Everyone Is Building Their Own CRM. Here’s Why That’s Almost Always a Bad Idea.

Scroll LinkedIn for five minutes and you'll find someone announcing they've built their own CRM, their own pipeline tool, their own AI-powered sales layer. The implied message is always the same: off-the-shelf is for companies that lack ambition or engineering talent. Real innovators build.

It's a compelling story. It's also, in most cases, a very expensive mistake.

The irony is that one of the most technically sophisticated AI companies in the world — Anthropic, the team behind Claude — recently chose to scale on Salesforce rather than build their own stack. When a company whose entire business is building powerful, novel software decides that a commercial CRM is the right call, it's worth asking why.

What building your own CRM actually costs

The conversation usually starts with the licence fee. Salesforce isn't cheap, and procurement teams notice that number. What they don't model — at least not until it's too late — is the true cost of the alternative.

Building a CRM from scratch means building for:

None of this is insurmountable. But the engineering hours compound fast, and they compound before you've written a single line of code that's specific to your business model — which is the only place where a custom build actually creates competitive advantage.

Why Salesforce keeps winning

Salesforce has been solving the above problems for twenty-five years. The platform you're licensing today carries the weight of millions of customer deployments, thousands of edge cases caught and resolved, and a security and compliance posture that would take years and significant capital to replicate.

But the more interesting reason Salesforce wins isn't legacy infrastructure — it's the rate of change. The platform is moving faster than most organisations can track: Agentforce, Data Cloud, the Einstein Trust Layer, native vector database capabilities. Each release adds capability that a bespoke build would need to engineer independently, at cost, and then maintain.

The Anthropic example is instructive precisely because Anthropic isn't a company that shies away from complexity. They operate at the frontier of AI. And they looked at the build-vs-buy question and concluded that Salesforce was the right foundation on which to grow. That's not a concession — that's a clear-eyed prioritisation of where to invest engineering effort.

Where the real work — and the real value — lives

This is the point that often gets lost in the build-vs-buy debate: choosing Salesforce doesn't mean switching off your judgment or your ambition. It means redirecting both.

The organisations we see genuinely winning with Salesforce aren't the ones who've deployed it straight out of the box. They're the ones who've invested in:

Agentforce is the current front line of this. The companies deploying AI agents that actually work aren't starting from scratch — they're layering intelligent automation on top of a well-structured Salesforce foundation, with clean data, clear processes and appropriate guardrails. That's a very different problem from building a CRM. It's also a much more solvable one.

The build-vs-buy question is really a question about focus

If your core business is CRM software, build a CRM. If it isn't — and for the vast majority of organisations it isn't — the question is really: where should your engineering and leadership attention go?

The answer, almost always, is the work that differentiates you: your product, your data, your customer relationships, your go-to-market motion. Not the infrastructure layer that Salesforce has already built, hardened and continues to evolve at scale.

Anthropic chose Salesforce. Not because they couldn't build something themselves. Because they understood exactly what their engineering effort should be for.