One of the hardest challenges in public systems is that progress depends on institutions making decisions together. Even when everyone agrees on the ambition, they often begin from different assumptions, incentives and timelines.
That was the challenge we encountered during A Thousand Days to Shine, a multi-partner early childhood development initiative in western Kenya focused on strengthening nurturing care during a child's first three years of life, particularly by increasing fathers' involvement in early caregiving.
The programme brought together county governments, implementing organisations, technical experts, funders and communities around a shared ambition. While everyone agreed on the outcome, they arrived with different expectations of where to begin, different timelines and different measures of success. It had political support, experienced partners, funding and technical expertise, yet the work repeatedly struggled to find momentum.
Within the wider initiative, I led the cross-partner innovation process behind Nyathi en Mwandu ("A Child is Wealth"), the behaviour change campaign that became the public face of the programme. It quickly became clear that the challenge was not designing a better campaign, but helping capable institutions make decisions together.
Authority can convene a partnership, but it cannot carry one.
On paper, the partnership had everything it needed to succeed. The organisations involved respected one another, and our role carried credibility because it had been endorsed by the funder. Participation came easily. Ownership did not.
The turning point came when one of our county partners remarked that "we should board the bus." It became clear that trust would not be built through periodic visits from headquarters. It had to be earned by working alongside our partners.
I directed the establishment of an embedded project office within one of our partner organisations. That decision changed far more than where we worked. Conversations became more open. Decisions became shared. Trust developed through everyday collaboration rather than scheduled engagement.
Institutional credibility may bring organisations together, but ownership is earned through the everyday decisions leaders make long after a programme begins.
Stakeholders rarely resist new ideas. They resist uncertain outcomes.
One of the greatest tensions in the programme was the pressure to demonstrate visible progress while we were still investing in understanding the problem through research, behavioural insight and community engagement.
Looking back, I don't believe partners were resisting research. They were resisting the uncertainty that came with not yet knowing the answer.
The leadership challenge was therefore not defending a methodology. It was helping institutions remain committed while understanding was still emerging.
Uncertainty is often mistaken for a lack of progress. Helping partners distinguish between the two proved just as important as the technical work itself.
Leadership is often deciding what success you're willing to walk away from.
As the programme gained visibility, opportunities emerged that promised greater political attention and public recognition. They also risked shifting ownership away from communities and towards political interests.
Protecting the long-term legitimacy of the work meant declining opportunities that offered short-term momentum but threatened its independence. Those decisions were not easy, yet they kept the programme anchored to its original purpose.
That restraint mattered. The programme continued beyond changes in political leadership, with county governments strengthening their ownership and neighbouring counties expressing interest in adapting similar approaches.
Sometimes the opportunities leaders decline matter just as much as the ones they pursue.
Looking back, I realise these observations were never really about early childhood development. They were about how institutions make decisions when no single organisation can solve a problem alone. That has become one of the defining questions of my career. Whether working with governments, foundations, businesses or communities, I've found that lasting change depends less on individual expertise than on helping institutions understand problems together before deciding how to address them.