Session 5 · Week 3 · WHAT Layer
←→ navigate   N notes
⭐ Vanderbilt Owen MBA · Spring 2026
Session 5: Deep Research
The Intelligence Layer
Right Product · Right Story · Right Time · Right People
Right Voices · Right Message
Monday, March 23, 2026
OP
Oliver Park
Instructor · ✦ Inkwell Labs
🔍 Deep Research 📊 Market Intelligence 🎯 7 Rights Framework 🚀 Build Better
Last Week

Jeff Jonas showed us how to
BUILD the graph

"Jeff Jonas — entity resolution pioneer, Senzing founder. He taught us the engineering foundations of knowledge graphs at scale."
🔗 Entity Resolution
Connecting disparate data points that refer to the same real-world entity. Oliver Park = O. Park = Oliver L. = same person.
📈 Context Accumulation
The graph gets smarter over time. Each new data point adds context, improves entity matching, reveals new patterns.
🧠 Sensemaking
Connecting disparate data into intelligence. The graph reveals relationships humans would never see manually.
🎯
Jeff's Gift to Us
The knowledge graph infrastructure is solved. We can now build graphs that scale, resolve entities, and accumulate context automatically.
Now: How do we USE it?
Building graphs is engineering.
Using them for business decisions is strategy.
🎯 The Core Framework
The 7 Rights Framework
"The right product, with the right story, at the right time,
to the right people, from the right voices, with the right message."
🎯
Right Product
What does the market actually need?
📖
Right Story
What narrative resonates?
⏰
Right Time
When is the market ready?
👥
Right People
Who is the ideal customer?
🎙️
Right Voices
Who do they trust?
💬
Right Message
What language converts?
Deep research answers ALL seven. AI + graphs make it possible at scale.
🤔 Definition
What IS Deep Research?
"Not Googling. Not ChatGPT. Not a survey."
1. Maps the full landscape
Market, competitors, consumers, channels — the entire ecosystem, not just your slice.
2. Identifies whitespace
Gaps no one is filling. What exists vs. what's needed. Gap analysis via graph topology.
3. Finds patterns humans miss
Cross-domain connections. AI can process millions of data points to surface non-obvious correlations.
4. Predicts trajectories
Where the market is going, not where it is. Trend analysis, early signal detection.
🕸️
5. Surfaces the "Hidden Graph"
The non-obvious insights live in the connections between disparate domains. AI traverses relationship networks that humans would never map manually.
🧪 Tool Introduction
Google NotebookLM
"Your AI research assistant — grounded in YOUR sources, not the open web"
🔗 Open NotebookLM →
📚
Upload Any Source
PDFs, Google Docs, websites, YouTube, audio files. Up to 50 sources per notebook.
🔒
Zero Hallucination
Answers cite ONLY your uploaded sources. Every claim is traceable. No random web content.
🎙️
Audio Overview
Generates a podcast-style discussion of your research. Two AI hosts debate and explain your findings.
⚡
One-Click Outputs
Study guides, FAQs, timelines, briefing docs, and outlines — generated instantly from your sources.
💡 For This Course
Upload your course readings, team research, and industry reports into one notebook. Ask it to find contradictions, synthesize themes, and generate briefing docs for your assignments.
🎯 For Deep Research
Upload 10 competitor analyses and ask: "What market gap do none of these companies address?" — NotebookLM synthesizes across ALL sources simultaneously.
🏗️ The Stack
The Deep Research Stack
🌐
Layer 1: Open Source Intelligence (OSINT)
Public data, filings, news, social media
📊
Layer 2: Structured Data
Industry databases, market reports, patent filings
🗣️
Layer 3: Voice of Customer
Reviews, forums, Reddit, social listening
🕸️
Layer 4: Graph Intelligence
Relationship mapping, entity resolution, link prediction
🤖
Layer 5: AI Synthesis
LLMs to process, pattern-match, and generate insights
🔄
Each layer feeds the next
Raw data becomes structured knowledge, which becomes relationship intelligence, which becomes actionable insights.
The Graph CONNECTS them all
Without the knowledge graph, you have five disconnected data sources. With it, you have an intelligence system where each layer enriches the others.
🎯 Right Product
Defining Market Need
"The graveyard of startups is full of solutions looking for problems."
Deep research for product-market fit:
• Demand signals: Search volume trends, Reddit/forum complaints, review sentiment
• Whitespace mapping: What exists vs. what's needed (gap analysis via graph)
• Adjacent innovation: What worked in analogous markets?
Mushroom coffee didn't come from coffee companies or mushroom growers — it came from wellness companies reading health forums
Microgreen powder demand is visible in search trends 18 months before major brands entered
🔧 TOOLS & METHODS
• Google Trends + Exploding Topics
• Patent analysis (what's coming in 2-3 years)
• Amazon/Shopify review mining
• Reddit/forum scraping
• Competitor graph mapping
💡 GRAPH INSIGHT
Your knowledge graph should have a
"Product Landscape" subgraph
📖 Right Story
Narrative Research
"People don't buy products. They buy stories."
Deep research for narrative:
• Origin stories that resonate (founder journey, mission, values)
• Category framing: Are you a "wellness brand" or a "farm"? VERY different stories.
• Cultural moment mapping: What macro trends make your story timely?
Patagonia: "We're in business to save our home planet" (not "we sell jackets")
Radical Shoots: "Integrity & Flavor in Food" (not "we grow microgreens")
💰
The story determines the PRICE CEILING
Not the product. A "superfood" commands 10x the price of "lettuce" even if it's the same plant. The narrative creates the value perception.
🎯 RESEARCH FOCUS
Which narratives are winning in adjacent categories? What cultural moments are creating new story opportunities?
⏰ Right Time
Temporal Intelligence
"Timing isn't luck. It's research."
How to research timing:
• Technology adoption curves: Where is the category on the S-curve?
• Regulatory windows: New laws create new markets
• Cultural moments: Health scares, documentaries, viral content
• Seasonal patterns: When do people search for your category?
📊 EXAMPLE: BCBST Food-as-Medicine
The 2024 BCBST food-as-medicine program created a $100M+ procurement channel that didn't exist 12 months earlier. Predictable if you were tracking regulatory signals.
🕸️ GRAPH APPROACH
Build a "Market Timeline" subgraph connecting:

Events → Consumer Behavior Shifts
Regulations → Market Opportunities
Cultural Moments → Demand Spikes
⚡ TIMING SIGNALS
Patent filings (2-3 year lead time)
Congressional hearings (12-18 months)
Documentary releases (immediate spike)
Search trend shifts (leading indicator)
👥 Right People
Customer Intelligence
"Your ideal customer is hiding in the data."
Deep research for customer definition:
• Psychographics over demographics: WHY they buy, not just WHO they are
• Jobs-to-be-done framework: What job is the customer hiring your product to do?
• Behavioral clustering: Purchase patterns, content consumption, community membership
• Lookalike modeling: Find more people like your best customers
🕸️ GRAPH APPROACH
Customer Entity → has_behavior → buys_organic
Customer Entity → visits → farmers_markets
Customer Entity → follows → @spinachtiger
📊 NASHVILLE EXAMPLE
• Median age 34.5, median HHI $77,371
• 82% of Americans prioritize wellness
• 75% bought natural/organic in last 6 months
• 90% of Gen Z/Millennial shop natural
• 900+ healthcare companies = health-conscious workforce
💡 THE DEEP INSIGHT
Nashville's farm-to-table restaurant density means CHEFS are the first customer, not consumers. Chefs are the validators.
🎙️ Right Voices
Influence Mapping
"Who do your target customers actually listen to?"
Deep research for influence:
• Not just follower counts — TRUST mapping
• Micro-influencers (5K-50K) outperform mega-influencers on conversion
• Chef validation > celebrity endorsement for food products
• Press/media hierarchy: Which publications does YOUR customer read?
🕸️ GRAPH APPROACH
Build an influence graph:
Customer_Segment → trusts → Voice → reaches → Audience
🎯 TRUST NETWORKS
Center: Inner circle (family, friends, doctor)
Ring 2: Trusted experts (chefs, nutritionists, trainers)
Ring 3: Community voices (local bloggers, micro-influencers)
Ring 4: Media (local press, niche publications)
Ring 5: Broad reach (social media, advertising)
⚠️ THE MISTAKE
Most brands start at Ring 5 and wonder why nobody converts. Start at Ring 2.
💬 Right Message
Language Intelligence
"The words your customer uses are not the words you use."
Deep research for messaging:
• Review mining: Extract EXACT language customers use to describe the problem
• Search query analysis: How do people SEARCH for what you sell?
• A/B testing at scale: Test headlines, claims, CTAs with AI
• Regulatory language: What can you legally say? (Structure/function vs. health claims)
Customers say "brain fog" not "cognitive function"
Customers say "gut health" not "gastrointestinal wellness"
🗣️
Use THEIR words, not YOUR words
Professional terminology creates distance. Customer language creates connection. Mine reviews, forums, and support tickets for authentic vocabulary.
📊 RESEARCH TOOLS
Google Keyword Planner
Answer The Public
Reddit/Forum scraping
Customer support transcripts
Review sentiment analysis
🛤️ Right Path
Path of Least Resistance
"The shortest distance between your product and your customer."
Channel intelligence via deep research:
• Where does your customer already shop? (Don't make them go somewhere new)
• What's the purchase trigger? (Sampling? Recommendation? Content?)
• Friction analysis: Map every step from awareness → purchase. Where do people drop off?
🕸️ GRAPH APPROACH
Customer → discovers_via → Channel
Customer → purchases_at → Location
🚀 CHANNEL SEQUENCING
Phase 1 (0-6mo): Farmers markets + Chef B2B
Phase 2 (6-12mo): Local retail + DTC subscription
Phase 3 (12-18mo): Regional grocery + National DTC
Phase 4 (18-36mo): Institutional + Marketplace
💡 INSIGHT
The path of least resistance is almost never your first instinct. Each phase FUNDS and VALIDATES the next.
📋 Framework
The Deep Research Canvas
A single-page framework combining all 7 Rights
Right
Question
Research Method
Graph Entity
🎯 Product
What does the market actually need?
Demand signals, whitespace mapping
Product_Gap → needs → Market_Segment
📖 Story
What narrative resonates?
Cultural moment mapping
Brand → tells_story → Cultural_Moment
⏰ Time
When is the market ready?
Regulatory signals, S-curve analysis
Event → creates → Market_Window
👥 People
Who is the ideal customer?
Behavioral clustering, psychographics
Customer → has_behavior → Buying_Pattern
🎙️ Voices
Who do they trust?
Influence mapping, trust networks
Customer_Segment → trusts → Influencer
💬 Message
What language converts?
Review mining, search analysis
Customer → uses_language → Vocabulary
🛤️ Path
What's the path of least resistance?
Channel mapping, friction analysis
Customer → discovers_via → Channel
This is your workshop deliverable. Use it for every venture, every campaign, every strategic decision.
🤖 Live Tool
VanderBot: Your Deep Research Copilot
"Let's do deep research together — live, right now."
🤖 What VanderBot Does
🔍 Deep Research — Searches, synthesizes, and maps market intelligence across the web
🕸️ Graph Building — Generates entities, relationships, and triples from research
📊 7 Rights Analysis — Applies our framework to any market or product
📝 Document Generation — Creates briefs, analyses, and presentations
⚡ Live Exercise
Step 1: Open VanderBot
vanderbot.com
Step 2: Prompt it with your team's market:
"Do a deep research analysis of [your market] using the 7 Rights framework"
Step 3: Review the output together — what surprised you?
Step 4: Ask a follow-up: "What's the non-obvious insight?"
💡 Pro Tip: VanderBot gets better with specificity. "Nashville functional food market" > "food market." Add constraints: budget, timeline, geography.
🧪 Hands-On
NotebookLM Exercise
"Build your team's research notebook — right now"
📋 Steps
1. Go to notebooklm.google.com
2. Create a new notebook for your team's project
3. Upload 3-5 sources:
Course readings, your team's research, industry reports, competitor websites (paste URLs)
4. Ask it:
"Based on these sources, apply the 7 Rights framework to identify our market opportunity"
5. Generate an Audio Overview
Listen to the AI discuss your research
💡 Power Prompts to Try
"What contradictions exist between these sources?"
"What's the strongest evidence for product-market fit?"
"Generate a timeline of key market events from these sources"
"Who are the key entities mentioned across all sources?"
🔗 The Workflow
VanderBot discovers → NotebookLM synthesizes → Your Graph captures → Your Canvas organizes
🛠️ Workshop
Build Your Deep Research Canvas
"Apply the 7 Rights to YOUR team's project"
📋 INSTRUCTIONS
1. Choose your product/venture
(from your graph project)
2. For each of the 7 Rights, answer the key question
Use the canvas framework
3. Identify the TOP research method
What's your approach for each Right?
4. Map at least 3 entities per Right
In your knowledge graph
5. Find ONE non-obvious insight
From connecting Rights across your graph
⏰
Teams of 4
Use the Deep Research Canvas template. Focus on finding connections between the 7 dimensions.
💡 THE SECRET
"The best insights come from the CONNECTIONS between Rights, not the Rights themselves."
🎯 DELIVERABLE
Completed Canvas + 1 Cross-Connection Insight
What did you learn from connecting two Rights?
💡 Key Takeaways
Session 5 Insights
🎯
Deep research is systematic intelligence
Not casual Googling. AI-augmented, graph-enhanced intelligence gathering with specific methodologies.
📊
The 7 Rights framework ensures no blind spots
Systematic coverage of product, story, time, people, voices, message, and path dimensions.
🗣️
Use your CUSTOMER'S language, not yours
Review mining and search analysis reveal authentic vocabulary that creates connection.
🛤️
Path of least resistance beats brute force marketing
Smart channel sequencing where each phase funds and validates the next.
🔗
The knowledge graph CONNECTS all seven dimensions
Entities and relationships across all 7 Rights create comprehensive market intelligence.
💡
The non-obvious insight lives in the CROSS-connections
The best strategic intelligence comes from connecting Rights together, not studying them separately.
Your Strategic Intelligence Framework
Use the 7 Rights for every venture, every campaign, every strategic decision. The graph connects it all.
📝 Assignment + Looking Ahead
Assignment + Looking Ahead
📋 DUE BEFORE SESSION 6
AI Reliability Analysis (2 pages)
How do you assess the reliability of AI systems when the stakes are high?
🎯 EXTRA CREDIT
Add a "7 Rights" subgraph to your team's knowledge graph project
🗓️ COMING UP
Session 6 (Wed): AI Reliability Audit
Session 7: Midterm — March 30
"The companies that research deepest,
win fastest."
— ✦ Inkwell Labs
1 / 19
Speaker Notes