SAVEOur theory of change
Why we intervene where we do. Choose a version.
Our idea in plain words
The problem
The Earth has limits. Scientists have described nine of them. We have already gone past seven.
Some parts of the Earth can tip over, such as coral reefs, the Amazon rainforest or the ice on Greenland. Picture leaning back on your chair. It is fine for a long time. Then one millimetre too far, and you fall. When these parts of the Earth tip over, it cannot be undone for centuries.
What is missing
Many people want to stop this from happening: scientists, organisations, people working in ministries. But they often do not know each other. They work side by side without meeting. The research sits in one room, and the decision is made in another.
There are moments when a decision is still open: just before a vote, a negotiation or an important meeting. If the right information is on the right desk at that moment, it can change a lot. Later, it cannot.
What SAVE does
We bring these people together. We do three things:
- We look closely. Who knows what? Who knows whom? Who decides? We draw all this on a map.
- We bring people to one table. Exactly when a decision is still open.
- We prepare others. We show people how to talk to politicians well themselves.
Why it works
We do not have to convince anyone. Most of the people we connect already want change. What they lack is a connection to each other. And society can tip too, but in a good direction. Often a small, well-prepared group is enough for something new to take hold. That group just has to be faster than nature.
How we measure ourselves
We do not just count meetings. We ask: would this connection have happened without us? Did anyone do something afterwards? And we say so honestly when many people contributed to a success.
What we still need
Today, many contacts depend on our founder. For SAVE to work without him, we need paid positions and good tools. That is why money we can use freely, over several years, helps us most.
Donate nowCoordination infrastructure for a civilisation under systemic strain
Starting point
Two independent bodies of evidence point to the same conclusion. The first is the physical trajectory of the Earth system. Planetary boundaries research (Rockström et al. 2009; Steffen et al. 2015; Richardson et al. 2023) tracks the safe operating space across nine domains, and the Planetary Health Check 2025 places the exceeded count at seven of nine. Lenton et al. (2019) and the Global Tipping Points Report document that several tipping elements are at or approaching their thresholds; the 2025 edition concludes that warm-water coral reefs have already crossed their thermal tipping point. Unlike social systems, these physical systems do not reset once a threshold is crossed.
The second is the organisational trajectory. Isolated interventions rarely produce large-scale change without a common agenda, shared measurement and a backbone organisation whose explicit job is coordination itself (Kania and Kramer 2011). Within the climate movement, coalition secretariats such as Climate Alliance Germany and CAN Europe perform this function. Far fewer organisations are funded to perform it across the boundaries between science, civil society, media and government.
Where we intervene
We intervene at the coordination layer: the relationships, trust and information flows between actors who are already convinced of the need for systemic change but are structurally prevented from reaching each other when it would matter. This sets us apart from theories that bet on a society-wide shift in mindset, whose timeline is hard to reconcile with the physical trajectory, and complements approaches that seek to win over business leaders as first movers.
Theoretical grounding
- Leverage points: Meadows (1999) ranks changes to information flows and paradigms well above changes to parameters such as subsidies. Actor mapping, convening and advocacy training operate almost entirely at this higher level.
- Sensitive intervention points: Farmer, Hepburn, Ives et al. (2019) argue that complex adaptive systems have moments at which small, precisely timed interventions can trigger disproportionate cascades. Delivering briefings just before decisive windows is the operational form of this thesis.
- Network brokerage: Granovetter (1973) and Burt (1992) locate value creation at the seams between otherwise unconnected clusters. We take seriously the simulation-based critique by Watts and Dodds (2007) that diffusion is often driven by many well-positioned actors rather than a few influentials: brokerage must be institutionalised as a teachable method (Robinett 2014), not left as a personal gift.
- Social capital: Putnam’s (2000) distinction between bonding and bridging capital is central. Almost every actor in our portfolio has abundant bonding capital and very little bridging capital towards the others. Bridging capital is the scarce resource SAVE produces (see also Bourdieu 1986; Coleman 1988).
- The asymmetry between physical and social tipping points: Otto et al. (2020) and Centola et al. (2018, identifying a critical mass of roughly 25 per cent in an experimental setting) suggest that social systems can be tipped towards a new equilibrium by a small, well-targeted minority; Sharpe and Lenton (2021) and Nyborg et al. (2016) describe how such cascades scale upwards. Coordination speed is therefore the variable that decides whether the reachable social tipping point arrives before the near-irreversible physical one.
Outcome pathway
- Outputs: maps, briefings, convenings, trainings. Necessary bookkeeping, never the headline.
- Counterfactual connections: for every connection we broker, we record whether it would not have happened at all, would have happened later or in weaker form, or would have happened anyway, and we ask the counterpart to confirm wherever possible.
- Uptake: documented actions by the recipient, such as follow-up requests or frameworks appearing in planning documents.
- Contribution to decisions: short process-tracing cases, reported as contribution, never as sole cause.
- Impact: a lower probability that physical tipping points are crossed. This cannot be measured for any single organisation, which is why we report the four links before it.
Two assumptions carry this chain, and we test them most actively: that decision-makers are receptive when input arrives inside a live window, and that brokered connections persist once made, without SAVE in the middle.
Open questions and structural risks
- Founder dependency: much of SAVE’s brokerage credibility currently rests with one person. Flexible, multi-year, unrestricted funding is how a personal network becomes an institutional one: salaries, shared knowledge management and AI-assisted scaling of repetitive brokerage tasks.
- The attribution problem: the causal contribution of coordination to any single policy outcome is inherently diffuse. We recommend judging us on intermediate metrics rather than claims of sole causation.
- Finite trust capital: every introduction spends reputation that is only renewed through follow-through. This caps how fast the model can grow without losing quality.
References
- Bourdieu, P. (1986). The Forms of Capital.
- Burt, R. S. (1992). Structural Holes. Harvard University Press.
- Centola, D. et al. (2018). Experimental evidence for tipping points in social convention. Science 360.
- Coleman, J. S. (1988). Social Capital in the Creation of Human Capital. American Journal of Sociology 94.
- Farmer, J. D., Hepburn, C., Ives, M. C. et al. (2019). Sensitive intervention points in the post-carbon transition. Science 364.
- Granovetter, M. S. (1973). The Strength of Weak Ties. American Journal of Sociology 78(6).
- Kania, J. and Kramer, M. (2011). Collective Impact. Stanford Social Innovation Review.
- Lenton, T. M. et al. (2019). Climate tipping points, too risky to bet against. Nature 575.
- Lenton, T. M. et al. Global Tipping Points Report, University of Exeter, 2023 and 2025 editions.
- Meadows, D. (1999). Leverage Points: Places to Intervene in a System. Sustainability Institute.
- Nyborg, K. et al. (2016). Social norms as solutions. Science 354.
- Otto, I. M. et al. (2020). Social tipping dynamics for stabilizing Earth’s climate by 2050. PNAS 117(5).
- Putnam, R. D. (2000). Bowling Alone. Simon & Schuster.
- Richardson, K. et al. (2023). Earth beyond six of nine planetary boundaries. Science Advances 9(37).
- Robinett, J. (2014). How to Be a Power Connector. McGraw Hill.
- Rockström, J. et al. (2009). A safe operating space for humanity. Nature 461.
- Sakschewski, B., Caesar, L. et al. (2025). Planetary Health Check 2025. PIK.
- Sharpe, S. and Lenton, T. M. (2021). Upward-scaling tipping cascades to meet climate goals. Climate Policy 21(4).
- Steffen, W. et al. (2015). Planetary boundaries: Guiding human development on a changing planet. Science 347.
- Travers, J. and Milgram, S. (1969). An Experimental Study of the Small World Problem. Sociometry 32(4).
- Watts, D. J. and Dodds, P. S. (2007). Influentials, Networks, and Public Opinion Formation. Journal of Consumer Research 34(4).
Make the next conversation possible.
Every donation pays for preparation, travel to negotiations and the time that good connections take. Donors in Germany receive a tax receipt.