How Passion and Validation drive SaaS Success with Olga Voigt

Most conversations about product-market fit stay abstract. This one didn’t.

In this episode of the Predictable Revenue podcast, host Collin Stewart sat down with Olga Voigt, co-founder of Zibble AI, to break down what it actually takes to build and sell a product in today’s market, especially in AI.

What emerged was an entirely different lens: focus on value first, and let the product catch up.

Product-Market Fit Is Not the Starting Point

Most founders start in the same place: build the product, then find the market. That’s backward.

Product-market fit isn’t something you unlock after months of building, and it’s not a milestone you hit. It’s a spectrum, and more importantly, it’s a multiplier. When it’s weak, everything feels hard. When it’s strong, everything pulls forward.

The mistake is thinking the product gets you there.

So founders overbuild, chase features, and polish UX. They try to ship something “complete” before putting it in front of real buyers. And then they wonder why no one bites.

Because customers don’t buy products, they buy outcomes.

If your product doesn’t clearly deliver that, or worse, if you can’t articulate that, you don’t have a product problem. You have a value problem, and no amount of building fixes that.

Value-Market Fit > Product-Market Fit

The real shift is understanding the value so clearly that the product becomes secondary. That’s how Olga approaches it, she doesn’t start with features. 

She starts with three questions: 

Get that right, and everything else gets easier: your messaging, your sales, even your roadmap. Miss it, and you’re guessing.

This matters even more in AI. 

The product you build today won’t look the same in three months. Models change, capabilities expand, and features become obsolete quickly. 

If your strategy is anchored to the product, you’re forced to rebuild constantly, but value holds, it’s the only stable layer in a fast-moving stack, and it’s what customers actually pay for. You see this most clearly in how you sell.

Most founders run demos. 

They walk through features, click buttons, and explain workflows. Olga does the opposite. She starts with a simple question: “What decision do you need to make this quarter?” Then she uses the product to solve it in real time.

No pitch, no feature tour, just a real problem, worked in real time, with a tangible outcome at the end. That’s the difference between showing a product and proving value.

The Fastest Path to Traction: Sell Before You Build

You probably think validation happens after the product is built. In reality, it starts much earlier.

The path is simpler and harder than it looks: prototype, show it, gather feedback, and iterate. Then do it again. And again.

That’s what Olga did: before committing to a fully built product, they showed early versions to 30 – 40 real users. Not surveys, not hypothetical conversations, actual interactions with people who could buy.

This is where most false positives show up.

Early reactions are almost always encouraging… People say it’s interesting, they say they’d use it, some even say they’d pay for it. But none of that means demand is real.

Demand only becomes real when friction shows up.

That’s when “this is great” turns into “we’ll get back to you.”

Zibble hit that wall, too. 

Initial excitement didn’t translate into immediate adoption. They had to go deeper: prove accuracy, build trust in the AI, and validate that the new workflow could reliably replace the old one. That process took months.

This is the part most founders underestimate. Validation isn’t just confirming that people like the idea. It’s proving they’ll change behavior, take on risk, and pay for the outcome.

Service over Product Is a Cheat Code (But Comes With a Cost)

If you’re coming from services, you have an advantage most founders don’t… You already have customers, you understand their problems, and you’ve been paid to solve them. And when you introduce a product, you’re not starting from zero, you’re selling to people who already trust you.

Zibble’s first customers weren’t strangers. They were existing service clients who had already experienced the value in a different form. This is the closest thing to a shortcut you’ll find, but it’s not free.

You can’t scale a service business and a product at the same time, not with the same focus, the same team, or the same energy. At some point, you have to choose where your attention goes.

And that choice has consequences.

As you invest more into the product, the service side starts to slow down, growth stalls, and revenue may even dip. It’s uncomfortable, especially when services are what’s funding the business today. But that’s the trade.

Services stop being the main engine and start becoming fuel, supporting the thing you’re trying to build next.

Great Sales Isn’t Demoing… It’s Diagnosing.

Most founders think a good sales call means a polished demo. It doesn’t. A typical demo is a feature walkthrough: You click through the product, explain what each part does, and hope something resonates. 

The problem is that you’re asking the buyer to connect the dots themselves and figure out how this fits into their world. Most won’t. Strong sales works the other way around. 

It starts with the problem, not the product.

Instead of showing features, you diagnose a real situation: 

Then you use the product to solve that specific problem in real time.

The structure is simple:

Done well, the call doesn’t feel like a demo, it feels like a working session. The buyer leaves with clarity, not just information. That shift changes everything. You’re not pitching a tool, they’re experiencing the outcome firsthand.

The Real Signal of Product-Market Fit

Stop looking for product-market fit in the wrong places. You track signups, monitor dashboards, and analyze growth curves to validate what you’ve built. The problem is that those are lagging signals and are easy to misinterpret. The real signal is much simpler: pull.

You feel it in how people respond. 

That difference matters more than any metric because it reflects urgency, not curiosity.

You also see it in behavior. 

After most sales calls, prospects disengage, they might follow up later, but the momentum fades. 

When there’s real pull, the opposite happens: People stay engaged, they keep exploring, they start using the product beyond the initial conversation, and begin integrating it into their workflow. That’s the signal.

Product-market fit is something your customers demonstrate through urgency, continued use, and a willingness to move forward without being pushed.

AI Products Require Trust Before Growth

One of the biggest blockers in AI is trust.

Founders often assume that if the output is fast and impressive, adoption will follow. In reality, most buyers are hesitant: they worry about accuracy, hallucinations, data security, and whether they can rely on the output for real decisions.

That hesitation shows up late in the process, right when you expect momentum…

But as soon as the product gets close to real use, the questions change… 

Zibble ran into this directly. 

Initial interest was strong, but it wasn’t enough to drive adoption. They had to spend months validating the system, proving that the AI could consistently deliver reliable results and pass internal security reviews before customers were willing to move forward .

That’s the part many founders underestimate.

In AI, go-to-market is more than generating demand, it’s also reducing perceived risk. And that shifts the GTM role entirely. 

Education comes first. 

You have to show how the system works, where it’s reliable, where the boundaries are, and how it fits into existing workflows. Only then does sales become effective. If people don’t trust the output, they won’t use the product, no matter how good it is.

Founder-Led Sales Is Non-Negotiable

Early traction didn’t come from funnels, campaigns, or paid acquisition. It came from people.

Zibble’s first customers were existing relationships, clients they had worked with before, conversations they had already earned, trust they didn’t have to manufacture from scratch.

That’s how most early-stage sales actually happen.

At this stage, there is no repeatable system yet, there’s no polished GTM engine, there’s just the founder, a problem, and a series of conversations to figure out if the solution truly matters.

That’s why founder-led sales isn’t optional.

Founders are the only ones close enough to the problem to understand the nuance, adapt the message in real time, and connect the product to real business outcomes. They’re also the only ones who can navigate early objections, reshape the pitch, and feed those insights back into the product. You can’t delegate that.

Trying to hire sales too early usually creates distance from the market. The message gets diluted, feedback loops slow down, and you end up optimizing a process that doesn’t work yet.

The first phase of growth isn’t about scaling sales. It’s about discovering what actually sells. And that requires the founder in the room, running calls, asking questions, and closing the first deals themselves.

Conclusion

Product-market fit isn’t something you engineer in isolation, it’s the byproduct of consistently delivering real value to real customers.

The founders who figure this out don’t start with features or funnels… They start with problems, prove value early, stay close to the customer, and earn trust before trying to scale.

Everything else (product, sales, growth) builds on top of that. Get the value right, and the rest has something solid to stand on.

NO TIME TO READ?