Making sense of vibe selling: What might this AI buzzword really mean for enterprise sales?

If you’ve come across the “vibe coding” trend and wondered what to make of it, brace yourself, because the idea has bled into sales. Suddenly, the word we use to describe our playlists might be a sales approach.

Industry trends move fast — even for the CPO of an AI-native platform. So before “vibe selling,” fully takes root, let’s pause and explore: what does it really mean, and where can it matter?

Coined earlier this year by OpenAI co-founder Andrej Karpathy, vibe coding means less writing, more reacting. Instead of us humans running at 100% CPU, we tell AI what we want, look at the output, and iterate all through conversation. Perfect for prototyping and side projects, it caught fire in developer circles because it felt accessible — even fun — to hand over the reins to AI and refine on the fly.

On the business side, we can think of it as handing over work to other autonomous agents.

But this isn’t a breakthrough in core technology. It’s breakthrough framing. Underneath the term is the same generative AI code-assist that’s been around for a while (think GitHub Copilot, Amazon CodeWhisperer), but with a cleaner, simple front end.

“Vibing” quickly jumped its original context and is expanding into other functions like marketing and sales. Depending on who you ask, vibe selling might hint at different approaches, but most see it as letting AI carry out certain activities, while we sit back and let the show unfold.

The answer is yes, but not for certain tasks. In coding, we’re interacting with a machine, so it makes sense to let the AI run the whole show at times. In sales, the work requires interacting and building credibility with humans, which is hard to vibe, let alone in live conversations.

But it’s well-suited for many other things. Writing emails is a clear example: don’t write them, vibe them. We can let a context-rich AI solution create the content, and then guide it to perfection. Gradually, the same approach takes on more complex tasks. Think of researching accounts, drafting documents, or even creating playbooks, with AI doing the heavy lifting while we vibe to the desired output. This approach resembles coding where instead of creating piece-by-piece, we describe the outcome and guide refinement through conversation. And increasingly, it will happen right where we’re working (in-platform) so it’s seamless, not separate.

It’s like having a green gopher in the background to handle the busywork and enable stronger sellers to handle the higher-order tasks.

Revenue AI — Gong’s space — is where the vibe selling narrative meets real enterprise needs. Not because the label matters. Enterprise CROs care about outcomes, not jargon after all. But because the workflows are real: AI agents are taking the grunt work off sellers’ plates so they can focus on live buyer interactions.

That’s an important question because overlap definitely comes into play. Vibe selling and agentic AI both point to the same promise: sellers focus on outcomes, while AI handles execution. Today, vibe selling activates various agents in a semi-agentic way: using prompts and approvals by humans at each stage. As I’ve argued previously, this isn’t a bad thing for enterprises.

But agents are advancing quickly and true agents won’t always wait for human guidance. Once their output earns our trust, some may operate independently. That distinction matters for sophisticated buyers who are already evaluating and deploying agents. It’s a place where we’ve got to separate signal from noise.

That’s the line to hold. My take is that vibe selling has a role in writing and research, but not yet in front of buyers. That’s the natural extension of vibe coding: offload the build, iterate on the output. And it should happen inside the technology that sellers use, perfectly embedded in their workflows.

Over time we may start trusting agents to conduct customer-facing work such as running software demos, or helping resolve timely issues. Whether we call that vibe selling or not, it does represent fully agentic workflows that are fast becoming a practical reality for the future we’re shaping.