The decision to kill (or sunset) Salesforce CPQ is one of history's boldest product management decisions.

It was completely “gangster,” driven by a clear-eyed, ruthless commitment to the future. 

Do you remember that scene in The Godfather Part II when Michael takes Fredo fishing? That kind of gangster!

Who would risk a $700M revenue stream? That’s bigger than 99.9% of tech companies. Who has the guts to roll the dice on that? Marc Benioff.

Salesforce killed CPQ because it couldn’t keep up with where the market was heading, and their replacement plan is already showing cracks. But their legendary commitment to the AppExchange ecosystem could ultimately be their saving grace and become vital to the future of RevOps.  

When Salesforce decided to sunset Salesforce CPQ, it wasn’t a quiet phase-out — it was a hit job. It was a deliberate, unapologetic move to end a set of products that had become irreconcilable with the company's vision for the future. While the decision stunned some customers, the logic behind it is bold and telling.

The exact moment CPQ began its sunset phase, investments surged into Revenue Cloud Advanced (RCA) and agentic. Even as RCA’s execution is still evolving, Salesforce’s strategic intent was clear: they saw where the market was headed, and CPQ wasn’t it.

This wasn’t about declining revenue. 

Salesforce CPQ was still a moneymaker. But it was also a patchwork of legacy decisions, acquired technology, and years of customization layered on top of an aging architecture. Customers had bent the system in every direction, often with the help of armies of SI consultants, until it was nearly impossible to support or evolve in a consistent, scalable way.

Salesforce had a choice: keep shoring up a product that couldn’t meet the needs of an AI-first world, or clear the table and start over. They chose the latter.

Let’s be clear: Salesforce isn’t just pivoting to AI, it’s betting the company on being the data model at the heart of a new generation of AI enterprise applications.

AI needs data. The more data you are struggling with, the greater the impact. Standing up flexible, secure data models across the globe is staggeringly hard. It’s the foundation on which the $58B Salesforce empire was built.

Building a flexible application layer on top of it is even harder. Salesforce has an unbelievable record for safety and reliability that enterprise customers will not soon forget. And if they gripe about the cost of the platform, they know it’s a fraction of the cost of delivering that level of reliability and security themselves.

However, transactional AI doesn’t work well with ambiguity or edge cases. For agentic workflows — like a customer requesting a midterm ramp adjustment or a co-termed renewal — the system has to understand what's happening without relying on a human to rewrite business logic. Ideally, these agents would also be interoperable, forming a marketplace of reusable solutions. 

Salesforce CPQ and billing customers were running wildly divergent implementations. Some were deeply customized, and others barely functioned. That variance made it impossible to layer on consistent, intelligent automation.

So Salesforce made the tough call: stop selling the product that couldn’t be the foundation for an AI-native platform.

What followed wasn’t a clean replacement. Salesforce RCA is a replatformed offering — it’s really Vlocity, which is really Siebel — and it’s still carrying over assumptions from those legacy CPQ models and has struggled to meet the needs of most customers.  

The vision is to provide quoting, self-service and billing in one product. That is what Nue provides as well: seamless omnichannel flexibility on Salesforce.  It’s not just CPQ, it’s much more.

However, RCA is struggling. The workflows aren’t fully reimagined for agentic automation (or even humans, for that matter). Interfaces for midterm changes, proration, or pricing ramps are clunky or missing. Implementation partners are struggling. Some large SaaS prospects can’t even get a reference customer in their industry. 

The result? 

A credibility gap and a vacuum of capability at the exact moment when companies are trying to retool their revenue infrastructure. Meanwhile, Nue is winning over many ex-CPQ and Revenue Cloud customers and has prominent reference customers from startups to OpenAI and public companies.

Building enterprise software is hard.  We have great respect for any team that takes on as big a challenge as RCA and Nue are embracing. Over time, RCA will no doubt improve. Nue will be the best choice for some customers. RCA for others. However, Salesforce doesn’t need to win every deal to achieve its strategic goals.

Nue and RCA are natively and exclusively on Salesforce, and so is the data!

With its AppExchange program, Salesforce has built a strategic advantage for themselves. They are fostering an ecosystem that encourages independent software vendors like Nue to innovate, even when those vendors are competing with Salesforce’s own products. It’s a bold, long-game approach. The genius of it? Nue is well ahead of RCA right now, and even if Nue continues to be the dominant revenue lifecycle management solution, Salesforce still wins — because it’s happening on their platform. 

The true battle isn’t just for a single product category — it’s to become the platform of choice for building enterprise AI agents. With AppExchange, Salesforce is ensuring they are on the front lines of the agentic revolution.

Salesforce may have killed its CPQ product and may be stumbling with its replacement, but its commitment to a robust and varied app community ensures the best omnichannel CPQ and billing solution still runs on Salesforce. 

The move was ruthless. Strategic. Necessary.

As the new crop of AI-agent vendors emerges, Salesforce will still deliver the infrastructure, protecting the data and taking the lion’s share of the revenue.

Don’t judge the strategy by the first generation of AgentForce by the way — but more on that in a future post.

Curious about what comes after CPQ? Check out Nue, the modern revenue platform built for what's next.