Think in Systems: Follow the Connections
Lesson 2 of 20: Practise mapping interdependencies, recognising feedback loops, and anticipating second-order effects before making a decision.
AI-generated neural narration · Normal speed (1.0×) · Original Elect Society educational film
Take it into practice.
Open Elect GTO ↗Read the full transcript
The Decision Beyond the Decision
Elect’s mission is to realise 100% of human potential on an omnipresent scale. That is an ambition, not a proven or guaranteed result. Working toward it requires seeing people, resources, and institutions as connected systems rather than isolated pieces. A decision can succeed locally while creating trouble elsewhere. Faster service might overload the people who maintain it. More access might expose an overlooked bottleneck. Systems thinking begins with a generous question: if this change works exactly as intended, who or what will have to adapt?
Map What Depends on What
Imagine a neighbourhood library lending drills, ladders, and sewing machines. On Saturday, twelve people queue beside a shelf of returned tools. The coordinator proposes doubling online reservations to improve access. But reservations depend on available equipment, equipment depends on inspections, and inspections depend on two trained volunteers. Each successful booking also creates a future return. Sketch those links before approving the change. The website may handle twice the traffic while the inspection bench cannot. One capable component does not establish that the whole service is ready.
Find the Reinforcing Loop
Now follow a possible spiral. More reservations bring more returns. More returns create an inspection backlog. The backlog makes availability estimates less reliable, so disappointed borrowers send extra messages. Volunteers spend time answering those messages instead of inspecting tools, which enlarges the backlog again. This is reinforcing feedback: an initial change produces effects that amplify it. Reinforcing does not mean beneficial. The same structure can support growth, as reliable service builds trust, trust attracts volunteers, and volunteer capacity improves reliability. Look for the circle, not merely the sequence.
Balance Without Overcorrecting
A balancing loop counteracts a change. The library could reduce new reservations when uninspected returns exceed a threshold, giving volunteers room to catch up. Yet timing matters. If the coordinator checks the backlog only monthly, overload may persist unnoticed. If reservations are repeatedly opened and closed after every return, borrowers face confusing swings. A weekly review and a modest buffer might provide steadier control. Choose a signal close to the actual constraint, allow time for changes to work, and avoid treating every fluctuation as an emergency.
Ask What Happens Next
The immediate effect of limiting reservations is fewer bookings. A second-order effect might be that people with flexible schedules secure most remaining slots, while shift workers lose access. Another might be fewer repair projects completed across the neighbourhood. These outcomes are possibilities to investigate, not predictions to announce as facts. Every boundary leaves something outside the map. Ask whose experience is missing, what behaviour the rule encourages, and where costs could move. Protecting volunteer capacity matters, but so does distributing access fairly. Neither goal automatically resolves the other.
Measure Without Mistaking the Mission
Suppose the dashboard celebrates completed loans. Staff could raise that number by favouring easy items and experienced borrowers, while newcomers receive less help. The metric improves; the service may not. Pair loan counts with inspection delays, volunteer workload, and feedback from people unable to book. These signals help test whether the system serves its purpose. They do not measure anyone’s actual human potential. Keep observations separate from assumptions: a shorter queue is visible evidence, while greater fairness still requires checking who gained access and who did not.
Run a Bounded Experiment
Try this exercise with a decision you face. Write the intended benefit in the centre of a page. Add five connected people, processes, or resources. Draw arrows showing dependencies, circle one feedback loop, and mark one delayed effect. Then design a small, reversible test. For the library, recruit another inspection volunteer before adding a few reservations for two weeks. Agree who can pause the trial and what workload would trigger that pause. Set a date for borrowers and volunteers to review delays, workload, and access together. Use their findings to decide whether to continue, adjust, or stop the change.
Choose for the Whole
Tomorrow, choose one ordinary decision: adding a meeting, speeding up approvals, or launching a community event. Ask four questions. What must work for this to help? What feedback could amplify or restrain it? What happens after people adjust? How will we notice harm and change course? You do not need a perfect model of everything. You need a useful map, honest uncertainty, and a manageable next step. Make sure someone has the authority to respond when evidence challenges the plan. Thinking in systems turns ambition into practical care: improving one part without losing sight of its effects on the whole.
Based on Operating stack. Explore the source material.