while it makes sense that companies are unlikely to want to maintain a bunch of auxiliary saas tools just because Claude Code exists, it might be the case that Claude Code massively reduces the barrier to entry for software companies, and in theory the maintenance costs as well. So while companies will still outsource a lot, their options for outsourcing could go up a tonne, so even though companies are still spending money on external options, those options see margin pressure that dilute SaaS from 70% margin in the era where building great software was hard, to ~0%, if building great software suddenly becomes trivial.
ezekg•3w ago
> ... margin pressure that dilute SaaS from 70% margin in the era where building great software was hard, to ~0%, if building great software suddenly becomes trivial.
People keep saying that, but I honestly don't see it. The problem is that this idea ignores the fundamentals of build vs buy. And it ignores the cost of time. If you have a real business and you put an engineer onto vibe-coding a replacement for a CRM that costs you $500/mo, the economics just don't work out even if they could do it in a week. You'd spend more on paying that engineer -- whose core-competency is likely not CRMs but rather your product -- to build and maintain the internal tool than you would offloading it to a vendor.
Maybe this will happen with indies, but real businesses will continue to buy, for the same reasons they opt for managed open source products that are "free" -- because real businesses understand the cost of time. And real businesses usually don't want to sell to indies anyways, for the same reason: they value their time at near-zero.
And this isn't even factoring in cost of tokens, the cost of storage, and the cost of servers! I don't see a world where building/running/maintaining software costs nothing.
r_thambapillai•3w ago
ezekg•3w ago
People keep saying that, but I honestly don't see it. The problem is that this idea ignores the fundamentals of build vs buy. And it ignores the cost of time. If you have a real business and you put an engineer onto vibe-coding a replacement for a CRM that costs you $500/mo, the economics just don't work out even if they could do it in a week. You'd spend more on paying that engineer -- whose core-competency is likely not CRMs but rather your product -- to build and maintain the internal tool than you would offloading it to a vendor.
Maybe this will happen with indies, but real businesses will continue to buy, for the same reasons they opt for managed open source products that are "free" -- because real businesses understand the cost of time. And real businesses usually don't want to sell to indies anyways, for the same reason: they value their time at near-zero.
And this isn't even factoring in cost of tokens, the cost of storage, and the cost of servers! I don't see a world where building/running/maintaining software costs nothing.