A "flywheel" is a good way to think about CTR's relationship to rankings, because the effect isn't linear — it compounds, and it compounds in both directions.
The upward spin
A page ranking at position 6 with an unusually high click-through rate for that position sends a signal that it's more relevant than its current rank suggests. If that holds up over enough searches, the page can move up. Moving up increases its baseline visibility, which — if the page keeps earning clicks at a healthy rate for its new position — reinforces the same signal from a stronger starting point. Each turn of the wheel makes the next turn a little easier.
Where it stalls
The flywheel only spins if the page holds up its end once people arrive. A CTR boost that isn't followed by real engagement — people staying, reading, taking the next step — doesn't compound the same way. This is the part that's easy to miss: the click is the first half of the signal, not the whole thing. That's why our own service engagement window isn't just "click and leave" — it includes real time on page, scrolling, and internal navigation, because that's the half of the signal that actually determines whether a CTR gain sticks.
Why this favors patience over a one-time spike
A single burst of clicks on a page that isn't ready for the traffic can even work against you — a spike in visits followed by a spike in pogo-sticking is a worse signal than no spike at all. The flywheel model is why we talk about this as a sustained campaign rather than a stunt: the compounding only happens if the engagement quality holds up over the weeks it takes Google's systems to trust the new pattern.