The Secrets of Conversation Masters: A Three-Level Framework for Making People Like You

Vanessa Van Edwards distills 20 years of conversation research into a three-level framework to fight the modern connection crisis.
Behavioral researcher Vanessa Van Edwards argues that asynchronous communication, AI substitution, and scripted social interactions have triggered a modern "connection crisis." Drawing on psychologist Dan McAdams's Narrative Identity Theory, she breaks conversation into three levels — facts-to-stories, personal goals, and self-narrative — and identifies five narrative archetypes to predict behavior. Her practical toolkit covers opening lines, graceful exits, speaking time management, and the key distinction between entertainers and appreciators, all grounded in one Carnegie-style conclusion: be interested, not interesting.
In an era where AI is gradually taking over human communication, conversational ability is becoming a rare form of career insurance. Behavioral researcher Vanessa Van Edwards, in her new book Conversations, makes a core argument: we are in the midst of a "connection crisis" — conversations are becoming harder, more awkward, and less satisfying. She spent 20 years studying conversation masters, breaking down invisible conversational blueprints into learnable skills.
This article is based on Steven Bartlett's in-depth podcast interview with Vanessa Van Edwards on The Diary of a CEO, summarizing the conversational frameworks and practical tools she presents.
Why We Are Losing the Ability to Converse
Vanessa views conversation as "the vehicle to everything you want": if you want a promotion, you need to connect meaningfully with your boss; if you want to find a soulmate, you need deep and meaningful dialogue; if you want to escape loneliness, you need friends who genuinely bring you joy. She cites data showing that one in six people globally suffers from loneliness, and only 30% of Americans engage in social interaction daily.
Even more alarming is a three-part erosion. First, we listen to podcasts as if they were conversations — but they're actually interviews, performative content that doesn't follow the mechanics of real dialogue. Second, asynchronous communication is skyrocketing — Vanessa jokes about sending nine-minute voice messages to friends, calling voice notes "the most selfish form of conversation" because there's no turn-taking mechanism. Third, people are increasingly talking to technology. She shares a telling detail: her husband told an AI "do better, that's terrible work," then turned to ask her something, at which point she warned him, "You better not say 'do better' to me."
She compares conversation to fitness: "We care so much about physical health, but social health is critical to emotional health." Conversation is a skill you can build by "doing reps" — it's not something you're born with.
The Three-Level Conversation Framework: From Facts to Self-Narrative
Vanessa's core framework draws on the research of psychologist Dan McAdams, dividing the connection process into three levels. She ran a "rapid social experiment" with 500 participants to validate it: she gave them six conversation openers and asked them to rate each. The two worst-performing questions were "What do you do?" (ranked last) and "How are you?" — because they're social scripts that put the brain on autopilot and lead to factual dead ends.
The best opener was "What's been the highlight of your day?" because it invites the other person's brain to search for something positive while simultaneously breaking autopilot.

Level 1: General Traits — Turning Facts into Stories
The first level is basic factual exchange, but the key insight is that "facts are stop signs, stories are doors." Instead of asking "How are you?", try "Do you have any passion projects going on?" or "What's something exciting you've been working on lately?" Vanessa emphasizes this requires a bit of "conversational courage," but it opens the door to stories and clues about values.
She borrows a technique from conversation master Howard Stern — creating an "information gap," such as "You won't believe what happened to me yesterday." The brain is hardwired to be uncomfortable with the unknown. In an age where your attention competes with endless content feeds, your questions need to be "dopaminergic" enough to make someone put their phone down.
Level 2: Personal Concerns — Doing "Dream Building"
The second level moves into motivations, goals, values, and worries. Vanessa calls this "dream building": helping others reveal their secret desires and secret side projects. She offers the "pet or pest project" framework — a pet project is something they're proud of and eager to pursue; a pest project carries shame, something they want to fix but haven't dared bring up.
Effective questions include "What's something you've always wanted to learn?" (triggers a growth mindset) and "What's your biggest goal right now?" She also offers practical advice for CEOs and teams: open weekly meetings with "tell me something good" — a three-to-five-minute ritual that quickly creates frictionless team connection, because "communication friction costs companies money."
Level 3: Self-Narrative — Reserved for the People Who Matter Most
The third level is "the story you tell yourself about yourself." Vanessa distinguishes between two types of narrative: redemptive narratives (where you overcome whatever happens to you) and contamination narratives (where you feel like everything bad always happens to you). Research shows that people with redemptive narratives live longer, have stronger immune systems, and are happier, because they have a strong internal locus of control.
She emphasizes that Level 3 must have a "gated entry" — sharing too much too fast leads to "vulnerability regret," which is especially inappropriate in professional settings.

Dan McAdams is a personality psychologist at Northwestern University whose "Narrative Identity Theory" holds that humans form a stable sense of self by constructing internalized stories about their own lives. The theory describes the formation of close relationships as a layered "unveiling" process: people first exchange objective facts, then share subjective goals and values, and finally touch the deep narratives that form the core of the self. Vanessa's three-level framework is a practical translation of this academic model — transforming layers of connection that would naturally take years to develop into a consciously accelerated conversational design. Understanding this psychological foundation helps explain why "What do you do?" is so ineffective: it can only trigger Level 1 factual retrieval and cannot open a channel to deeper connection.
"Locus of control" is a personality dimension concept proposed by psychologist Julian Rotter in 1954, describing a person's tendency to believe in the degree of connection between their own behavior and life outcomes. People with a strong internal locus of control believe outcomes are primarily determined by their own actions; those with a strong external locus tend to attribute outcomes to luck, fate, or others. Research consistently shows that a high internal locus of control correlates with better mental health, greater resilience, and higher job satisfaction. Redemptive narratives are beneficial to health precisely because they structurally reinforce an internal locus of control — no matter what happens, the protagonist of the narrative is always an agent with agency, not a victim of fate. This also explains why Vanessa views "Do you think you're lucky?" as an effective tool for probing someone's locus of control.
"Do You Think You're Lucky?": A Near-Polygraph-Level Question
Vanessa's favorite Level 3 question is "Do you think you're lucky?" The answers come in three varieties: people who immediately say "yes" tend to have high gratitude, high sense of control, and self-confidence; people who say "no" shouldn't be talked out of it (that's toxic positivity) but rather understood; and people who hesitate either have never reflected on it or don't yet feel safe enough with you to answer.
She cites psychologist Richard Wiseman's research: people who consider themselves lucky actually create good fortune through specific behaviors — lower anxiety, higher extroversion, extensive socializing, openness to new experiences, and relentless optimism. Vanessa has even considered opening a "luck lab" workshop to help people "harvest" a sense of luck, which she believes is more powerful than gratitude. "Gratitude has a slight tinge of shame to it," she says.

Five Narrative Archetypes: The Key to Predicting Behavior
Vanessa distilled five archetypes from hundreds of narratives. These aren't fixed personalities but fluid "ways of seeing your own story":
- Ambitious Hero: Overcomes obstacles through sheer willpower; internally never satisfied; prone to "shiny object syndrome" and burnout. Steven identifies as this type — his chosen character is Chris Gardner from The Pursuit of Happyness, a film that gave him hope when he was surviving on stolen food at 18.
- Contented Pragmatist: Loves the status quo, thrives on habits and tradition; often misread as lacking ambition; well-suited to stable, regulated industries.
- Selfless Helper: Finds meaning in giving; prone to compassion fatigue and poor boundaries; often attracts "high-conflict personalities" who take without giving back.
- Creative Disruptor: Hates rules and exists to break them; a natural "dream killer" and the most innovative challenger in the room.
- Unlucky Struggler: The embodiment of the contamination narrative; think George Costanza from Seinfeld — often funny but perpetually tormented.
She offers a clever excavation tool: "Which book, movie, or TV character are you most like?" The answer isn't what matters — the follow-up "why?" is. The traits they name next reveal how they see themselves.

How to Exit Gracefully — and Where Aura Comes From
For the question "How do I exit a bad conversation without offending anyone?" Vanessa offers a "graceful exit" formula: a future hook + a compliment or action. First, ask what the other person has coming up next (shifting their brain into "what's ahead" mode), then use an exit phrase paired with a physical action ("I'm going to grab a coffee refill"), while simultaneously stepping slightly back and angling your body toward the door as a nonverbal signal. She stresses being decisive, because "the lingering awkwardness of hovering is the enemy of charm."
On the subject of aura and "rizz," her core conclusion is: don't try to seem impressive. She shares a personal story — at a VIP party, everyone's self-introductions escalated in grandeur ("I'm a billionaire," "I cured cancer"), until it was her turn and she said, "I'm Vanessa, a recovering awkward person." The whole room burst out laughing and visibly relaxed. Afterward, nearly everyone came over to talk to her.
She summarizes the two elements of charm: clearly signaling "I like you" (people dramatically underestimate the liking gap), and leading with warmth before competence (82% of first impressions are based on warmth and competence). She also introduces the concept of the "maximum resonance point" — speaking at the natural lower end of your vocal range while exhaling, which conveys ease and confidence.
Are You Talking Too Much?
A study of over a thousand Americans found that 95% had worked with someone who talked too much. These individuals spoke for an average of 90 minutes per day, and 71% of respondents felt that overly talkative colleagues affected their ability to get work done. Vanessa offers a diagnostic: time yourself in recorded Zoom meetings — 15 to 45 seconds is an appropriate speaking turn. Also watch for nonverbal cues, like someone opening their mouth to speak and then waiting — the "open fish" signal.
She also distinguishes between two types of people: entertainers, who want to be asked questions, love telling stories and giving advice; and appreciators, who say less and love hearing others' stories. When talking to a VIP or a boss, figuring out which type they are is critical — entertainers want you to ask questions; appreciators want you to tell stories.
She closes by invoking Dale Carnegie's wisdom: be "interested" rather than "interesting." Research shows the most popular children are those with the longest "people I like" lists. If you enter a conversation with the mindset of "how can I like you more," connection will follow naturally. As AI increasingly erodes human communication, this conversational blueprint may be an essential tool in our fight against loneliness.
"Liking gap" is a finding from a 2018 study published in Psychological Science by Harvard researchers including Erica Boothby: after a conversation, strangers actually like each other more than they anticipated, but nearly everyone underestimates how much the other person likes them. This cognitive bias causes people to systematically underestimate their own appeal in social situations, leading them to act overly cautious or deliberately "impressive" — which runs directly counter to Vanessa's advice on charm. The research also found that the gap grows larger the longer the conversation, suggesting that people's self-assessments in social situations are consistently more pessimistic than reality. Simply knowing about this bias is itself a practical social calibration tool.
Related articles

JobRunner MCP Server: AI Agents Can Now Fix Failed Java Jobs
JobRunner launches an open-source MCP server enabling AI agents to view cluster status, diagnose failures, and recover Java jobs—incident response from any cloud coding environment or existing agent.

MLAI Digital Launches 9 Security Agents: MCP Server Safeguards the Full Lifecycle of Agentic AI
MLAI Digital launches Security MCP Server with nine built-in specialized security agents covering vulnerability discovery, attack simulation, prompt injection detection, API and infrastructure assessment, and accelerated remediation—embedding security validation across the full Agentic AI lifecycle from development to production.

Running Isaac Lab on H100 GPUs: Exploring Better Approaches
Exploring optimization methods for running Isaac Lab on NVIDIA H100 GPUs, including MIG configuration, mixed precision, TensorRT acceleration, and community experience to improve robotics simulation and reinforcement learning efficiency.