The Flywheel Effect — by Dan Lynn
By Dan Lynn
One of the most useful ideas I've ever come across is Jim Collins' flywheel — from Good to Great and his follow-up monograph, Turning the Flywheel. Understand it, and you understand how great companies actually get built — and how to build one yourself.
Picture a massive, heavy flywheel. At first it takes enormous effort just to move it an inch. You push and push, and it barely turns. But you keep going — and slowly it builds momentum. Then something remarkable happens: the wheel's own weight starts working for you. It begins to turn itself, faster and faster, until it's nearly unstoppable. As Collins puts it, there's "no single defining action, no grand program… no miracle moment" — the process is more like "relentlessly pushing a giant, heavy flywheel, turn upon turn," until you hit breakthrough.
That's how good companies become great. Not one heroic move — a thousand consistent pushes in the same direction, compounding.
But here's the part most people miss, and it's the whole game: a flywheel is not a list of good things to do. It's a sequence. Collins calls it strategic compounding. Each component makes the next one almost inevitable — and, the magic part, the last one loops back and accelerates the first. That's what turns plain effort into self-feeding momentum.
The clearest example is Amazon. Collins actually coached Jeff Bezos on this in 2001, and Amazon's flywheel is legendary:
Lower prices → more customers → more sellers wanting access to those customers → bigger selection → greater scale → a lower cost structure → even lower prices.
See what happens? The end feeds the beginning. Every turn makes the next turn easier. Amazon didn't win with one big idea — it built a loop that, once spinning, powered itself for two decades.
So how do you build your own?
Find your components — the handful of things that truly drive your business. Get the order right — arrange them so each one makes the next almost inevitable. Order is everything; it's a sequence, not a checklist. Close the loop — the last step has to feed back and strengthen the first. If it doesn't loop, it's not a flywheel; it's just a to-do list. Keep it tight — usually four to six components. More than that and it's not a flywheel, it's a spreadsheet. Keep turning — and keep refining. A flywheel is never finished; as your business and the world change, it evolves. The takeaway I've carried into every company I've built and every one I advise: don't chase a scattered list of initiatives. Find the loop where each step makes the next inevitable, get the order right, and push. The early turns are brutal and feel like nothing's happening. But keep pushing — because a well-built flywheel eventually turns itself.
Questions this Key Learning answers
How do I build a flywheel for my business?
The author recommends identifying the handful of components that truly drive the business, then arranging them so each one causes the next. The final component must strengthen the first, closing the loop. Keep the sequence tight—usually four to six components—and continue refining it as the business and the world change.
How can I tell if my flywheel is actually working?
Our analysis suggests testing whether every arrow is genuinely causal. Write each link as a causal statement, challenge it, and track a leading indicator for each step rather than only watching final revenue. If one outcome does not reliably strengthen the next condition, the flywheel may be merely a diagram rather than a self-reinforcing loop.
Should I change direction when business growth still feels slow?
Not necessarily. The author emphasizes that early turns can be brutal and appear ineffective, while our analysis notes that feedback loops often contain long delays. Before changing direction or adding unrelated initiatives, distinguish a slow loop from a broken one by examining whether each step is producing evidence that it causes the next.