What makes a Seed VC say “no” in 30 minutes?

From user research to stress-tested founder scenarios, Partech's Alison Imbert reveals what separates fundable conviction from AI-generated shine at Seed stage.

AI has made it easier than ever to build something that looks impressive. Polished demos, sleek interfaces, and convincing prototypes can now be produced in days, not months.  But how do investors cut through the shine to identify a product with real value, a defensible technology, and a team capable of building something that lasts?

To find out, I spoke to Alison Imbert, Partner at Partech. 

From Marseille to Partech: an accidental path into VC

Partech is a global venture capital firm with roots dating back to 1982, originally founded in San Francisco and now headquartered in Paris. Today, it operates as a multi-stage investment platform backing technology startups from seed to growth, with offices across Europe, Africa, and the US. The firm manages roughly €2.5 to €3 billion in assets and supports a portfolio of more than 200 companies across 40 countries.

Partech operates under a “full-stack” model: it runs multiple funds covering early-stage venture, growth equity, and impact investing. This enables it to stay involved with companies as they scale. Its investment focus spans sectors like fintech, enterprise software, AI, healthtech, and data platforms, with a strong emphasis on software and digital infrastructure. 

Imbert grew up in Marseille, coming from a working-class background. She made her way into engineering school in Paris because she loved maths. She began her career in consulting, working as a Senior Consultant at Roland Berger in Paris for 4 years. She focused on private equity and data science projects.  After four years, Imbert returned to her technical roots, completing a master’s in data science while leading a small six-person team. Imbert transitioned to VC by pure chance.

She admits that in 2018, in France, “the ecosystem wasn’t very mature. It wasn’t really considered ‘sexy.’” 

She ended up at Partech,  managing the €53 million Paris-Saclay Seed Fund, sharing, “I loved the team — Romain, Boris, the energy. The partnership was very direct, very challenging, but I liked it. I felt like I would learn a lot and be more concrete compared to consulting, where you do a lot of slides and long projects.”

Imbert’s role was investing in top engineering talent, a space where  Imbert already had a network. While running the fund, she invested in companies like Epsor, Joko, METYCLE, Orus, and Pennylane. 

Since then, Imbert helped raise a separate, larger fund that combines its Seed strategy with the main Partech Seed fund.

Why early conviction — and go-to-market — define the job

Imbert sees early-stage investing as a balance between conviction and challenge. It means backing founders before there is proof, while rigorously testing how they will build, sell, and defend their position as AI reshapes the landscape. She shared:

“What I love most about VC is supporting founders at the very beginning. You’re the first person to believe in them — sometimes when they only have an idea. You build very strong relationships.”

However, the relationship is challenging on both sides, with Imbert admitting that founders won’t just agree if you say something wrong, “So you have to keep learning and stay sharp.”

Imber is bullish about B2B SaaS, declaring that “AI“is an amazing opportunity to reinvent a lot of SaaS and challenge incumbents. Who knows,  maybe Salesforce won’t look the same in 10 years.

She also really enjoys the go-to-market side of startups — whether SMB or enterprise.:

“I like product too, but where I think I add the most value is on go-to-market. Recently, I’ve been focusing more on enterprise and large corporates because there’s more moat. It’s harder, but if you succeed, the asset’s value is stronger.

Today, when everyone can build quickly, the key question is the moat.”

What actually matters at Seed stage investor pitching

Investing at Seed stage means going beyond the conventional pitch decks and demos — “the demo is worth nothing. It’s so easy to build,” asserts Imbert. 

Further, the cost reduction in building means that funds have changed how they assess companies. 

Imbert likes to focus on user research: 

“AI won’t give you deep market insight — only direct conversations with customers will. I want founders who are obsessed with understanding the pain of their customers and willing to spend the next decade working on it.”

One question she always asks founders is: How many clients or prospects did you interview in the last two months?

“I don’t care about polished materials. I want to see messy notes. The depth of insight is what matters.

For example, I worked with founders who travelled across France to meet customers much older than them, working the same way for decades.

They built relationships, gathered insights, and showed real obsession with solving the problem. That’s what matters.”

While some founders can say 20 to 30 interviewees, she explained that others say 100, and they can explain what they learned. 

“Those are the founders I want to invest in. Because then you can go deeper: How is the pain articulated? Who is the buyer versus the user? What really matters in the workflow? Those insights only come from talking to customers.”

She also provided an example of a founder who pivoted not just because the market wasn’t there, but because he realised he didn’t want to sell to a specific persona in the long term. 

“That level of self-awareness matters.”

It follows that, in the long term, proprietary data becomes a real asset.

“We want to understand how founders will access and build unique datasets that improve their product over time,” shared Imbert. “That’s something we think about a lot.”

In other words, in a market flooded with impressive-looking products, the edge comes from founders who have done the work — and can show exactly what they’ve learned.

Distribution is the new moat

Imbert also cites the importance of distribution. While some founders rely on capital to dominate distribution, others develop more defensible strategies with network effects, brand, or critical access to users. 

“But founders need to understand this shift: differentiation is no longer just the product.”

When it comes to AI-embedded products, there’s a clear tension between buyers and users. Decision-makers often push for adoption — “we need AI” — driven by board-level mandates.

But on the ground, users can be far more hesitant, wary of changing established workflows or concerned about what it means for their roles. For founders, this creates a dual challenge: understanding both the strategic intent at the top and the practical resistance at the bottom. The product must solve a real, immediate pain point for users — something valuable enough that they actively choose to engage with it.

“Onboarding is critical. You can’t force change — you have to guide it,” asserts Imbert.

“I’ve seen this with payroll software. Owners want efficiency, but users resist switching from tools they’ve used for decades. Success depends on how well you manage that transition. Strong top-down demand doesn’t guarantee adoption.”

Putting founders under pressure with customer credibility

Imbert likes to test how founders interact with potential customers during user research. 

“Are they credible? Do people trust them? This is especially important with very young founders — and we’re seeing more of them now.”

Today, with many first-time and very young founders, the question of building credibility is vital, especially when selling into legacy organisations and needing to earn the trust of experienced buyers. To test this, Partech puts startups into real situations. 

“We introduce them to CTOs or CISOs from companies we know and see how they perform.”

From there, feedback becomes a key signal: were they credible? Did they handle difficult questions with confidence and clarity? The assessment doesn’t stop at live interactions.

“We also stress-test them with hypothetical scenarios — For example, how they would react to a fast-moving competitor. We look at how they think, not just what they say.”

Ultimately, what matters is how founders combine credibility with strategic thinking and adaptability — whether they can navigate complexity, respond under pressure, and evolve as the market changes.

More than one kind of smarts

It takes more than raw intelligence to build a company at scale. And inherent to this is both intellectual and emotional intelligence. Imbert asserts that founders need IQ and EQ, and loves founders who are real talent magnets —people everyone naturally gravitates toward.

“Take Arthur Waller, CEO of Pennylane. Everyone likes him. Put him in front of his team, and people like him. Put him in front of founders, investors, journalists — same thing.

He’s incredibly approachable, very articulate, and just easy to connect with. Everyone in the room wants to interact with him and be part of what he’s building. That’s the kind of profile we look for.

Some people think arrogance signals strength, but you can be very strong without being arrogant.”

Technical depth as a defensible edge

Partech increasingly invests in areas where technical depth creates a moat — cybersecurity, infrastructure, robotics. In this space, Imbert looks for founders with real technical authority — not just credentials, but recognition from their community. She cites the example of the second-time founder and CEO of Alpic, Pierre-Louis Theron (previously of  Streamroot, which was acquired by Lumen).

Alpic is building an MCP-native cloud platform that enables companies to expose their products as agent-accessible services. It provides the infrastructure for deploying, managing, and scaling Model Context Protocol (MCP) endpoints, enabling AI agents to securely discover, authenticate, and execute actions across software systems.

Imbert shared:

“I like to see signals of authority within the community. It’s a way for me to assess technical depth — if respected builders in the space are close to the founders, value their insights, and actively engage with them, that carries real weight. It’s not just about credentials or a polished CV; it’s about recognition from people who actually know what they’re doing.”

Humility matters

Imbert contends that the market is moving so fast that founders need to adapt quickly, admit when they’re wrong, and pivot.

“We’re not just looking for smart people with a plan; we want people who can learn. They need to listen to customers, employees, investors — and then make their own decisions. Being able to say “I was wrong” is critical.”

Humility aside, Imbert is equally focused on whether co-founders share the same fundamental vision for the company. She interviews co-founders separately and asks whether they’d take a €20 million exit tomorrow. 

“Misaligned answers usually end the conversation,” she admits. 

In the end, what makes a seed VC say “no” in 30 minutes has less to do with what’s been built than what hasn’t. For Imbert, the real signal lies in how deeply founders understand their users, how clearly they think about distribution, and how honestly they assess themselves and each other. 

Conviction isn’t about polish — it’s about proof of work. And if that foundation isn’t there, no amount of AI-generated shine will change the outcome.