Seed Stage Startup Evaluation Skills
This conversation was generated using Cook'd AI's career assistant and reviewed by Fallon Han from our editorial team for accuracy.
I'm an undergraduate in Singapore and I'm preparing for a venture capital internship at a seed-stage fund. I want to practice a founder-style startup evaluation. Can you simulate an interview focused on product intuition, customer discovery, and market timing? I want to learn how an early-stage investor would think.
Absolutely. I will play the interviewer and push you like a seed-stage investor would.
Scenario: a startup claims it is building an AI tool for small finance teams to automate monthly reporting and variance explanations. The founders say they are early, have a few pilot users, and believe the market is ready now.
I will start with the same kind of question you would ask a founder:
Why does this product need to exist now, and what evidence would make you believe the timing is truly right?
I would ask what changed in the last 12 to 18 months that makes this painful enough to solve now. For example, have finance teams adopted more cloud systems, have reporting expectations increased, or has AI improved enough to reduce manual work without breaking trust? I would want evidence like repeated customer complaints, growing spreadsheet complexity, or a clear shift in workflow that creates urgency.
Good. That is the right direction, but I want you to go one level deeper.
A strong investor does not just ask whether the pain exists. They ask whether the pain is urgent, frequent, and expensive enough that customers will actually change behavior.
Follow-up: how would you test customer discovery for this idea in a way that separates polite interest from real willingness to pay?
I would not rely on generic interviews alone. I would ask the founders to talk to people who own the problem, like finance managers or heads of FP and A at small companies.
In the interviews, I would look for specifics: how they currently prepare reports, how long it takes, where errors happen, what gets delayed, and what they have already tried. Then I would ask what budget they have today, what software they use, and whether they would pilot the product or commit to a paid trial. Real willingness to pay would show up if they can name a current workaround cost or agree to next steps like data access and onboarding.
Much better. You are now connecting discovery to behavior, not just opinions.
Let us pressure-test product intuition. Suppose the founders say, "Our AI can save 10 hours a month per analyst." As an investor, why might that not be enough? What would you want to hear instead?
Ten hours saved is useful, but it may not be enough if the workflow is still risky, hard to trust, or not tied to a painful decision.
I would want to hear about the highest-value use case. For example, does it help a team close books faster, catch material errors, or explain variance in a way that improves management decisions? I would also want to know if the product fits into an existing workflow with low switching cost, and whether the output is trusted enough to be used in actual reporting rather than just as a draft.
Strong answer. You are thinking like a seed investor now.
Final founder-interview question: if the company says the market is huge, how would you judge whether this is a real seed-stage venture opportunity or just a nice product with limited pull?
I would assess whether there is a narrow but urgent beachhead market with clear expansion potential. If the first customer segment is too broad, it may be hard to win.
I would look for a specific wedge, such as startups with lean finance teams or cross-border SMBs with messy reporting. Then I would ask whether the founders have a believable path from one repeatable use case to a larger platform over time. If customer pain is strong, onboarding is simple, and there is a path to repeatable sales, I would view it as a real venture opportunity. If not, it may be a useful tool but not a fundable seed story.