How to talk through an idea with a mentor: what you're really testing

When you walk into a mentor's office with an idea, you're not really asking for feedback. You're running a test.

The best mentors know this. They're not there to tell you whether your idea is good—they're there to help you figure out what you don't yet know, and what evidence you'll need to move forward. But that test only works if you arrive knowing what you're actually trying to learn.

Most founders don't. They come with a polished story: the problem, the market size, the solution, the team. The pitch feels complete. Then the mentor asks a simple question—How do you know?—and the conversation reveals that half of what you've just said rests on assumption, not evidence.

That's not a failure. That's the point of the conversation. But you'll get more out of it—and your mentor will be able to help you more—if you come in clear about which parts are solid and which parts are still open questions.

What you're testing, and why it matters

Before you sit down, ask yourself: What would have to be true for this idea to be worth building?

Not "Is this a good idea?" That question is too vague. Instead, break it down into the assumptions underneath it.

A few questions will surface them:

• On the problem: Do people with this problem actually want to solve it? How much does it cost them today, in time or money? Are they actively searching for a solution, or would they have to be convinced one exists?

• On the market: How many people have this problem? In your geography? Is the market growing, shrinking or staying flat? If you solve it, who do you reach first?

• On the solution: Why would your approach work better than what they do today? What would it cost to build and support? What could go wrong—and what would you do then?

• On traction: Who have you talked to, and what did they say? (Not "They said they'd buy it." Actual conversations. Actual actions, even small ones.)

• On money: How much capital would you need, and for what? Who might fund it? Do you have a sense of the unit economics—how much you'd spend to acquire a customer, and what they'd pay?

• On execution: Who would build this? What's the hardest part—the thing that, if you got it wrong, would kill the whole idea? How would you know if you'd got it right?

Most founders have answers to some of these. Guesses on others. Blanks on the rest. That's normal. The mentor conversation is where you sort out which is which.

Come with your uncertainties, not your confidence

The instinct when you meet with someone experienced is to sound certain. Don't. It works against you.

A mentor can help you test an assumption only if she knows what you're assuming. If you say, "The market is $500 million," and she doesn't know whether you got that from a report, a calculation or a hunch, she can't help you figure out whether that number matters or how to prove it.

Instead, come with the honest version:

• "I've found three reports on the market size. They range from $300 million to $800 million. The difference is whether they count companies under a certain revenue threshold. I don't know yet which definition matters to our business."

• "I've talked to five people in the industry, and four said they'd consider a solution like this. But I haven't shown them pricing or asked them to commit to a pilot. I think that's the next step."

• "We don't have a technical co-founder yet. The lead time for building the core piece is the thing I'm least sure about. I've sketched a timeline, but I'd like your read on whether it's realistic."

This kind of clarity does three things. First, it shows the mentor exactly where to focus. Second, it tells her what kind of help you need—market research? Technical feasibility? Pilot design? Third, it signals that you're serious: you've already thought about these questions and you're not looking for a cheerleader.

A good mentor will respect that. They'll spend the conversation helping you figure out which assumptions to test first, and what would count as evidence.

Separate what you know from what you're betting on

Before the conversation, spend 15 minutes writing down:

What you're confident about: Which of the core assumptions have you already tested? A pilot customer who's committed? A technical proof of concept? Founder experience in the space?

What's plausible but unproven: Where are you making bets? The belief that a particular feature would be valuable? The hunch that you can reach customers through a specific channel? The assumption that your cost structure would work at scale?

What's still completely open: Where is there no data yet? The regulatory environment? The willingness to pay? Who the first partner should be?

This exercise matters because it shapes what you ask the mentor. If you're confident the problem exists but unsure whether your solution beats the alternatives, don't spend 20 minutes rehashing the problem. Ask: If I wanted to pressure-test my solution against the main competitor, what would I need to show?

If you're uncertain about the total addressable market but pretty sure about the first 100 customers, say so. Ask: What would be enough evidence that this market segment exists and is reachable?

Good mentors can smell the difference between a gap in your thinking and a gap in your evidence. They can help you decide which gaps matter most and how to close them efficiently.

What proof looks like before you're ready to pitch

You don't need a finished business case to have a useful mentor conversation. You do need to know what counts as evidence.

For the mentor to help you, get specific about what would move the needle:

• Not "People are interested." Instead: A pilot with two customers for 12 weeks, tracking usage and willingness to renew.

• Not "The market is big." Instead: A specific segment (this geography, this role, this industry) where you can count active users and define addressable revenue.

• Not "We can build this." Instead: A working prototype of the core function, built in-house, that solves the problem for a real user.

• Not "We have team expertise." Instead: Two founders, one with domain experience in X and one with delivery experience in Y, who have worked together before.

These aren't business-case-level requirements. They're clarity checkpoints. When you bring this level of specificity to a mentor, the conversation shifts. Instead of debating whether your idea is good, you're debating what the next test should be, what it will cost, and what you'll do with the answer.

That's the conversation you want. Because at that point, you're not really asking the mentor for permission. You're asking for guidance on the shortest path to proof.

The mentor's role in your validation pipeline

Think of a mentor conversation as a milestone, not a destination. You're checking in on your thinking, getting reality-tested on your assumptions, and deciding what to do next.

Before you meet: Do the hard thinking. Write down what you believe, what you're guessing about, and what you don't know. If you can, get a little evidence first—even a dozen conversations with people in the space, even a rough cost estimate.

During the conversation: Listen for where the mentor pushes back. That's not a sign your idea is bad. It's a sign of where your logic might be weakest or where your evidence is thinnest. Ask follow-up questions. Take notes.

After: Don't just ask "So, should I do this?" Instead, ask "What should I test first, and how will I know it worked?" Then move. Your next test—pilot, conversation, prototype, or data point—should come from that conversation.

The founders who move fastest are not the ones with the best ideas. They're the ones who are most honest about what they don't know, most systematic about testing it, and most willing to let evidence change their direction.

A mentor can accelerate that process, but only if you come ready to be challenged—and clear about what you're testing.

---

If you're working through an idea and would benefit from talking it through with someone who's helped founders do this before, ElkBee offers a free session to talk through validation, funding, or growth. Book one at www.elkbee.com/book.