One Thing at a Time

I’m a longtime subscriber to James Clear’s 3-2-1 Newsletter, and I highly recommend it to everyone.

My favorite section is “2 Quotes from Others,” and one from this past week got me thinking:

The 4th Earl of Chesterfield, Philip Dormer Stanhope, on focus:

“There is time enough for everything, in the course of the day, if you do but one thing at once; but there is not time enough in the year, if you will do two things at a time.”

Source: Letter to his son, Letter IX (April 14, 1747).

This is absolutely true. I have long subscribed to the notion that multitasking is mostly a costly myth.

In fact, research by Gloria Mark, Victor González, and Justin Harris found that interrupted work was resumed, on average, after about 23 minutes:

After an interruption, knowledge workers took an average of about 23 minutes and 15 seconds to return to their original task.

You can even check the cost of context switching.

That same 2005 CHI study found that workers spent an average of only three minutes on a task before switching to another, or getting pulled into Slack or text or email or anything else. I’d bet real money that three-minute window has compressed even further since 2005.

The research is remarkably consistent: multitasking comes with a cost. Our brains aren’t built to focus deeply on two things at once.

We even have a history of resisting technologies that multiply what one person can accomplish. Just 17 years after Stanhope wrote that letter, James Hargreaves invented the spinning jenny, a multi-spindle machine that let one worker spin several threads at once rather than one. Later versions could handle as many as 120. The productivity gain was extraordinary. So was the backlash: Hargreaves’ machines were smashed, and he fled to Nottingham.

But suddenly, AI.

AI can’t fix the deeply human problems eroding our ability to focus. But it can change how much work that focus can initiate.

So maybe the limitation was never the amount of work that could happen in parallel. It was the amount of work one person’s attention could initiate.

You’re still only focusing on one thing at a time.

The difference is that your work no longer has to.

Last week, a customer suddenly needed 672 AEM pages checked for spelling.

That’s the sort of task that would normally consume weeks of someone’s time.

Instead, I handed it to Grady with a single prompt, walked away to work on other customer needs, and came back 27 minutes later to a spreadsheet of proposed fixes ready for review.

The result:

• 672 URLs reviewed
• 83 pages with likely findings
• 98 total occurrences
• 589 pages with no likely misspellings
• Page copy, metadata, and alt text included
• No AEM content changed

The customer could then review the 98 proposed corrections and approve Grady to make only the changes they agreed with.

In a matter of minutes, you’re doing one thing a few times and setting AI off to accomplish a remarkable amount of work on your behalf. Something like:

So no, I—and you—still can’t really do more than one thing well at a time. Every shift in context still comes with a cost.

But we can now be responsible for much more work being accomplished during that same span of attention. Meetings can stop feeling like interruptions to execution because execution can continue while we plan and strategize what comes next.

For most of knowledge-work history, doing more required more human attention. Every additional task needed another person—or another piece of your already-fragmented focus.

That constraint is beginning to change.

With AI, you multiply. Focus on one task long enough to set the agent in motion, then move deliberately to the next. The work can continue in parallel even while your attention remains sequential.

AI doesn’t make human multitasking effective, just as the spinning jenny didn’t make people spin faster. Both multiplied what one person’s effort could produce in the same amount of time. AI is doing that for knowledge work instead of yarn.

So choose your focus wisely. AI may give you many more moments of choice and make each of those choices matter considerably more.

One person. One focus at a time. Exponentially more work accomplished.