Rethinking capability in an age of constant change: What changed in 6 months?
Six months ago, we brought together a group of senior leaders to talk about talent, skills, and capability. Last week, we did it again in New York.
Same format. Similar audience. Same intent.
Completely different conversation.
Six months ago: We were fixing learning
In November, the discussion was sharp, but familiar.
Leaders were wrestling with performance, but still looking at it through a learning lens.
They told us:
- Learning wasn’t translating into action
- Skills were decaying too quickly
- Managers weren’t reinforcing development
- AI had arrived, but there was no shared way to think about it
The conclusion was clear:
Learning had to evolve.
It had to become more applied. More structured. More embedded in the flow of work.
We talked about practice, reinforcement, community, how capability is built socially, not individually.
We talked about moving beyond content and toward business outcomes.
But at its core, the system still held:
Learning was the lever. We just needed to pull it differently.
Last week: We stopped talking about learning altogether
In New York, something shifted.
Not incrementally. Fundamentally.
No one debated learning design. No one mentioned course catalogs. No one asked how to increase completion rates.
Instead, leaders spoke in much more direct and uncomfortable terms:
- “We’ve invested heavily in skills, but execution still breaks under pressure”
- “We don’t know what ‘good’ looks like anymore, even quarter to quarter”
- “We don’t have time to stop and learn, and we can’t afford not to”
The conversation didn’t evolve.
It moved up a level.
This wasn’t about improving learning anymore.
It was about whether organisations can actually perform in an environment where the old models no longer apply.
The shift: From knowing to doing
If I had to summarise the biggest change:
Six months ago, leaders were asking:
“How do we make learning show up in the work?”
Last week, they were asking:
“How do we build capability across fast enough to keep up with the pace of change”
That’s a very different problem.
There’s a growing realisation across organisations:
Knowledge is no longer the constraint. The speed of skill application is.
And once you accept that, everything changes.
Capability models are evolving
One of the most striking themes last week was how quickly traditional models are losing relevance.
Career pathways. Competency frameworks. Maturity curves.
These were designed for a world where:
- roles were stable
- skills evolved slowly
- progress was linear
That world doesn’t exist anymore.
Now:
- skills are context-specific and temporary
- teams operate in constant partial readiness
- leaders make decisions without a clear definition of “good”
The goal is no longer mastery.
It’s adaptability.
Time is the new constraint
Six months ago, we talked about embedding learning into the flow of work. Last week, the tone changed.
Time is now the most expensive cost in capability development.
Pulling someone out of delivery, even for valuable learning, is seen as a trade-off with real economic impact.
Which creates a tension every leader felt:
Learning is essential.
But delivery is immediate.
That tension is forcing a shift toward:
- point-in-time learning
- applied experiences
- learning that is in the flow of work, not separate from it
AI has moved the goalposts
Six months ago, leaders were unsure how to govern AI.
Last week, that question evolved into something deeper:
What is the human role now?
As AI moves into workflows:
- some skills move out of people and into systems
- execution becomes partially automated
- the human role shifts toward judgment, validation, and decision-making
But here’s the tension:
AI increases speed, but also increases risk.
Systems that are “mostly right” still fail in ways that matter.
So the challenge is no longer enablement.
It’s control, trust, and accountability.
Leadership has become the bottleneck
This might have been the most uncomfortable insight of the entire session.
Technology is no longer the limiting factor.
Leadership is.
Leaders told us:
- their teams are more technically capable than they are
- they’re being asked to lead without clear answers
- they’re managing hybrid teams of humans and AI
For the first time, leadership effectiveness isn’t about expertise.
It’s about:
- judgment
- sensemaking
- decision-making under uncertainty
And many leaders aren’t confident they’ve been prepared for that.
The return of social learning (for a different reason)
Interestingly, one thing has made a comeback. Social and instructor-led learning. But not because digital learning failed. Because people are overwhelmed.
In a world moving this fast, leaders are actively seeking:
- spaces to compare notes
- shared understanding
- reassurance that they’re making the right calls
Learning is becoming less about content, and more about collective sensemaking.
And quietly… a new risk is emerging
The final shift is more subtle, but perhaps the most dangerous.
Speed without guardrails.
As teams experiment with AI and new ways of working:
- initiatives are happening outside formal structures
- fragmentation is increasing
- governance is lagging behind innovation
Leaders are trying to balance two competing forces:
Move fast.
Stay coherent.
And right now, most aren’t convinced they’re getting it right.
So what actually changed?
Looking back across both sessions, the shift is undeniable.
Six months ago:
We need to fix learning.
Last week:
Learning isn’t the problem anymore.
The real issue is something much bigger:
Organisations don’t have an operating system for performance in this new environment.
My takeaway
If I distill everything down to one line:
- We’ve spent years optimising how people learn.
- Now we need to redesign how people perform.
That’s a very different challenge.
And it’s the one every leader in that room is now trying to solve.
The question I left with
- If knowledge is everywhere,
- If skills are constantly shifting,
- If AI is changing the rules in real time…
What does “capability” actually mean now?
Because whatever the answer is, it won’t look anything like what we were talking about six months ago.