A number of months in the past, I sat throughout from a CEO who was genuinely proud. He had simply applied an AI-powered individuals analytics platform: real-time sentiment information, predictive turnover scores, and engagement dashboards. Lovely system. His HR crew had been lower by a 3rd. “It does what they used to do,” he advised me.
Six months later, two of his highest-performing senior managers give up in the identical quarter. No flags. No warning scores. Nothing on the dashboard. Simply two individuals who had felt, for a very long time, that nobody knew them, who had lastly stopped ready for that to vary.
The associated fee? One was a crew lead carrying $4 million in consumer relationships. The opposite had spent two years grooming junior expertise. Between severance, recruiting, onboarding, and the enterprise that walked out with them, the corporate spent near $600,000 changing individuals the dashboard mentioned had been wonderful.
The system wasn’t flawed about what it measured. It simply couldn’t measure what mattered most.
The work no one can see
Sociologist Allison Pugh spent years learning the individuals we belief with our most human moments: physicians, academics, chaplains, therapists. The idea at its core could also be a very powerful factor your group has by no means heard of. Her ebook The Last Human Job argues that the one irreplaceable work people will do in an AI-saturated future is relational: empathy, attunement, and real presence. She calls this work “connective labor.”
Connective labor is the work of really seeing one other individual.
Not managing them. Not assessing them. Seeing them.
It’s the check-in that surfaces a struggling worker earlier than they spiral right into a quiet give up. The trustworthy dialog that defuses battle earlier than it splits a crew. The chief who notices, with out a dashboard or a survey immediate, that one thing is off with somebody she’s recognized for 3 years. Connective labor is invisible. It’s relational. And it’s load-bearing.
Pugh’s analysis focuses on professions we’ve at all times related to that sort of care. She doesn’t spend a lot time in center administration. However that’s precisely the place connective labor quietly holds organizations collectively, and the place it’s now beneath essentially the most stress I’ve seen in 35 years of labor.
What AI is doing to your individuals leaders
AI is automating the transactional scaffolding of the supervisor’s job. Scheduling. Reporting. Coordination. Efficiency monitoring. And what’s left after the automation strips away the legible, measurable, delegatable work is connective labor. The toughest half. Essentially the most invisible half. The half nearly no group is aware of identify, measure, or put money into.
I take advantage of the time period individuals chief intentionally, as a result of a supervisor undersells what this position requires. Folks leaders are those who create the circumstances that make work sustainable: security, belonging, problem, and connection to one thing that issues. These aren’t delicate aspirations. They’re operational necessities. Once they’re absent, efficiency erodes, turnover climbs, and battle festers quietly till it doesn’t.
This can’t be automated. Seeing an individual requires being seen in return. It’s a two-way act. A sentiment algorithm can detect the sign that one thing is flawed. It can’t sit throughout the desk from somebody and assist them work out what it’s, what they want, and whether or not there’s nonetheless a motive to remain. AI can’t encourage an individual to develop when progress is difficult. It can’t disagree kindly in a approach that exposes a blind spot and forces honesty. AI doesn’t care about our particular thriving, and it is vitally good at agreeing with us.
These actions are a individuals chief’s job. And we’re systematically making it unattainable to do.
Why it retains disappearing
Connective labor will get devalued as a result of it’s laborious to see. It doesn’t produce a deliverable. When a individuals chief spends 40 minutes on an unplanned dialog with somebody who’s struggling, nothing is logged for that effort. When she notices somebody has gone quiet in conferences and makes some extent of checking in, the result (belief rebuilt, an issue caught early, an individual who feels much less alone) has no line merchandise.
So organizations cease investing in it. They lower the slack that made it potential, stack extra reporting, cascade extra instruments, and shrink individuals leaders’ actual consideration whereas increasing their official headcount.
The U.S. Surgeon General’s 2022 advisory named loneliness a public well being disaster and pointed on to the office as one of many main locations adults search connection. That report landed a couple of years in the past. Most organizations responded with one other survey.
After which they surprise why their greatest individuals maintain leaving.
Most organizations don’t have an engagement downside. They’ve a connective labor deficit. They’ve automated across the factor that holds individuals collectively, and no one’s named it but.
What this prices
A producing consumer of mine was pleased with their lean administration construction. Every individuals chief carried about eighteen direct experiences, a sturdy HR system, weekly automated check-in prompts, and a quarterly pulse survey.
What they didn’t have was time. Their individuals leaders had been spending roughly twelve minutes per week per individual. The relationships had been cordial. Purposeful. Shallow. Once I interviewed the crew, one phrase stored developing in several varieties: invisible. Replaceable. Nice.
Nice is essentially the most harmful phrase in a corporation. Nice means nothing is on hearth. It additionally means nothing is being tended to.
When two rivals began recruiting, they misplaced seven individuals in eight weeks. Good, skilled, high-performing individuals who determined that feeling genuinely recognized elsewhere was value a lateral transfer. Their individuals leaders weren’t failing. They had been structurally prevented from doing the work which may have stored these individuals. The connective labor had been engineered out of the system. No person meant for that to occur. It occurred anyway.
What fit-for-human organizations do otherwise
This isn’t an argument in opposition to AI. I’m not nostalgic for inefficiency. I need organizations to make use of each device out there to release human capability. That’s precisely the query: free it up for what?
If the hours saved by automation get absorbed by extra reporting and extra throughput, you haven’t gained something. You’ve made the transactional elements of the job sooner whereas leaving the relational elements precisely as starved as earlier than.
Organizations which might be match for human life do three issues otherwise. They identify connective labor explicitly and deal with it as a business-critical operate. They shield the circumstances for it—individuals leaders want unscheduled margin; the sort of area that lets an actual dialog occur. And so they make it seen in how they consider and develop their individuals. If connective labor by no means reveals up in the way you acknowledge the individuals who do it effectively, you might be actively choosing in opposition to it. You’re going to get what you measure.
Pugh’s analysis was by no means actually solely about physicians and chaplains. It was about what it prices, to each events, when somebody is seen versus processed. AI won’t save your tradition. The individuals chief who is aware of her individuals will.
Let’s make sure that she nonetheless has time to do the work that solely she will do.
