I think it depends on how you define "idea". If the idea was to automate indexing and improve search, they were hardly the first and it was not a good idea tha

I think it depends on how you define "idea". If the idea was to automate indexing and improve search, they were hardly the first and it was not a good idea that propelled them anywhere, but rather brilliant execution. If by idea you mean something like "the bulk of the pagerank algorithm" then yes, that counted for a lot, but I think fleshing out that algorithm from the concept of "automating indexing" is part of what most people would call execution and even part of what a lot of people would call programming (developping the algorithm is often a key step of writing the program), and even then they executed the details like their page design and UI choices well. They succeeded largely because they were hardworking people who executed well on a hard problem, though as PG says in the essay there was also a luck factor in play as there is in most of life.