Removing Bottlenecks to Productivity and Profit

Written By John E. Grant  |  Productivity  |  1 Comments

Note: I originally wrote this post for Lawyerist back with it was primarily a blog site. At some point they re-orged their content and this one didn't make the cut, so I'm re-publishing it here.

Everybody wants productivity and profit, and there are many theories on how to achieve it. Companies spend billions of dollars trying to sell you the promise of them.

Most of the time, those products aren’t actually making you more productive or profitable. Instead, they only produce what renowned management theorist Eli Goldratt called local efficiency: an improvement to a part of a workflow that doesn’t translate to an improvement in the business overall.

Find Your Bottleneck

There is typically only one bottleneck constricting the flow of an entire system (occasionally, but rarely, there are two) according to Goldratt’s Theory of Constraints. There may be several parts of our workflow we think are bad, but only one can be the worst. A chain is only as strong as its weakest link.

Two important corollaries follow from the single-bottleneck theory:

  1. Improving the flow of work at your bottleneck will improve the flow of work for your entire system. (We’ll come back to this since it is the key to checking your work.)
  2. Any improvement you make to a part of your workflow that is not the bottleneck cannot improve your system overall.

Since local improvements feel like overall improvements, this second corollary is harder for most people to grasp, but you can think of it like this: removing a boulder at the part of the river where you are standing will make the water look like it is flowing smoothly from that vantage point. In the big picture, however, you are probably just raising the water level at the gorge a bit further downstream.

Understand the Whole Flow

The first step to improving your legal process is to make sure you are setting out to improve the system overall, not just a small part of it. In other words, think globally, act locally.

There are, generally, only a few large-scale goals legal practices are shooting for: get more work, improve client satisfaction, and, ultimately, make more money.

Once you’ve established your broad goals, think about how your day-to-day activities work to achieve those goals. If you were so inclined, you could do some fancy Lean Six Sigma stuff and map out your end-to-end value stream, but frankly, that would be a waste of time. It is better to start with your big-picture workflow and only dive into the details when a legitimate need arises.

And that big picture workflow probably isn’t too complicated. Most law practices have a workflow consisting of roughly the following stages:

  1. Marketing
  2. New client/matter intake
  3. Initial research into fact and law
  4. Document drafting
  5. Negotiation/settlement
  6. Execution/trial
  7. Matter close-out

This is highly generic, but that’s the point. It is a heuristic technique, not a perfect one. There may be some recursive steps for litigation or complex deals, or skipped steps for simple matters, but most legal tasks fit into one of these top-level categories. Call them what you want, add or remove categories to fit your practice, but try to keep it under 10 stages at first. Otherwise, you risk getting lost in the weeds.

Make Your Work Visible

Once you have your categories, the next step is to try to find your bottleneck. This is a process of trial and error, but there are two ways to get started:

  1. Go with your gut. You know your workflow better than anyone, so you may have a good sense where your sticking points lie. But be suspicious of your gut too. Your pet peeve may not be the thing that is actually keeping you from achieving your larger goals. Better to take a late Cold War approach: trust, but verify.
  2. Give yourself better information. A better way to get started is to obtain better information, and the best way to do that is to make your work visible. Making your work visible is the best way to access what Nobel Laureate Daniel Kahneman calls System 1 Thinking, which allows us to draw intuitive conclusions about what we are perceiving.

The simplest way to make your work visible is to build a workflow kanban board. Find a whiteboard and make a big vertical column for each of your workflow stages. If you don’t have a whiteboard, any wall will work. Use sticky notes for the column headings and painter’s tape for the lines.

Now grab a pack of sticky notes and write out a note for every matter you are currently working on. You don’t need a lot of detail right now, just the client and matter will do (if you stick with kanban, you may add details later). For each matter, place the sticky in the appropriate column on your board.

Kanban workflow example for law firms.

Once you’re done, take a step back and look for patterns. Determine which columns are either especially full or empty. Full columns often indicate a bottleneck at that stage, and empty ones can mean the bottleneck lies immediately upstream. Ask yourself, do the patterns match your hunch? If so, great—you’re on track to identify your constraint. But don’t worry if not; you’ll get there.

Run an Experiment

Once you’ve got an idea about where your bottleneck is, it’s time to run an experiment to try to improve it. An experiment will tell you whether your adjustment improves the bottleneck and if the constraint to that part of your system has been relieved.

Some keys to running a good experiment on your business come from Eric Ries’s bestselling book The Lean Startup. First and foremost, you need to figure out what you’re going to measure to validate your efforts. If, for example, you hypothesize your bottleneck lies in your marketing efforts, you may decide to run an experiment to drive more traffic to your website. Before you launch that effort, however, you’d better take a baseline of your recent traffic patterns to measure against.

Second, make a minimum viable change to your efforts. Don’t spend $10,000 on a complete website redesign when you can spend a fraction re-working a single page. Remember, you don’t know for sure you’re working at the bottleneck yet. Improving a part of your system that is not the constraint is a waste of money. So come up with a simple improvement you can do as a trial balloon, and then use that trial to determine if you’re on the right track.

Finally, if you’ve run your experiment and measured an improvement in that part of your workflow, congratulations! But you’re not done. Your last step is to check your work against your big-picture goals to see if you’ve made a difference. Using the same web traffic example, say you managed to drive 100 more unique visitors to your website in a month. That’s great, but did those visits result in phone calls? Those phone calls to new clients? Were the new clients of the same quality as your existing mix?

Rinse and Repeat

Ultimately, if you move the needle on your global goals then you are probably at your bottleneck, which is great. Run another experiment on the same part of your workflow and see if you can measure another system-wide improvement. Repeat this process until your local improvements no longer generate system-wide results. At that point, your bottleneck has moved, and you need to go back to square one (albeit with an improved end-to-end workflow to show for it!)

If your local improvement doesn’t generate system-wide impacts, that’s a clear signal you weren’t working on your workflow’s true constraint. If you only made a small investment in your efforts then no big deal. You may not have improved upon your goals, but you’ve gained an important piece of knowledge, so your investment has paid dividends.

Think of your small experiments as investments of your time, effort, and resources. You could go for a big payoff, but those usually come with a big risk. It’s better to make lots of little investments, learn more about your bottleneck, and then let your improvements compound over time. It may not be as sexy, but it will almost certainly pay a higher return over time.

  • {"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}