When productivity becomes the problem: Three actions for HR leaders

Seramount recently hosted an executive boardroom at UNLEASH America 2026 focused on workforce productivity.

What emerged wasn’t a tidy playbook. It was something more honest: a room of HR leaders naming structural forces bearing down on their organizations and working toward language for problems that rarely get called by their real names.

Here’s what the discussion surfaced, and what it means for the decisions in front of HR leaders.

The real problem is visibility, not performance

HR leaders aren’t struggling with productivity. They’re struggling with visibility. That distinction matters enormously for how you act.

Before the pandemic, productivity was inferred from proximity. Butts in seats equaled value creation. The data didn’t support this, being seen felt like evidence.

Hybrid and remote work changed work models, yet most organizations never replaced measurements with something better. Anxiety filled the gap.

Research told a different story: 83% of employees report remote work helps them work more efficiently, 61% of HR leaders say hybrid has improved productivity, while Stanford research found a 33% improvement in retention in hybrid roles with no productivity loss.

Strict RTO mandates, by contrast, have been linked to a 13% increase in turnover.

There was a brief but telling moment post-COVID when productivity rose for the first time in 25 years.

Not because of better tools or mandates, but because people literally had to rethink how their work got done. Then command-and-control came back, friction increased, and the gains evaporated.

That’s not a historical footnote. It’s a direct signal about the need for clarity and that leadership matters.

This is deeper than location. When clarity about priorities is missing, employees aren’t slowing down, they’re accelerating, treating everything as urgent because they have no reliable signal about what isn’t.

Cognitive load is compounding, while decision quality/quantity/confidence is likely dropping. Time in seat is not value creation.

Activity is not impact. Visibility is not trust. The time is now to think of the stronger leading indicators of productivity and more importantly, performance.

Urgency stopped being episodic – it became the operating model

The boardroom surfaced that three distinct forces are converging simultaneously. Each alone is manageable but, together, they have made urgency structural.

Organizational acceleration: AI is being adopted faster than workflows are being redesigned to absorb it.

Leadership compression: managers are squeezed between executive pressure and team capacity, with limited authority to reset in either direction. Urgency flows down, but permission to slow down, adapt, and evolve rarely travels with it.

External load: economic uncertainty, political unrest, and generational stressors mean employees arrive depleted before the workday begins.

One leader in the room said it plainly: there is no additional capacity. The work will always be here. That isn’t a motivation problem. It is a system problem.

And burnout is the cost – an estimated $190 billion annually to U.S. employers, with burned-out workers 2.6x more likely to leave.

Eisenhower had it right: the urgent are never important, and the important are never urgent.

It’s the voice in my head as I listen to clients — and as one HR leader in the room lamented, when your highest performers are rewarded only with more work.

Urgency as an operating model blurs decision rights at every level. What looks like a productivity problem is actually a leadership culture problem hiding in plain sight.

The AI Paradox no-one wants to name

Nearly nine in 10 organizations are now requiring AI use in at least one business function. And yet only 39% report any enterprise-level impact.

The mechanism isn’t mysterious: AI accelerates whatever system it enters. If that system already has misaligned priorities, unclear decision rights, and structural urgency baked in, AI just makes all of it faster and more fragmented.

Organizations see genuine gains at the task level: faster execution on discrete work, rapid experimentation, higher output in specific functions.

But without operating model redesign, what scales is initiative overload and more activity without more progress.

Most organizations aren’t rethinking how humans and machines work together to achieve fundamentally different outcomes. They’re layering AI onto existing workloads and calling it transformation.

The ones actually seeing enterprise-level returns are doing something different: treating AI as a catalyst for redesigning how work gets structured, where and how decisions get made, ‘who’ is accountable, and where human judgment is critical for governance.

Three HR decisions that actually move the needle

The boardroom closed with a practical framework from Seramount’s change management practice — and a case study worth noting: roughly 40% of managers already believed untapped capacity existed in their organizations, without adding headcount or tools. The capacity was there. The clarity wasn’t.

The HR  decisions that matter in the next 90 days aren’t about tools or mandates. They’re these:

Decide what you’re measuring. Stop treating lagging output metrics as accountability and start measuring the conditions that make performance possible and encourage potential, innovation, and continued growth: clarity of priorities, cognitive load, collaboration quality, employee energy, contribution.

Decide where clarity comes from. Productivity resets don’t fail because of budget or tooling. They fail because friction stays hidden and alignment is assumed rather than built. Surface the hidden frictions at the individual, team, and organizational level. Connect people to their work and its impact. Give managers the language, guardrails, and agency to coach toward outcomes — not activity. That’s where your COVID-era productivity gains went. That’s how you get them back.

Decide what AI is actually for. If your AI strategy is ‘deploy tools faster,’ you’re accelerating the wrong thing. The organizations seeing real returns are allowing redesign of the work before deploying the tools. That sequencing should not be optional. It’s the variable that determines whether AI becomes a force multiplier of architected consequences (i.e., intended outcomes) or a very expensive layer of inherited consequences.

Productivity isn’t the problem, the system producing it is. And that’s a very different, and far more tractable, thing to fix.

All statistics included in this article, unless otherwise mentioned, were part of the boardroom discussion.