Systemic Change Agents
Edward Deming famously said “Every system is perfectly designed to get the results it gets”. The intent of the quote, of course, is to point out that systems produce what they’re incentivized to produce, even if their avowed intent might be something different. Embedded also is the implication that systems are hard to change, if indeed they are so beholden to their incentives.
Systems aren’t impossible to change though. In a previous piece, I made note of a particular set of stakeholders who are well-positioned to evince large-scale change on behalf of the systems they belong to. Typically, they’re associations, foundations, or other organizations who already wield some leverage in their respective systems. Whether by virtue of their convening authority, their standards-setting responsibility, or just their capacity to fund explorations in greater depth than other stakeholders, they have influence over complex systems in ways other stakeholders don’t.
A unique example of this is the GAVI Alliance (originally Global Alliance for Vaccines and Immunization), the organization that oversees and coordinates the vaccination of almost half the world’s children, primarily in poor countries. Founded in 2000, and organized technically as a secretariat, GAVI is tasked with executing the strategy of the public-private global health partnership that is funded by donors, the biggest of which are historically the UK government (through DFID), the US government (through USAID), and the Bill and Melinda Gates Foundation.
GAVI’s focus on childhood vaccination is well-justified. Vaccinations are one the most cost-effective means to positively affect human health — a small investment early in life can have enduring lifelong consequences. The early focus of GAVI centered on securing the funding to ensure consistent demand, and therefore consistent production, of needed vaccines. The second critical focus for GAVI was improving supply chain capabilities to ensure that vaccines were properly stored and transported until they could be administered to children.
Twenty years into the effort, and in the midst of developing GAVI 5.0 (the fifth iteration of their 5-year strategy), GAVI could claim some remarkable outcomes at scale (more than a billion vaccines delivered), while still seeing stubbornly poor progress in a number of regions. A common symptom for those poorly performing regions included weak infrastructure that couldn’t coordinate last mile logistics for vaccine campaigns or support vaccine delivery through existing health systems. Just as pernicious were human and cultural elements around misunderstanding and mistrust that kept families from getting the vaccines. For certain, GAVI’s one-size-fits-all approach to providing technical assistance to countries made little sense given the wide variation in circumstances and capabilities represented in individual nations.
My interaction with GAVI began in 2019. As the head for the Design Institute for Health, we had already been working with the Bill and Melinda Gates Foundation on approaches to systemic innovation in healthcare in emerging economies. They asked, as a concerned funder for GAVI, if we might apply some of our methods to help unlock new opportunities for improvement at GAVI. Importantly, GAVI had not requested the intervention, so although not unwelcome, we hadn’t been actively recruited by the organization to help. Taking an intentionally neutral posture, we offered to map GAVI’s process for managing its grant portfolio — the entire journey of engagement with the countries it assisted, from initial country application through to review, approval, funding, vaccine purchase and deployment, monitoring, evaluation, and reporting — a process that was renewed and repeated on an annual basis. Dozens of interviews with GAVI, its partners (including the WHO and other collaborators), and the countries themselves yielded a system map that looked distinctly different from what GAVI had believed their streamlined process entailed. Instead, it resembled a haphazard platter of spaghetti, with connections, exchanges, and contingencies, many of which GAVI wasn’t aware of. Particularly revealing were the unexpected activities and workarounds from the country side, something GAVI didn’t have visibility into, and certainly had not appreciated as the unintended consequences of the formal demands of their processes.
This map characterized the system as it existed, instead of as GAVI imagined it, yielding a trove of insights. Among the systemic realizations were:
- The extent to which the assessments and risk identification that were a substantive part of the annual application process were redundant and unnecessary, when most of the knowledge was already known to GAVI through the monitoring and evaluation conducted in the previous cycle (the vast majority of countries were returning applicants).
- The existence of an entire ecosystem of consultants and advisors were paid by countries to help prepare the complicated applications, therefore siphoning off important funding into a middle layer that didn’t have direct benefit to vaccine deployment.
- The wide variation in assistance needed between countries — those that had built infrastructure and were transitioning to independence needed nothing but an occasional check-in to perpetuate their ongoing progress. Other fragile or conflict countries needed comprehensive support across all activities.
- The similarity in challenges faced by countries in the same region, and the untapped ability for them to inform, educate, and advise each other.
- Collisions inside GAVI between internal groups whose roles and responsibilities often overlapped, and the wasted effort necessary to resolve those conflicts or accountability.
- The sheer volume of uncoordinated streams of communication directed at countries by the different GAVI groups, which rendered most of it useless since countries couldn’t actually divine who within GAVI they should interact with for different needs.
Some of the challenges revealed had obvious remedies that could be deployed directly within GAVI, but others involved the complex set of stakeholders in the broader ecosystem. Addressing them would require a more collaborative approach to problem-solving. Late in 2019, we designed and facilitated a multi-day workshop in Geneva which included staff from GAVI, partners (WHO, UNICEF, and others), country leaders, and funders.
Instead of the usual report-and-discuss approach, we asked the participants to develop tangible solutions in active work sessions. Each group, assembled with diverse representation across stakeholders, was given a distinct scenario with the now-identified challenges embedded as components of a complex and realistic problem. Uniquely, each member of the team was assigned a persona that was different from their real-life role, to not only induce empathy for the challenges their partners faced, but also so they could appreciate the systemic ripple effects of different proposed solutions.
The solutions built by each team were then presented back to the group as whole, where common approaches could be identified and amplified, and differences could be considered. That discussion established consensus for change, and built a foundation for the GAVI 5.0 strategy, which took a differentiated approach to countries, dependent on their need. The consequence of that differentiated strategy meant that attention and resources could be redirected to the countries that needed it most, while also resulting in a more streamlined process for countries who had been long-term partners and were on a more predictable glide path to self-sustained independence.
Even as our engagement on GAVI’s work with external partners came to resolution, it’s worth noting that the system mapping process revealed a number of internal organizational challenges that also required attention. Most notably, we continued to work with GAVI to evolve the definition of roles and responsibilities amongst its internal groups, new decision-making protocols across the organization, and a comprehensively revamped communication capability to better engage with the countries it served.
Since working with GAVI, we’ve continued to hone the methods we utilized in that work. System mapping consistently enables a full network of stakeholders to see the system together, and by doing so, to identify bottlenecks, barriers, and opportunities that cannot be seen from a single silo. The only way to chart forward progress is then to work together on real-world scenarios with real-world constraints, producing the kinds of collaborative solutions that produce beneficial effects across the network. Importantly, at the heart of those efforts is always an organization that has unique leverage on the system in which it resides.
We’re seeing similar patterns emerge in our current work — in health care delivery, clinical research, clean energy, and carbon emissions. We’d love to hear about your efforts and your learnings as we all build a canon of knowledge on making change in complex adaptive systems.
Copyright 2025 New Origin Studio. All rights reserved.
