Crisis leadership and the pivot: Ellen Kullman on leading through uncertainty

Executive overview

When a crisis hits, most leaders freeze waiting for clarity that never comes. Ellen Kullman — former CEO of DuPont, now CEO of Carbon — has navigated multiple crises across two major CEO roles and distilled her approach into four durable principles.

The core move: stop waiting for the situation to resolve, focus on what you can control, and actively write the trajectory you want rather than riding out events. At Carbon, this meant pivoting from Adidas shoe soles and NFL helmet liners to shipping 20,000 medical face shields a week within a month of the COVID crisis hitting.

The organisations that survive crisis are not the ones with the best playbook — they're the ones that move fastest inside the space they can actually control.

The four crisis principles

1. Focus on what you can control — Paralysis comes from fixating on the uncontrollable. Narrow the team's focus to pricing, availability, supplier and customer relationships — things where action is possible.
2. Create your own trajectory — The world will not return to what it was. Actively decide what to invest in and where to go; don't wait to "ride it out."
3. Communicate constantly — It takes roughly 15 repetitions before people believe a leader is serious. Consistency, not volume, drives belief. Listening is as important as broadcasting.
4. Centre on mission — In every crisis, anchoring people to core values and company purpose gives them hope and a focal point for their energy.

Carbon's COVID pivot

• Carbon moved from decorative and athletic applications to medical PPE within days of the crisis hitting.
• Face shield design emerged from conversations with Verily (Alphabet); within two weeks, 7,500 shields shipped.
• Production scaled to ~20,000 shields per week from Carbon's own labs alone.
• Design files were open-sourced; 300+ printer-owning customers downloaded and began producing locally — including companies in the Netherlands.
• The pivot demonstrated 3D printing's core advantage: the same printers, same materials, same network — instantly redeployed.
• Nasal swabs were a more complex case: clinical requirements, PCR machine compatibility, and FDA considerations meant a slower, more deliberate three-week process before launch — an intentional decision to "slow down to go fast."

What failed in past crises

• Painting the future picture without connecting it to the present — people can't make the leap if the gap is too wide.
• Ignoring existing culture: leaders who dismissed what an organisation had built ("that's not how I did it back at...") failed. Those who honoured the good parts while adding new elements succeeded.
• Announcing change as edict rather than experiment — framing initiatives as experiments ("let's try this and see") generates more buy-in and makes failure less personal.

How to run experiments well

• Establish a specific hypothesis before starting — "better than we are today" is not a hypothesis.
• Define what winning looks like in concrete terms; this creates shared language that depersonalises outcomes.
• Review weekly: what worked, what didn't, what changes next week.
• Common language matters — across teams, customers, and vendors — so that evaluation becomes objective rather than political.

The future of distributed manufacturing

• Traditional manufacturing requires months to retool; 3D printing collapses that to weeks or days.
• Surge capacity model: maintain a base production load, then instantly multiply output by activating the wider printer network using cloud-distributed design files.
• Geographic resilience: when a natural disaster disrupts one manufacturing site, files can be transferred and production restarted anywhere in the world.
• Short-term earnings focus has historically prevented investment in this surge capacity — the COVID crisis exposed that as short-sighted.
• The deeper shift: software and iteration are now embedded in physical manufacturing, collapsing design-to-production cycles from 18 months to weeks.