Forum Ventures Playbook: Idea Validation

That initial spark, the idea that keeps you up at night, is powerful. You have (ideally) lived the problem firsthand and your solution feels obvious. However, one of the most common reasons early-stage startups fail during the zero to one phase is building something nobody urgently needs. And therefore won’t pay for. 

Before you design your MVP or write any code, or speak to any investors for that matter, you need to validate one thing clearly: Is this a real, painful, and expensive problem for the right customer?

That work starts with customer discovery.

Customer discovery is a system, not a single conversation

Effective customer discovery is not casual chatting or a one-off interview. It is an intentional, repeatable process designed to validate your value hypothesis and identify which customers you should actually build for.

A strong discovery loop follows five steps:

1. Research the prospect and their company
2. Build a discovery call plan
3. Execute the call using a structured framework
4. Recap and follow up
5. Learn and iterate

Discovery is not a single event. It often takes multiple conversations with the same company to fully understand the problem, the impact, and the buying dynamics

We’ve also developed a guide with 15 key things you need to keep in mind for customer discovery calls. 

How many customer conversations do you actually need?

There is no perfect number, but there is a minimum for signal.

As a baseline:

• Aim for at least 30 companies
• One key contact per company for SMBs
• Multiple stakeholders for mid-market and enterprise accounts

You are not looking for opinions. You are looking for patterns that repeat without prompting. And multiple calls across multiple companies and teams will enable you to see those patterns.

The SPIN framework for customer discovery interviews

The most effective discovery interviews follow a natural progression known as SPIN (Situation, Problem, Implication, Needs-Payoff). This framework helps move conversations from surface-level context to real urgency.

Situation

Understand the customer’s current environment and workflow.

• What tools are you using today to manage this?
• How does this process work end to end?

Problem

Surface friction, inefficiency, and unmet needs.

• What feels frustrating or manual about this today?
• What have you tried in the past to fix it?

Implication

Quantify the cost of the problem.

• How much time or money does this cost your team each month?
• What happens if this does not get solved?

Need-Payoff

Let the customer articulate the value themselves.

• If this problem disappeared, what would change for you?
• When does this become critical to solve?

This progression moves conversations from polite interest to real signal.

Why customers buy: Pain, fear, and gain

Customers take action for three reasons:

• Pain: Something is actively costing them time, money, or credibility
• Fear: They see a risk ahead and want to avoid it
• Gain: They can clearly visualize a better future

Your job in discovery is to understand which of these is driving urgency and how strong that motivation really is.

How to get in front of the right customers for discovery

Customer discovery fails quickly if you talk to the wrong people or cannot get conversations at all. Before outreach, founders should be clear on three inputs:

• Ideal Customer Profile
• Buyer or user persona
• Value hypothesis

From there, discovery outreach becomes a process rather than guesswork.

How to get in front of the right customers for discovery

Customer discovery breaks down when you talk to the wrong people or cannot articulate why you want to speak with them.

Before you reach out to anyone, founders should clearly define three things: their Ideal Customer Profile, the buyer or user persona, and their value hypothesis. Without this clarity, outreach becomes noisy and interviews produce weak signal.

Ideal Customer Profile (ICP)

Your Ideal Customer Profile describes the type of company that is most likely to experience the problem you are solving and pay for a solution.

Example:

A B2B SaaS company with 50 to 200 employees, $5M to $20M in annual recurring revenue, based in North America, using HubSpot or Salesforce, and actively scaling its go-to-market team.

Defining your ICP helps you avoid talking to companies that are too early, too large, or structurally misaligned.

Buyer or User Persona

The buyer or user persona is the specific person inside your ICP company who feels the pain most acutely and can describe it in detail.

Example:

A Director of Revenue Operations who owns CRM data, reporting, and forecasting. They spend hours manually pulling data, struggle with trust in their numbers, and feel pressure to deliver accurate insights to leadership.

Discovery interviews should focus on people who live with the problem, not observers who are insulated from it.

Value Hypothesis

Your value hypothesis is a testable belief about how you create value for your customer. This is what customer discovery is designed to validate or disprove.

Example:

We believe that Revenue Operations leaders at mid-stage B2B SaaS companies spend 8 to 12 hours per week manually preparing reports. If we automate reporting and surface clean insights automatically, they will save time, trust their data more, and be willing to pay for a solution.

A strong value hypothesis gives your discovery conversations direction without turning them into a pitch.

A simple prospecting flow for early discovery

Once you are clear on who you want to talk to and why, you can move into a repeatable prospecting process.

1. Generate a target account list using firmographic, technographic, or behavior data
2. Build a contact list within those accounts
3. Choose one or two outreach channels to test
4. Write customer development messaging, not sales copy
5. Run a short outreach cadence
6. Learn and iterate

The goal at this stage is to efficiently generate high-quality signal before optimizing for scale.

Example discovery outreach script

Discovery outreach should feel human, lightweight, and learning-oriented. And NOT like a sales email. 

Here is a simple example founders can adapt:

Subject: Quick question on how you handle [problem area]

Hi [Name],

Thanks for connecting. I am working on a product focused on improving how teams handle [problem area], and I am speaking with a small number of people who deal with this firsthand.

I am not selling anything. I am trying to understand what works today and what does not so we do not build the wrong thing.

Would you be open to a 15-minute call sometime in the next two weeks? Your perspective would be incredibly helpful.

Happy to work around your schedule.

Thanks,
[Your name]

This type of messaging consistently outperforms sales-led outreach early on because it lowers defenses and invites honesty.

Customer discovery questions by buying stage

Not all questions serve the same purpose. Strong discovery flows from awareness to consideration to decision.

Awareness: Understanding the problem

• What tasks within this area feel most frustrating or time-consuming?
• Can you walk me through the last time this issue came up?
• How does this affect your day-to-day work?

Consideration: Understanding current solutions

• How do you currently solve this today?
• What do you like and dislike about the tools you use?
• If you could design the perfect solution, what would it look like?

Decision: Understanding urgency and impact

• What would make you consider switching from your current solution?
• How much time or money would be saved if this were fully solved?
• If this is not addressed this year, what does that mean for the business?

Patterns here reveal whether a problem is a nice-to-have or mission-critical.

Common customer discovery mistakes to avoid

Founders often sabotage discovery unintentionally.

Common pitfalls include:

• Pitching the product too early
• Asking hypothetical or leading questions
• Talking to people without direct exposure to the problem
• Stopping after only a few conversations

Strong discovery requires volume, structure, and repetition to surface real insight.

Using AI to increase discovery velocity

At Forum’s AI Venture Studio (we just announced a $20M raise!), we’re extremely bullish on leveraging AI through the zero-to-one phase, and we’ve been able to speed up the process from a couple of months to less than a week.

However, AI does not replace customer discovery. It helps founders learn faster with fewer resources.

Used well, AI increases discovery velocity by compressing the time between hypothesis, conversation, and insight. The goal is not automation for its own sake. The goal is to reach clarity sooner and avoid building the wrong thing.

Below are a few practical ways founders use AI during early discovery.

Simulating output before building infrastructure

Before building a full product, founders can use AI to simulate the output a customer would receive.

For example, instead of building a reporting engine, a founder can manually run sample data through an AI workflow and present the resulting insight as a weekly report. In discovery calls, the question becomes: if you received this every week, how would it change your job?

This allows founders to validate value before committing engineering time.

Drafting and stress-testing discovery questions

Founders often start discovery with weak or leading questions.

Ai can help draft multiple versions of discovery questions, rephrase them to remove bias, and test whether a question pushes the conversation toward a desired answer. This improves the quality of signal without increasing call volume.

The result is better conversations, not more conversations.

Summarizing conversations and surfacing patterns

After several interviews, founders often struggle to synthesize what they heard.

Ai can be used to summarize call notes, cluster recurring themes, and highlight repeated language across conversations. This makes it easier to spot patterns, refine value hypotheses, and decide what to test next.

The value here is not automation. It is faster insight extraction.

Testing lightweight workflows manually

In some cases, founders can use AI to deliver the core value manually before building software.

If a product promises better insights, predictions, or recommendations, founders can generate those outputs manually using AI and deliver them via email or shared documents. If customers engage, ask follow-up questions, or request continued access, that is strong validation before writing code.

Ai is most powerful in discovery when it helps founders learn faster, not when it distracts them from talking to customers.

When are you done with customer discovery?

You are ready to move forward when:

• The same problems come up unprompted
• Customers use similar language to describe pain
• They ask about pilots, pricing, or timelines
• You can clearly identify who should and should not be a customer

At this point, discovery naturally evolves into the next phase of company building: founder-led sales and design partnerships. At the same time, you’ll also need to validate your idea for VC-Investment. Here’s a full guide on how to do that.

Final thought

Customer discovery is not a pre-seed checkbox. It is a continuous loop of listening, testing, and refining.

Founders on the zero to one journey who invest early in structured discovery build better products, raise stronger rounds, and waste far less time building in the dark.

About Forum Ventures

Forum Ventures® is the leading early-stage fund, accelerator and venture studio for B2B SaaS startups in New York, San Francisco, and Toronto. We support founders through the zero to one phase by becoming their fractional cofounder and building heads down together. To date, we have made 500+ pre-seed and seed investments globally. Those founders have gone on to raise $1B+ in follow-on funding from funds like Kleiner Perkins, Andreesen Horowitz, General Catalyst, NEA, Emergence Capital, Menlo Ventures, Craft Ventures, Founders Fund and more.

‍

Frequently Asked Questions: B2B Idea Validation

What is the primary purpose of customer discovery for B2B startups?

Customer discovery is a systematic process used by founders in the zero to one phase to validate a value hypothesis before building a product. Its purpose is to confirm that a specific group of customers faces a real, painful, and expensive problem that they are willing to pay to solve, thereby reducing the risk of building something nobody wants.

How many customer interviews are needed to validate a new business idea?

For B2B validation, a baseline of 30 different companies is generally required to achieve a "strong signal." In the SMB (Small/Medium Business) sector, one interview per company is often enough, but for Mid-Market or Enterprise solutions, you should interview 2–3 different stakeholders within the same company to understand the internal buying dynamics.

How does the SPIN framework help in founder-led discovery?

The SPIN framework (Situation, Problem, Implication, Need-payoff) is a structured questioning sequence that guides an interview from surface-level facts to high-stakes pain points. It helps founders uncover the "Implication" (the actual cost of the problem in time or money) and the "Need-payoff" (what the customer gains if the problem is solved), which are the primary drivers for a B2B purchase.

Why is pitching a product during a discovery call considered a mistake?

Pitching during discovery often results in "false positives" because prospects may give polite, non-committal praise to avoid conflict. To get honest data, founders should focus entirely on the customer's current workflow and pain points. Validation occurs when a prospect describes the problem you intend to solve as a top priority without being prompted.

What are the best ways to find prospects for discovery interviews?

The most effective way to secure interviews is through targeted, non-sales outreach. This involves identifying an Ideal Customer Profile (ICP), finding the specific persona who experiences the pain, and sending a lightweight message (via LinkedIn or email) explaining that you are "researching the space" and "not selling anything." This approach lowers defenses and positions the founder as a learner.

How can founders use AI to speed up the idea validation process?

AI can be used as a "force multiplier" in three ways:

• Concierge MVPs: Using AI to manually generate the "output" of your proposed software to see if customers find the data valuable.
• Question Bias Testing: Running interview scripts through AI to identify and remove leading or hypothetical questions.
• Pattern Recognition: Using AI to analyze transcripts from dozens of calls to find recurring themes and specific language used by prospects to describe their pain.

What are the signs that a B2B idea has been successfully validated?

Validation is reached when you achieve "pattern resonance." This is when you can predict what the prospect will say next, they use identical language to previous interviewees to describe their frustration, and—most importantly—they begin asking you for a timeline, a pilot program, or a price.