Survivorship Bias and Growth
The best product doesn't always win. The one everyone uses wins. Later we all tell ourselves a story about why and how that product was really the best all along in an impressive display of survivorship bias. In fact, the history of technology is littered with quantitatively superior products that lost in the marketplace because they weren't well timed, well marketed, or well supported: VHS vs Betamax. Gasoline vs Electric. English vs Esperanto. Metric vs Imperial.
"Habit and user expectation remains a stronger moat than people appreciate." - Ben Thompson
Winning products tend to continue to win until major disruption or mismanagement. The reason for that is simple: everyone else has to conform before they can compete. You become the de facto standard against which other things are measured. It is hard to differentiate for the better when people have already collectively endorsed a fiction that the best solution has already been found (and, what luck, it's the one that won). Society is a dynamic part of product/market fit, and once a large number of humans have learned come to expect a certain behavior, there just isn't enough value in learning a new system even if it is better. This isn't just about switching costs between products but rather more fundamentally about interaction paradigms. Do you really think an app garden is the apex of all possible phone UIs? Is a vertical scrolling feed really the best way to stay up to date on friends? Does the QWERTY arrangement on a keyboard make any sense today?
This is the reason companies focus so much on growth. It isn't just that growth signifies a fit with a market. But rather because growth actually makes the market fit stronger. Even when the product doesn't involve network effects directly it still benefits from the habits and expectations of the population at large that accompany growth.