for competence-based technology testing and training for lawyers and staff. There is rarely good reason to assume that either has natural facility with basic technology. In fact, one explanation for why some lawyers learn not to delegate is the designated person to whom they are supposed to delegate is spectacularly bad at the delegated tasks.
I’ve tested lawyers, law students, and staff. Of these, staff has the highest level of variance. Some staff (especially from the ranks of trainers and word processors) are amazing. Perfect accuracy, unparalleled speed. My LTA is a joke to them (one advantage of competence-based testing being that they can prove this quickly and test out of training they do not need). But these superheroes have foils–often, but certainly not always, from the admin ranks–who appear to have almost no clue what they are doing on the most basic functions of the most common desktop software.
I don’t need to imagine (because it is my personal story) a young lawyer determining they should just do everything themselves because their support is excruciatingly slow and error prone. Why have your work wait in a queue only to spend more time correcting the returned product than is required to do the work yourself? Staff quality is a key factor in delegation. Training staff is important.
This, of course, is another advantage of competence-based testing. Ascertaining who knows what is great for creating individualized training curricula and raising tech competence, but it is also foundational for team assembly and workflow design. Transparent, objective measurement of proficiency can give lawyers the confidence that the person to whom they are delegating is much better and faster than they are at the task being delegated.
Excel, as always, is my favorite example. I love Excel. I am convinced that Excel is “the most important computer application of all time.” As a data-driven, in-house lawyer, I spent more time in Excel — company financial and performance data — than I did in Word. But I recognize that my experience is not typical and concede that not every lawyer needs to be an Excel expert. Ubiquitous expertise is not what I asked of my outside firms. Instead, I expected them to (a) have a few, identified Excel experts on their team and (b) a workflow designed to direct spreadsheet-intensive work to those experts.
How To Delegate
That technology is now part of a lawyer’s ethical duty of competence is well-trod ground. (see here, here, and here). Less commented upon (at least in the sources I read) is the interplay between evolving ethical duty of competence and the rules governing delegation. Specifically, Rule 5.1 covers delegation to other attorneys (e.g., from partner to associate), and Rule 5.3 covers delegation to nonlawyers (a term I dislike but that is used in the text of the rule). While “certain tasks are delegable, the lawyer remains responsible for ensuring that those tasks are performed competently, diligently, and otherwise in conformance with the lawyer’s own ethical obligations.” Since the lawyer’s own ethical obligation of competence encompasses technology, the lawyer cannot avoid that duty by delegating the task to someone else.
Granted, these paired obligations can be challenging to fulfill in highly technical areas. While the facts of the particular were egregious,that lawyers will be sanctioned for relying on vendors during the ediscovery process has some troubling implications. But, fortunately, I do not have to deal with edge cases. When it comes to technology training, I’m primarily focused on the basics, for now.
In this regard, I will forever be in the debt of a commenter who responded to one of the first stories on my testing of outside counsel’s tech competence. This commenter has become the posterchild in my discussions of delegation. Because one area of my portfolio included managing a fair amount of small-bore litigation, I would test whether associates and staff could use the automatic Bates-numbering function built into their PDF software. One commenter to the ABA Journal piece that mentioned this was horrified that I expected an exalted J.D. to know such things:
All I am saying is that there is no need to charge a client $250+/hr for me to put page numbers on a stack of documents. I will do the substantive document review, prepare the written responses to the document requests, etc. My secretary or paralegal can do the photocopying and numbering. If a firm has the resources (and I recognize that not all do), I maintain that this would be the proper way to handle such tasks.
All I am saying is that I’m grateful. If I had presented this as a hypothetical, no one would have believed me. The lawyer was so locked into the way things had always been done that the concept of a technology-based solution was outside the realm of comprehension. Commenting on an article referencing the fact that the computer can apply Bates labels in seconds, the lawyer could only envision the laborious manual process and be appalled at the idea that a lawyer might be involved it. God bless the Interwebs!
If a lawyer knows how to use the Bates-label function, the process takes seconds to complete. But even if a lawyer doesn’t know how to use the function, the lawyer still needs to have an idea that such a function exists in order to delegate properly. There is a material difference between (i) expecting your secretary or paralegal to return digital documents in a couple of minutes and (ii) accepting scanned images after several hours. Speed and costs are a concern (copies cost and paralegals bill) but so is quality. The manual process introduces hundreds, if not thousands, of opportunities for error (e.g., two pages stuck together) and drastically reduces the quality of the documents (from digital document to scanned image).
I’ve seen that dynamic over and over again. The attorney not only doesn’t know how to complete the project in a few minutes, but also doesn’t know that the project can/should be completed in a few minutes. The support person also lacks the requisite training and resorts to the most manual, brute-force approach to getting the work done. How the work actually gets done is frequently far more important than whether or not it is delegated. The most common examples come from Word (e.g., auto-numbering, cross-references, tables of contents) and PDF (e.g., generation, assembly, footers, redactions). The most spectacular examples come from Excel (e.g., filter, sort, math functions, lookup, pivot tables).
I once encountered a paralegal who was in her second week of trying to reconcile three mammoth spreadsheets. One contained customer reference numbers and customer contact information. A second contained customer reference numbers and customer purchase information, including product serial numbers. A third contained a list of serial numbers of purchased products with respect to which the client needed to contact the customers (think warranty, class actions, recalls, etc.). The paralegal was filling out customer contact information on the third list by (i) using the Find function to locate the serial numbers on second list and (ii) then using the Find function again to locate the related customer ID on the first list. She was doing this for tens of thousands of products and was on pace to take several weeks. In a few minutes, I used Vlookup function to complete the entire project for her and reduce the opportunities for human error by several orders of magnitude. Until I came along, neither she nor the attorney who had delegated the task had any idea that she was doing anything wrong.
I harbor no expectation that everyone in the chain will be fully trained in every aspect of the software. The person actually doing the work should be trained in the relevant labor-saving, quality-improving features. The person overseeing them need not know the specifics; the delegator should have a general sense of what the work requires and how long it should take. A rough idea that the software can do X is a different level of competence than knowing how to make the software do X. But it is still a level above where most lawyers and staff currently operate.
We don’t know what we don’t know. Yet, everything is obvious once we have the answer. The people in the preceding stories are good for a chuckle if you are familiar with the functionality they aren’t. Except it isn’t funny. These are real people doing mind numbing and unnecessary work that matters. They are smart, talented, hard working, and well meaning. They just want to do a good job. They have the tools to do a good job. But they don’t know how to use those tools. This situation is unfortunate. It is also remediable. This is a problem we can solve in the near term. Part of the solution, however, entails understanding how technological competence intersects with proper delegation.
Lawyers should know how to complete the work they do not delegate. Lawyers should properly oversee the work they do delegate. Staff should know how to properly perform the work delegated. Not everyone needs to be an expert. But almost everyone would be better served by a higher tech baseline.