Discovery Interview

Discovery Interview is a community skill for conducting and analyzing customer discovery interviews, covering interview script creation, question framework design, response coding and analysis, insight extraction, and finding synthesis for product and market validation research.

What Is This?

Overview

Discovery Interview provides frameworks for running structured customer interviews that validate product assumptions. It covers interview script creation that builds conversation guides with open-ended questions and follow-up prompts, question framework design that applies Jobs-to-be-Done and problem-solution fit approaches, response coding that categorizes interview answers into themes and patterns, insight extraction that identifies recurring needs, frustrations, and workarounds across multiple interviews, and finding synthesis that produces evidence-based recommendations for product direction. The skill enables product teams to make decisions grounded in customer evidence rather than assumptions.

Who Should Use This

This skill serves product managers validating feature ideas before development investment, startup founders testing market assumptions with potential customers, and UX researchers conducting generative research to identify unmet needs.

Why Use It?

Problems It Solves

Product decisions based on assumptions rather than customer evidence lead to features nobody uses. Unstructured interviews produce anecdotes that cannot be systematically compared across respondents. Confirmation bias leads interviewers to hear validation for their ideas rather than genuine customer needs. Interview findings are summarized subjectively without a repeatable coding methodology.

Core Highlights

Script builder creates interview guides with open-ended questions that avoid leading the respondent. Framework selector applies JTBD or problem-solution fit templates to match the research objective. Response coder categorizes answers into themes with frequency counts. Insight synthesizer produces evidence-backed recommendations from coded interview data.

How to Use It?

Basic Usage

from dataclasses\
  import dataclass, field

@dataclass
class Question:
  text: str
  follow_ups:\
    list[str] = field(
      default_factory=list)
  category: str = ''

@dataclass
class InterviewScript:
  title: str
  objective: str
  questions:\
    list[Question]\
    = field(
      default_factory=list)

def build_jtbd_script(
  product_area: str
) -> InterviewScript:
  return InterviewScript(
    title=f'{product_area}'
      f' Discovery',
    objective=
      'Understand jobs '
      'customers hire '
      'solutions for',
    questions=[
      Question(
        text='Walk me '
          'through the '
          'last time you '
          'needed to do '
          'this task.',
        follow_ups=[
          'What was the '
          'trigger?',
          'What did you '
          'try first?'],
        category='context'
      ),
      Question(
        text='What is '
          'the hardest '
          'part about '
          'this process?',
        follow_ups=[
          'How often does '
          'that happen?'],
        category=\
          'pain_point')])

Real-World Examples

class ResponseCoder:
  def __init__(self):
    self.themes = {}
    self.responses = []

  def add_response(
    self,
    respondent: str,
    question: str,
    answer: str,
    themes:\
      list[str]
  ):
    self.responses\
      .append({
        'respondent':
          respondent,
        'question':
          question,
        'answer': answer,
        'themes': themes})
    for theme in themes:
      self.themes[theme]\
        = self.themes.get(
          theme, 0) + 1

  def top_themes(
    self, limit: int = 5
  ) -> list[tuple]:
    return sorted(
      self.themes.items(),
      key=lambda x: x[1],
      reverse=True
    )[:limit]

  def evidence_for(
    self, theme: str
  ) -> list[dict]:
    return [
      r for r
      in self.responses
      if theme
        in r['themes']]

Advanced Tips

Ask about past behavior rather than future intentions since people predict their own actions poorly but describe recent experiences accurately. Record the exact words customers use to describe their problems as these phrases are valuable for marketing copy and feature naming. Aim for interview saturation where new interviews stop producing novel themes, typically around twelve to fifteen interviews per segment.

When to Use It?

Use Cases

Validate a product idea by interviewing potential customers about their current workflows and pain points. Code interview responses into themes to identify the most common needs across a customer segment. Synthesize findings from a round of discovery interviews into evidence-based product recommendations.

Related Topics

Customer discovery, user research, Jobs-to-be-Done, qualitative research, product validation, and interview techniques.

Important Notes

Requirements

Access to target customers or potential users willing to participate in interviews. Recording equipment or transcription service for capturing interview content. Coding framework or spreadsheet for systematic response categorization.

Usage Recommendations

Do: use open-ended questions that start with how, what, and tell me about rather than yes or no questions. Separate interview facilitation from note-taking by having a dedicated observer when possible. Code responses independently before comparing themes across interviews.

Don't: pitch your solution during the interview which biases the respondent toward validating your idea. Rely on a single interview to represent an entire customer segment. Skip the coding step and jump to conclusions based on memorable quotes rather than systematic theme analysis.

Limitations

Interview findings represent qualitative evidence from small samples and should be validated with quantitative data before major product decisions. Self-reported behavior in interviews may not match actual behavior observed in usage analytics. Respondent selection bias affects findings when interview participants are not representative of the broader target market.