MCP Workflow Guide for Founders & GTM Teams

Research Without a Research Team

Most startups and GTM teams do not have dedicated researchers. Customer understanding falls to founders, product leads, and marketing managers who are already stretched thin.

Koji MCP changes the equation. Instead of scheduling, conducting, transcribing, and analyzing interviews yourself, you tell Claude what you need to learn and it handles the research infrastructure.

---

Use Case 1: Market Validation

Before investing months into building something, validate the problem:

"Create a customer discovery study about whether small business owners struggle with managing multiple social media accounts. Use the Mom Test methodology so we avoid leading questions."

Claude creates a study with:

• Problem statement focused on the pain, not your solution
• Questions that explore current behavior (not hypotheticals)
• Guardrails preventing you from pitching during research

Adding Quantitative Validation

Go beyond qualitative themes by adding structured questions:

"Add a scale question asking how painful this problem is (1-10, with 1 labeled 'Not a problem' and 10 labeled 'Extremely painful'), and a single-choice question asking how they currently manage social media (options: one platform at a time, scheduling tool, virtual assistant, agency, do not manage it)."

After 10-15 interviews, you get both qualitative depth and quantitative data:

"Analyze the interviews. What is the average pain score? What percentage use scheduling tools versus doing it manually? Do people with higher pain scores use different workarounds?"

Claude synthesizes the findings with per-question aggregations — averages, distributions, and frequency counts — alongside qualitative themes.

---

Use Case 2: Lead Qualification Through Research

The Lead Magnet methodology turns research into a sales channel:

"Create a lead magnet study about how marketing teams measure content ROI. Configure a lead form asking for name, email, company, and marketing team size. Add a scale question asking satisfaction with current ROI measurement (1-5) and a multiple-choice question about which metrics they track."

Use koji_configure_study to set up the lead form and branding in one call:

"Set the headline to 'How Do Marketing Teams Measure Content ROI?' with accent color #2563EB, enable the lead form, and set the URL slug to 'content-roi-research'."

After collecting interviews:

"Show me respondents from companies with 50+ marketing team members. What are their biggest content ROI challenges? What was their average satisfaction score with current measurement?"

You now have qualified leads with rich context and quantitative benchmarks for your sales team.

---

Use Case 3: Competitive Intelligence

Understand why customers choose (or leave) competitors:

"Create a JTBD study about what triggers companies to switch from [Competitor] to a new solution. Target the switching moment — what was happening when they decided to look for alternatives? Add a ranking question asking them to order these switching factors: price, features, support, reliability, integrations."

JTBD is perfect for competitive research because it focuses on the trigger events and decision criteria that drive switching behavior. The ranking question gives you quantitative data on which factors matter most.

---

Use Case 4: Positioning Validation

Test your positioning with real prospects:

"Create an exploratory study to understand how developers describe their ideal CI/CD workflow. I want to hear their language, not ours. Add a scale question asking how satisfied they are with their current CI/CD setup (1-7, with anchor probing so we understand what would improve it)."

The AI interviewer captures the exact words your audience uses to describe their problems. The anchor probing on the satisfaction scale reveals what improvements they actually want. Export the transcripts and use their language in your copy:

"Export all transcripts. Pull out the exact phrases respondents used when describing their frustrations. Also show me the satisfaction score distribution."

---

The Founder's Weekly Playbook

Week 1-2: Problem Validation

1. Create a discovery or Mom Test study with structured questions (pain scale + current solution choice)
2. Share the link in communities, LinkedIn, email outreach
3. Target 15 interviews

Week 3: Analysis

4. Review themes, sentiment, and per-question metrics via Claude
5. Generate a research report with charts showing pain distributions and solution breakdowns
6. Decide: build, pivot, or dig deeper?

Week 4+: Ongoing Discovery

7. Keep the study active for continuous feedback
8. Import new contacts as you meet prospects
9. Update reports monthly with latest quantitative trends

---

Tips for Non-Researchers

Do Not Lead the Witness

The worst mistake founders make in research: describing their solution and asking "would you use this?"

Koji's methodology frameworks prevent this automatically. The Mom Test methodology ensures the AI asks about past behavior, not hypothetical future use.

Five Interviews Is Enough to Start

You do not need statistical significance for early-stage validation. Five solid interviews often reveal clear patterns about whether a problem exists and how people currently solve it. Structured questions like pain scales give you quantitative signal even with small samples.

Use Voice Interviews for Deeper Insights

Founders often get richer data from voice interviews because respondents share more freely in spoken conversation. Enable voice as the default mode:

"Configure my study to enable voice interviews as the default mode"

Use Structured Questions to Make Data Actionable

As a founder, you need data that drives decisions. Structured questions bridge the gap between qualitative research and actionable metrics:

• Pain scale (1-10): Is this problem painful enough to pay for a solution?
• Willingness to pay scale: How much would they pay for a solution?
• Feature ranking: Which capabilities matter most to your target market?
• Current solution choice: What is your competitive landscape?

Each of these captures a number you can track over time while the AI interviewer still captures the qualitative reasoning behind each answer.

---

Next Steps

• Quick Start Guide — Get your first interview running
• MCP Setup — Connect Koji to Claude
• Structured Questions Guide — Configure question types and probing
• MCP Overview — Full integration overview
• PM Workflow Guide — For product-focused teams

Further reading on the blog

• B2B Customer Research: The Complete Guide for Product Teams (2026) — B2B customer research is harder than B2C — you are navigating buying groups of 10+ stakeholders, gatekeepers, and enterprise procurement cyc
• Concept Testing: The Complete Guide for Product Teams (2026) — Concept testing validates whether your idea is worth building before you build it. This guide covers methods, question templates, analysis a
• Customer Research Done Right: A Complete Guide for Product Teams — Customer research is the foundation of every successful product decision. Learn the types, methods, and best practices that help product tea

<!-- further-reading:blog -->