Off the Shelf: July 2021 (The Checklist Manifesto)

Of all the several books I read in July, the most impactful was The checklist manifesto, by Atul Gawande.

The Checklist Manifesto

Early in the book Gawande introduces research proposing that there are “three different kinds of problems in the world: the simple, the complicated, and the complex.” In my experience, simple problems are like the process of making pancakes. Practicing technique is helpful, but following a recipe (a form of checklist) is key to increasing your chance of success.

Gawande proposes rocketry as an example of a complicated problem. It can “be broken down into a series of simple problems” but there’s no pancake recipe to follow. While making pancakes can be done by just about anyone with an inventory of ordinary materials, a valid recipe and access to a few common cooking resources, sending a rocket to the moon involves many multiples of highly-trained people, a variety of teams, specialized materials and more expertise than can be easily collected in one place. Many of the problems can be made simple by breaking them down, but much remains complicated, and working through the logistics, timing, coordination and the results of Murphy’s Law is not straightforward.

He then proposes raising a child as an example of a complex problem. In my cooking experience, making pancakes (a simple problem) is influenced by many factors such as humidity, temperature, atmospheric pressure (or altitude) and consistency of materials. And SpaceX has shown us in 2021 that sending up a rocket is a repeatable process—complicated (with significantly many more variables), but repeatable. But even with experience, raising a child is complex — every child is unique, environmental factors are different — the list of variables is long and how they interact with one another is complex. (Not to mention that these are typically long-running projects with highly uncertain outcomes.)

Gawande then spends much of the rest of the book presenting examples and argues that even complex problems can benefit tremendously from developing the right checklist to the right problem and applying it at the right level in the right circumstances. And here is where the book gets really interesting.

Good and Bad Checklists

In one section, bad checklists are contrasted with good ones. The bad are “vague and imprecise”, too long and hard to use. They’re frequently uninformed and may tend toward too many details (micromanagement).

Good checklists, however, “provide reminders of only the most critical and important steps” and are precise and efficient, designed by informed people who are also likely to be the checklist’s users.

Pause Points

A good process that utilizes a checklist usually has at least a single (sometimes two or three) “pause point”, a strategically determined time when its user(s) must pause to review the checklist before moving ahead. These pause points seem to be unique to each flow and it appears that some trial and error is to be expected as the checklist is being designed. (Ideally, checklists should be artifacts included in the organization’s PDSA process for continuous improvement.)

These “pause points” are designed to draw attention back to the checklist, ensuring that users stay on script and don’t revert to strictly rote memory in their actions.

What Type?

Gawande highlights two types of checklists

• The “Do-Confirm” list allows the experts to utilize that rote memory but then draws them back to the checklist so they can confirm that everything was completed.
• The “Read-Do” list is designed to be consulted more on-the-go, with tasks being carried out, then checked off. These tend to be more lengthy and don’t rely as much on the user’s prior experience.

Conclusion

As I read through the book I found several places in both my professional and personal life where a checklist could make an immense difference. Some examples:

• Early in my backpacking training, I started with a packing checklist (Read-Do) but gradually drifted away as my experience level increased and my types and lengths of trips evolved. Incorporating a PDSA loop into maturing my list(s) would improve my planning and execution and would almost certainly keep me from missing a needed item on the second day of a long trek.
• At one client, the PC build process could benefit from a few Read-Do lists as the PC takes shape as well as one or more Do-Confirm lists before delivery. Here, the process could also benefit from one or more “pause points” including a “Cleared for Takeoff” check where significant steps are confirmed for completion before being handed to the end user.
• Another client’s Incident Response plans were well-designed but responders were missing critical steps once they were “in the moment”. As of this writing, I’m still not certain if a Read-Do list would be better than a Do-Confirm list, but we’ve started that discussion and will work through some ideas on table-top exercises.
• I began looking for tasks to be included in checklists that would improve the outcome with no increase in skill. Things we know how to do but don’t remember to do.

I highly recommend this book and have begun to mention it to others when I sense the need. Be certain to check out the appendix examples.

Some quotes that stood out to me:

The omissions [on the checklist] were intentional. . . . Although these are critical steps, experience had shown that professional[s] virtually never fail to perform them when necessary. So they didn’t need to be on the checklist — and in fact, shouldn’t be there. [Checklists] are quick and simple tools aimed to buttress the skills of expert professionals.

[Good checklists] do not try to spell out everything. . . . Instead, they provide reminders of only the most critical and important steps—the ones that even the highly skilled professionals using them could miss. Good checklists are, above all, practical.

---

Books and significant articles read this month: