Real Leverage Lives Under the Roadmap
Why shallow optimizations stall product teams, and where deeper leverage actually lives.
Even with all the “right” moves, sometimes progress stalls. You add the feature your customers have been demanding for years. You reorganize your teams around customer segments. You update the roadmap, you refresh your targets, and you do the all-hands to get everyone aligned. Still, the outcomes fall short. Your launch is delayed. Code quality isn’t what you expected. Does this feel uncomfortably close to home?
It’s not that your teams aren’t working hard. The interventions, intended to drive adoption, improve user experience, launch new features, reduce churn, grow revenue… whatever it might be… aren’t aimed at the parts of the system that actually govern those outcomes.
It turns out many product teams struggle to apply the most effective types of leverage to achieve their desired outcomes
Wait, what?
Donella Meadows, a systems scientist and environmentalist, spent her career studying complex systems from ecological sustainability to economic development, and was a fierce advocate for clear thinking, accessible models, and values-driven interventions. Meadows is best known for co-authoring The Limits to Growth (1972).1 Limits to Growth modeled the long-term consequences of unchecked economic and population growth on the planet's finite resources, sparking global debate and laying the foundation for modern systems thinking. Her influence spanned multiple domains, grounded in a belief that to change systems effectively, we must understand where and how they resist change.
In 1999, Meadows published a paper titled Leverage Points: Places to Intervene in a System2, and she argued that not all changes are created equal. Some interventions have exponentially more power to shift outcomes than others.
What Are Leverage Points?
In systems thinking, a leverage point is a place within a system where a small shift in one thing can produce big changes in everything. A system, in this context, refers to any interconnected set of elements, such as people, tools, policies, technologies, and processes, that are organized in a way to achieve some purpose. Systems have a purpose, boundaries, dynamic behavior, and emergent properties.
Systems thinking is a way to understand how those parts interact dynamically over time.
System dynamics is a discipline within systems thinking which uses modeling to understand the structure and behavior of complex systems. You may think I forgot the 's' in 'system' but no, it's correct as is ;)
Other key concepts:
Stocks: the accumulations in a system
Flows: the rates of change in or out of those stocks
Delays: time lags between cause and effect
Balancing feedback loops: mechanisms that stabilize a system
Reinforcing feedback loops: mechanisms that amplify changes
Meadows identified twelve leverage points, ranked from least powerful (#12) to most powerful (#1). This is often depicted as an iceberg, though because so many frameworks are depicted as icebergs, I decided to spare you, dear reader. I’ll tackle one of those iceberg frameworks another time.
Below, I’ve used urban mobility as the “system” to illustrate leverage points more concretely.
* Yes, that word Paradigm. I know. It’s not the 90s anymore. If you bristle at that word, just mentally replace it with “worldview” and stick with me a while longer.
You might intuitively understand that the further down you traverse the leverage points, the bigger the shift in the behavior of the system. For example, adjusting bus fare rates (Leverage Point 12) could incentivize or disincentivize ridership, but demand for urban mass transit is generally inelastic3, with some exceptions. Increasing bus service frequency during peak demand (Leverage Point 8) and/or congestion pricing for single-occupant motor vehicles (Leverage Point 5) would intuitively have a much greater impact on rider serviceability, among other things.
The Devil’s Advocate: How Do You Know?
This all sounds great, but where’s the data?
Q: Does Meadows’ framework lack empirical rigor?
A: Well, yes. It describes influence but doesn’t predict outcomes? It is heuristic in nature. Most of the literature on this topic is conceptual or case-based.
Although the leverage points framework has been foundational in systems thinking, empirical research validating its effectiveness remains limited. I’ve been loosely engaged with this topic for 17 years, and through the research I’ve done writing this article, I can confidently state that the majority of the published work is qualitative and theoretical (and most articles are behind paywalls). The call for more empirical research is a persistent pattern in the academic literature.
However, there is some movement toward more empiricism. In 2020, Dorninger et al., conducted a systematic review of 301 empirical studies on sustainability interventions in food and energy systems. Their analysis found that most interventions focused on shallow leverage points, such as adjusting parameters or feedback loops, while few addressed deeper structural or goal-level changes. Interventions at these deeper levels, such as altering system design or intent, were significantly less prevalent despite being more likely to yield transformative outcomes. They also observed a correlation between research methodology and leverage depth: engineering and technocratic approaches tended to stay shallow, while socio-political and social-ecological studies were more likely to address deeper systemic issues. The authors argued that prevailing scientific practices may constrain our ability to identify and implement transformational interventions. They called for a shift toward interdisciplinary, reflexive approaches that address the values and structures underpinning sustainability challenges.4
Moreover, system dynamics employs a modeling approach using the concepts of stocks, flows, delays, and feedback loops to analyze and intervene in complex systems. Numerous quantitative models have shown that adjustments to these elements can substantially impact system outcomes. I have firsthand experience with system dynamics models of moderate complexity demonstrating this behavior. Donella Meadows' leverage points theory provides a powerful conceptual guide for prioritizing such interventions, but direct, large-scale empirical validation of the hierarchical ordering she proposed is still limited.
Systemic Patterns in Product Management
How does this relate to product strategy and product management?
Most product teams focus on the top of the leverage list: tweaking parameters, adjusting buffers, etc. We’re talking about reordering priorities in the backlog and the occasional (or frequent even) reorg of the team structure, but few product leaders are empowered or incentivized to challenge the goals or paradigms that shape the system itself.
Consider the tools we use to guide product development: A/B testing, velocity charts, funnel metrics. These are all useful, but they anchor us at shallow levels. An A/B test might tell you which button color performs better, but it won’t tell you whether the user journey itself makes sense (there are other methods for that).
UX research often meets the same resistance. Insights are gathered, but the system resists change. Feedback is slow to move, blocked by politics, silos, or incentive misalignment. By the time research influences the roadmap, the moment has passed.
We keep adjusting the thermostat in a house with broken windows. Many of us have worked in organizations like that, where surface-level tweaks mask deeper misalignments.
This pattern echoes what McKinsey found in its research on transformation efforts: only 26% of organizations rated their change programs as “very or completely successful.” Success correlated with deep alignment between vision, leadership behavior, and capability building, suggesting that structural change alone is not enough.5
From Tweaks to Transformations
While formal academic studies on applying Meadows' leverage points to product management are scarce (maybe non-existent), there are practitioner-driven frameworks that demonstrate the relevance and adaptability of the concept.
In 2019, Mark Bruno outlined strategies for creating leverage in product management, such as identifying high-leverage tasks, avoiding rework, building a leveraged roadmap, and fostering a culture that emphasizes leverage. These approaches align with Meadows' concept of intervening at points in a system where small changes can lead to significant impacts.6
In 2021, Roger Swannell introduced a "Leverage Points Framework for Systems-Shifting Product Management," inspired by Meadows' work, which was subsequently adopted by Tristan Harris' Center for Humane Technology. This framework encourages PMs to move beyond user-centered practices and consider systemic changes that can lead to more profound and sustainable product impacts.7
The LNO (Leverage, Neutral, Overhead) framework, proposed by Aakash Gupta8 (and in collaboration with Courtland Halbrook) offers a method for PMs to categorize their tasks based on impact. Gupta and Halbrook, building on the earlier work of Brandon Chu9 and Shreyas Doshi10, said that by focusing on high-leverage activities and minimizing overhead, PMs can optimize their effectiveness, echoing the principle of targeting leverage points for maximum system influence.
It’s encouraging that others are thinking along the same lines and I’m happy to have discovered them while researching this piece.
Below, I’ve translated Meadows' leverage points into generic product management examples. You’ll see a progression from surface/operational leverage (tweaking variables) to deeper/strategic (meaning and transformation) leverage.
Moving Towards Deeper Leverage
Most product leaders can’t unilaterally redefine paradigms or shift system goals, but awareness still matters.
Even if you can’t act on every point, you can start asking better questions. You can surface the real dynamics beneath performance, and you can reshape how decisions are made, how information moves, and how success is defined.
Here are three examples of leverage points that product leaders can influence:
Information Flows - You don’t need to restructure the company to improve signal quality.
System Goals - Too many teams equate output with success.
Mental Models - Every team holds implicit beliefs about risk. Is it to be minimized? Reallocated? Explored? Changing how a team understands risk will reshape how it builds.
Shifting a system doesn’t always require a reorg or a roadmap overhaul. The real leverage often lies in something quieter: asking better questions, tracing patterns of delay and distortion, or noticing where useful feedback never lands.
As a conceptual exercise, imagine you’re a product or UX leader responsible for the console of one of the currently dominant hyperscalers (e.g., AWS, Azure, GCP, OCI, etc.). The console, for those who don’t know, is the unified front-end that cloud developers, admins, architects, and others use to manage their infrastructure services, monitoring, billing, etc. Cloud consoles are often used to troubleshoot issues and serve as a precursor prior to programmatic management through Terraform or similar tools.
How Might You Apply Meadows’ Leverage Points?
I’ve mapped Donella Meadows’ 12 leverage points to the domain of cloud infrastructure consoles. As a former member of the leadership team for the OCI console, I’m drawing here not from proprietary insights, but from publicly observable patterns and common challenges across all enterprise cloud consoles.
Let’s explore how this maps.
This table illustrates how a product and/or UX leader might apply systems leverage thinking to deepen impact, not as an evaluation of any specific team’s decisions, but as an exploration of how systemic shifts can elevate usability, reduce friction, and better align platform capabilities with user intent.
Product leadership isn’t about managing backlogs. It’s about understanding the patterns and pressures that shape them, including customer signals, team incentives, feedback loops, and structural bottlenecks, and then intentionally designing those elements to support outcomes that matter.
Often, it starts with a simple question: What are you actually trying to change?
Donella Meadows argued that we often overlook the parts of a system that matter most. It’s not always the visible, high-effort work that drives meaningful outcomes. More often than not, it’s the structural, informational, and cultural shifts that compound over time.
How does this change how you think about product strategy in your organization?
#ProductManagement #SystemsThinking #SystemDynamics #LeveragePoints #Leadership #Innovation #OrganizationalChange
Additional Resources
For a deeper but still gentle introduction to these ideas, check out:
Introduction to System Dynamics (MIT OpenCourseWare)
Sean Daken - Principal, Map & Vector | mapandvector.com |
Meadows, D. H., Meadows, D. L., Randers, J., & Behrens III, W. W. (1972). The Limits to Growth. Universe Books. https://www.clubofrome.org/publication/the-limits-to-growth/
Meadows, D. H. (1999). Leverage Points: Places to Intervene in a System. Sustainability Institute. https://donellameadows.org/archives/leverage-points-places-to-intervene-in-a-system/
Litman, T. (2004). Transit price elasticities and cross-elasticities. Journal of Public Transportation, 7(2), 37–58. https://digitalcommons.usf.edu/jpt/vol7/iss2/3
Dorninger, C., Abson, D. J., Apetrei, C. I., Derwort, P., Ives, C. D., Klaniecki, K., Lam, D. P. M., Lang, D. J., Riechers, M., Spittler, N., von Wehrden, H., & Fischer, J. (2020). Leverage points for sustainability transformation: a review on interventions in food and energy systems. Ecological Economics, 171, 106570. https://www.sciencedirect.com/science/article/abs/pii/S0921800919310018 (paywall) | PDF
Jacquemont, D., Maor, D., & Reich, A. (2015). How to beat the transformation odds. McKinsey & Company. https://www.mckinsey.com/capabilities/people-and-organizational-performance/our-insights/the-organization-blog/leading-organizational-transformation
Bruno, M. (2019, December 1). Creating leverage as a product manager. Medium. https://medium.com/@markbruno/creating-leverage-as-a-product-manager-d55f89977d97
Swannell, R. (2021, November 14). A leverage points framework for systems-shifting product management. Roger Swannell. https://rogerswannell.com/blog/a-leverage-points-framework-for-systems-shifting-product-management/
Gupta, A. (2021, November 8). LNO (Leverage, Neutral, Overhead) framework for product managers. Aakash Gupta. https://www.aakashg.com/lno-framework-for-product-managers/
Chu, B. (2023, January 9). Applying leverage as a product manager. Black Box of Product Management. https://blackboxofpm.com/applying-leverage-as-a-product-manager-ffad4a99db24
Doshi, S. (n.d.). The LNO effectiveness framework. Dualoop. https://www.dualoop.com/blog/shreyas-doshi-the-lno-effectiveness-framework