Data on percentage of successful startups that use monorepos?

Question

As part of my job, I realized that many of the most successful companies today such as Google and Meta actually started off with monorepos and never abandoned them. This type of very early decision can have huge impact on engineering velocity and I'm wondering if it might be something that ultimately makes some software companies more successful than others.I also noticed as I recently bounced from one job to another, that some companies partition responsibilities such as CI/CD, SRE, DevOps, and (now) MLOps into separate teams while others do not.Does Y-Combinator collect any data at all on success rate versus these sort of early org decisions? It's can be enormous pain across an organization to partition new teams or switch from polyrepo to monorepo and vice-versa, so there must be some data on this right?

arthurcolle · Accepted Answer

I think that slicing the universe of tech/startup companies into categories like 'successful' and 'not successful' and then creating some kind of causal story about how, maybe, 'the company in question uses a monorepo as part of their infrastructure mythology' is ultimately flawed, and puts way too much emphasis on historical artifacts that may have some bearing on the overall success of an engineering organization and its culture, but that are really just incidental to the truly key elements that defines an organization's success. I'm sure that Facebook's success lies not really in it using a monorepo, but the fact that it had a very pure hacker mentality, trying new things out, going with what worked, and making large strides in real tech and not trying to come up with complexity until there was a good reason to.If anything, I could see the opposite being true. It is so easy to properly do dependency injection and split things into services that there is no real reason to have a monorepo these days.

necovek · Answer

I am pretty sure it's random: there are successful multi-repo companies as well.
There are only a handful of companies the success of which compares to Meta or Alphabet, so statistics do not apply.
As a counter point, there is Apple which almost certainly doesn't use monorepo. Or Microsoft.

Data on percentage of successful startups that use monorepos?

I am pretty sure it's random: there are successful multi-repo companies as well.There are only a handful of companies the success of which compares to Meta or Alphabet, so statistics do not apply.As a counter point, there is Apple which almost certainly doesn't use monorepo. Or Microsoft.

I am pretty sure it's random: there are successful multi-repo companies as well.
There are only a handful of companies the success of which compares to Meta or Alphabet, so statistics do not apply.
As a counter point, there is Apple which almost certainly doesn't use monorepo. Or Microsoft.