Designing coordination mechanisms that improved communication and collaboration between NGOs after the earthquake.
In the days following the earthquake, dozens of NGOs and volunteer groups were operating in İzmir with overlapping efforts and no shared coordination layer. The work was to design, fast, the minimum structure needed to turn fragmented goodwill into a functioning relief system.
"Every organization on the ground wanted to help — but without a shared coordination layer, effort was duplicated while real gaps went unfilled."
— Field observation, first 72 hours
There was no shortage of willingness to help — there was a shortage of shared visibility. NGOs were solving the same problems in parallel while other needs went unaddressed entirely. The challenge was to design a coordination system fast enough to matter in a crisis timeline.
The relief ecosystem was made up of local and national NGOs, independent volunteer groups, municipal responders, and the affected communities themselves — each operating on different information, timelines, and trust networks.
Rapid stakeholder mapping in the field identified where coordination was breaking down: not at the point of intent, but at the point of information-sharing between organizations that had never worked together before.
In a crisis, speed beats elegance — but speed without structure just multiplies chaos.
The strategy was to design the smallest possible coordination layer that could be adopted immediately by organizations with no shared systems, no shared language, and no time for onboarding. That meant favoring simple, visible, low-tech mechanisms over anything requiring training or software adoption.
The resulting coordination mechanism gave every participating NGO a shared, real-time view of who was covering which need, in which location — closing the gap between duplicated effort and unmet need within days.
It was designed to be handed off and run without me — a system, not a dependency.
Designing under crisis conditions stripped away every comfort of a normal process — no time for research decks, no patience for polish. What mattered was whether a coordination mechanism could be understood and used immediately by people who had never met each other and were exhausted. It taught me that good systems design isn't about adding structure — it's about adding only the structure that earns its place under pressure.