Designing Your Workforce for the Never Normal: Four Lessons from UNLEASH World Paris 2026

UNLEASH World Paris 2026 brought thousands of HR, talent and business leaders to Paris Expo Porte de Versailles from 20 to 22 October. Our team split up to cover as many sessions as possible, and three of them kept returning to the same question: how do you make good workforce decisions when everything is changing at once?

Here is what stood out.

Uncertainty is no longer a phase

Do not wait for things to settle down. That is the thinking behind what Peter Hinssen of nexxworks calls the Never Normal: a world where change no longer arrives every few years but compounds every few weeks.

His Uncertainty Principle flips the usual instinct. The leaders who win, the idea goes, are not the ones who manage uncertainty but the ones who learn to use it. The pace is real: the session pointed to a Silicon Valley founder with 300 employees, only 30 of them human, as a sign of how quickly the shape of an organisation can change. Scalable learning, rather than scalable efficiency, was framed as the new competitive advantage, and the organisations still standing in 2040 will be the ones that spotted signals early and moved before they had permission.

As he wrapped up, Hinssen's message was blunt: this is not the time to be an innocent bystander. "Learn how to love uncertainty."

For anyone designing a workforce, the implication is hard to miss. If change is constant, a plan that is refreshed once a year is already out of date. What matters is how quickly you can spot the signals that matter and adapt your design in response.

The CHRO is becoming a risk partner

Few roles have changed as much as the CHRO's. Natalia Wallenberg of Ahold Delhaize, Amanda Poole of Bristol Myers Squibb and Andrew Davis of Sony Music Entertainment, in a conversation led by David Green of Insight222, described a job that has moved much closer to the center of the business. The pandemic turned HR leaders into something like a chief medical officer, and the wellbeing era that followed made them chief wellbeing officers. Now the pull is toward risk: owning the conversation about what is happening to work, what it means for workers and what it means for the workforce, while acting as the steward of how the business invests in its people. One panelist suggested the next title might be "chief adoption officer", because building adoption of new technology is becoming a capability in its own right.

Leading through that comes down to a few basics. Managers sit at the center of it, because they are the ones who help teams manage workload, try new ways of working and make sense of change. Values and culture are the anchor: when everything else is moving, purpose and mission are what organizations can hold on to, and they are not up for compromise. And because the environment is complex, the job is to keep things simple, communicating plainly so people can see where they fit, and having the courage to ask difficult questions and disagree rather than being too nice.

AI sharpens all of this. People do not want AI to happen to them, they want it to happen for them, and the organizations that win with it will not necessarily be the ones with the best models but the ones that build adoption and trust into the culture. That makes transformation a question of trust more than technology: the pace of change has to match the trust you have built.

The advice for CHRO peers started with a question to ask yourself: are you up for this moment? People can tell if you are not. From there, revisit your strategy, including your footprint, your approach and your workforce planning, and ask your business partners what the business is trying to accomplish and the single most important thing you can do to help.

Skills data only creates value when it changes a decision

Skills data on its own creates no advantage. That was the message from Jaguar Land Rover's Gill Aggus, speaking with Adam Gibson of The Adecco Group about turning capability risk into business advantage. Adam's research made the gap clear: AI proficiency is fast becoming the most sought after skill, yet fewer than half of organizations use skills data to shape hiring, development and deployment. Only around four in ten have a clear view of their current and future skills, and around a fifth use AI insight to inform headcount decisions. Seeing skills in more detail is not the same as making better decisions, so the starting point has to be the leadership decision you need to make, with the skills data scoped to answer it.

At JLR, that starts with strategy. Design is core to the brand, so the team used skills data to find the roles and skills that matter most to that capability, and directed investment in pay and development toward them. The move to a modern luxury positioning raised a bigger question: what does luxury mean across the whole organization, from vehicle and materials design to how cars are sold and serviced, and which roles need new skills as a result? The taxonomy has to evolve with the business rather than stay fixed.

The same logic runs through demand modeling. Manufacturing headcount is largely mathematical, built on people per vehicle with tolerances for different scenarios. More complex areas are modeled at job sub family and seniority level, then tested with business leaders: where will demand grow, where will efficiencies reduce the need, and who can be reskilled? Gill's approach is to get as far as possible with data, take a rough answer to the business and refine it together, and she was open that JLR's data will never be perfect. Adam's point was that this is about scenarios and visible assumptions, not false precision.

AI is changing the picture too, but task by task before job by job. Robotics moved welders into maintenance roles over about 20 years, and Gill sees AI doing something similar to knowledge work now. Her example was an Excel formula: AI wrote one perfectly and another completely wrong, which is why the skills that matter increasingly are asking the right questions and checking the output quickly. In areas like design, you still need strong experts to validate what AI produces.

Then comes the question of how to fill the gaps. JLR works with the 7 Bs: build, buy, borrow, bind, bounce, balance and bot. Build is the default, with strong investment in early careers and internal development, while buy comes in when the market or the strategy demands it, and the team tries to weigh the cost of each in close to real time. High graduate attrition led them to take another look at where graduate budget was going, and at how to retain people by showing lateral development paths as well as hierarchical ones.

None of it works unless people can see the value. For leaders, that is hard because so much of the work is about avoiding problems that never visibly happen, such as panic hiring and redundancy costs. For employees, sharing skills data has to benefit them: at JLR, it surfaces internal moves people would not have considered, such as manufacturing into engineering, which breaks down historic silos and keeps knowledge in the business.

For organizations without a clear role architecture, Gill described how JLR went from around 8,000 job titles to a much smaller framework, and said you can start top down with job families or bottom up with skills, as long as the two meet in the middle. Adam's addition was the question to ask before anything else: why do you want a skills journey, and what business problem does it solve? He also noted that external labor market constraints are a factor plans cannot ignore.

‍

What this means for designing your workforce

Three very different sessions, and a few common threads for anyone designing a workforce for what comes next.

Start with the business question. Data is only as valuable as the decision it informs. Whether it was skills data at JLR or the CHRO role in the boardroom, the strongest advice was to be clear about the decision you need to make first, then bring in the data that answers it.

Signals matter more than forecasts. In the Never Normal, the advantage goes to the organisations that see change early. That means looking beyond internal data to the external labour market, which was a constraint Adam Gibson called out explicitly.

Design for scenarios, not certainty. Imperfect data is fine, hidden assumptions are not. Get as far as you can, test it with the business, compare the scenarios that could play out, and keep refining.

People have to be part of it. Adoption, trust and experience came up again and again. Whether it is skills data or AI tools, it only works when people can see what is in it for them. A panel on employer brand made the same point from another angle: a brand is only as strong as the experience behind it, and trust is won or lost in everyday moments.

At TalentNeuron, this is the conversation we want to be part of. With AI, shifting skills and economic change all moving at once, organisations need to design and adapt their workforce, and compare scenarios for whatever comes next. We help by combining what you know about your own workforce with external labour market intelligence.

Want to talk about how to design your workforce for what comes next? Get in touch with our team.