Try OnsaI’m going to be upfront: this started as a joke.
Someone on our team said “so basically we built vibe sales” and everyone laughed. Then nobody could stop thinking about it.
If you haven’t been living under a rock, you know “vibe coding” - the term Andrej Karpathy coined in February 2025. The idea: you describe what you want in plain English, the AI writes the code, and you just… guide it. You don’t write syntax. You describe intent.
It went viral. Y Combinator started accepting vibe-coded startups. Half of Twitter was debating whether this was the future of software or the end of engineering standards.
Here’s what nobody’s talking about: the exact same thing is happening to sales.
Before I get to sales, let’s be precise about what made vibe coding real, not just a meme.
It wasn’t just “AI writes code.” People had that for years. What changed was the workflow:
The old way: You write every line. You debug syntax errors. You Google Stack Overflow answers. The skill is typing code.
The new way: You describe what you want. The AI writes. You review, iterate, guide. The skill is judgment.
The programmer didn’t disappear. They became a director instead of a typist. The barrier dropped from “can you write React?” to “can you describe what you need?”
That’s a structural shift. Not incremental.
Traditional B2B prospecting looks like this:
Step 1. Buy a ZoomInfo or Apollo subscription ($$$)
Step 2. Learn 47 filters
Step 3. Pull a list of 500 contacts
Step 4. Manually check which ones actually match your ICP
Step 5. Write personalized messages (or worse, use templates)
Step 6. Load into a sequence tool
Step 7. Monitor, iterate, repeat
That’s 6-8 hours before a single email goes out. And I should know - I did exactly this for months before building Onsa. (I wrote about my experience with Apollo in detail.)
Here’s what the new workflow looks like:
Step 1. Describe your ideal customer in plain English Step 2. AI researches, qualifies, and finds matching leads Step 3. You review and approve
That’s it. The skill shifted from “can you operate Apollo’s filters?” to “can you describe who you’re looking for?”
Sound familiar?
We considered it. For about 48 hours we were genuinely excited about the term. Then we stress-tested the idea with people who actually buy software.
The growth marketer in me loves it. First-mover SEO keyword. Viral potential. Meme arbitrage.
But here’s what a VP of Sales told me (paraphrasing): “If I got a cold email with ‘vibe sales’ in the subject line, I’d delete it before reading the second word. ‘Vibe’ signals unfinished product. I need numbers, not vibes.”
And he’s right. The people who need this product most - founders doing their own outreach, small sales teams, anyone who doesn’t want to become a ZoomInfo expert - they don’t care about memes. They care about results.
A meme is not a moat. The product is the moat.
I keep coming back to this framing:
“Describe your customer. Onsa does the rest.”
Six words. No borrowed concepts. Just the truth.
Here’s what that looks like in practice:
You paste a URL of your best customer’s website. Or you describe your ideal customer in plain English: “Series A fintech companies in the US with 20-50 employees who recently hired a Head of Sales.”
Onsa’s AI agents go to work. They research in real-time - not from a static database that was last updated 8 months ago, but from live sources. Company websites. LinkedIn. Public registries. Industry databases.
They build your ICP. They find matching leads. They write personalized outreach messages based on what they actually found - not from templates.
The result: a qualified lead list with messages ready to send. Time elapsed: minutes, not days.
One of our customers sells drone management software for construction companies. Their ideal customer: companies with licensed commercial drone operators.
Try finding that in Apollo. You’d search for “construction companies,” filter by size, and hope some of them happen to use drones. You’re guessing.
Our agents found the FAA’s public registry of licensed commercial drone operators. Then expanded: company websites, employees with titles like “Chief Remote Pilot,” their LinkedIn profiles.
The difference isn’t just speed. It’s precision. You describe the customer. The AI figures out where to look.
This is fundamentally different from how most salespeople use AI today. In our research across 36 real interviews, the top use case was still “research on prospects” - meaning people doing manual Google and LinkedIn searches, then asking ChatGPT to summarize what they found. That’s using AI as a faster search bar. What we’re building is AI that does the entire research process autonomously.
I talk to a lot of founders. Most of them hate sales. Not because they’re bad at it - because the tooling makes it feel like a second full-time job.
You shouldn’t need to become a ZoomInfo expert to find customers. You shouldn’t need to learn boolean search syntax. You shouldn’t need to stitch together 4 different tools and a spreadsheet to send 20 personalized emails.
The structural shift happening right now is this: the barrier to good prospecting is dropping from “technical sales ops skill” to “can you describe your customer.”
That’s the real change. Whether you call it “vibe sales” or not doesn’t matter. What matters is that the work is moving from execution to judgment. From operating tools to describing intent.
We launched Onsa on Product Hunt today. If you’re a founder doing your own sales, or a small team tired of paying for database subscriptions and learning complex tooling, I’d genuinely appreciate you checking it out.
Not because of memes. Because I think the way prospecting works is fundamentally broken, and we’re building something different.
You know your customer. Now go find them.
Is “vibe sales” an actual product category? No. It’s a useful metaphor for a real shift - from manually operating sales tools to describing your ideal customer and letting AI do the research. The concept is real even if the label is a meme.
How is this different from using ChatGPT for sales? ChatGPT helps you write emails or summarize research you already did. AI-powered prospecting tools like Onsa do the actual research - finding companies, qualifying leads, pulling contact info - autonomously. You describe the target, not the steps.
Does AI prospecting replace sales reps? No. It replaces the 6-8 hours of manual research and tool-switching that happens before any actual selling. The rep still reviews leads, runs calls, and closes deals. The skill shifts from data entry to judgment.
What about data quality compared to databases like Apollo or ZoomInfo? Static databases decay fast - people change jobs, companies pivot. Real-time AI research pulls current information from live sources at the moment you need it. We wrote a detailed comparison with Apollo based on 6 months of experience.
Can I try this without a sales background? That’s exactly the point. If you can describe your ideal customer in a few sentences, you can use Onsa. No filters to learn, no boolean syntax, no credit systems to manage.