AI summary by TwinMind

What to Say to Get Your Way | Jonah Berger | Talks at Google

Talks at Google

47m 48s

Introduction and Background

  • Dr. Jonah Berger, Wharton marketing professor, returns to Talks at Google 10 years after "Contagious"
  • Presenting new book "Magic Words: What to Say to Get Your Way" released earlier this month
  • Montgomery Blair High School class of '98 alumnus, previously visited Google after first book
  • Focus on tactical language use for daily professional/personal impact

Core Premise - Language Power

  • Everyone has something they want to change (customers, prospects, bosses, colleagues)
  • Subtle language shifts create big effects on impact
  • Adding 2 letters to requests increases compliance ~33%
  • "Recommend" vs "like" increases suggestion acceptance ~33%
  • Language in emails predicts promotion/firing likelihood
  • Loan application language predicts default probability

SPEAK Framework Overview

  • S: Similarity and difference
  • P: Posing questions
  • E: Emotion
  • A: Agency and identity
  • C: Confidence
  • C: Concreteness
  • Six key types of magic words for increased impact

Language of Agency and Identity

Actions vs Identities Research

  • Stanford study: "Help clean up" vs "Be a helper" - 30% increase in compliance
  • Voting study: "Go vote" vs "Be a voter" - 15% increase in turnout
  • Actions are temporary, identities are stable self-concepts
  • People engage in actions to confirm desired identities

Negative Behaviors

  • "Losing" vs "being a loser" - identity framing reduces negative behavior
  • "Cheating" vs "being a cheater" - identity prevents undesired actions
  • "Don't be a litterbug" campaign effectiveness

Describing People

  • "Runs" vs "is a runner" - identity suggests consistency
  • "Hard-working" vs "hard worker" - identity implies stable trait
  • "Creator" vs "creates content" - identity sounds like full-time role
  • Leader identity vs leadership actions

Problem-Solving Language

  • "What should I do?" vs "What could I do?"
  • "Should" focuses on single right answer
  • "Could" encourages multiple options, increases creativity
  • Wider viewpoint leads to better solutions

Language of Confidence

Certainty vs Uncertainty

  • Trump speech analysis: uses definites, absolutes, guarantees
  • "Will build" not "might build" - speaks like outcomes are certain
  • Financial advisor study: people prefer certain advisors even when overconfident
  • Certainty makes speakers seem more credible

Common Hedging Mistakes

  • Most people hedge: "might work," "seems like," "I think"
  • Hedging undermines impact, makes speakers seem uncertain
  • Hedges used as linguistic crutches, not intentional uncertainty communication

Confidence Techniques

  • Ditch unnecessary hedges
  • When expressing uncertainty, name it specifically: "Great idea but these 3 things need to happen"
  • Take ownership: "It seems to me" vs "It seems"
  • Use first-person pronouns (I, me) with uncertainty
  • Avoid filler words (um, uh, er, like)
  • Replace fillers with strategic pauses
  • Record yourself to identify filler word patterns

Context-Dependent Confidence

  • Certainty is a tool - know when to use it
  • Leader/salesperson presentations: high confidence appropriate
  • Small team meetings/vulnerability: hedging can show openness
  • Senior leaders: more certainty expected
  • Junior team members: overconfidence can be detrimental
  • Hedging makes you seem open to others' opinions

Language of Posing Questions

Asking for Advice Benefits

  • Harvard/Wharton research: asking for advice increases perceived intelligence/competence
  • People think they give great advice (egocentric bias)
  • When asked for advice, people think asker is smart for choosing them
  • Applies to leaders and team members equally
  • Collects valuable information while improving perception

Other Question Applications

  • Better follow-up questions
  • Using questions as deflection
  • Shaping conversational flow
  • Questions as spotlights focusing attention

Additional Framework Elements (Brief Overview)

Language of Similarity and Difference

  • How language similarity to others affects effectiveness
  • Movie script analysis shows engagement patterns
  • Applications for presentations and content creation

Language of Emotion

  • Better storytelling techniques
  • More effective self-pitching
  • Audience attention holding strategies
  • Analysis of tens of thousands of online content pieces

Language of Concreteness

  • Showing others you're listening
  • Making ideas more memorable
  • Entrepreneur fundraising success
  • Leadership perception improvement

Research Methodology and Tools

Data Sources

  • Tens of thousands of online content pieces
  • Thousands of conversations (people and companies)
  • Customer satisfaction and relationship building studies
  • Email language analysis
  • Loan application language parsing

Analysis Tools

  • Natural language processing evolution over 10+ years
  • Basic tools: dictionaries, Linguistic Inquiry and Word Count (LIWC)
  • Advanced tools: topic modeling, word embeddings
  • Automated text analysis for behavioral insights
  • Collaboration with computer scientists at Cornell Tech

Practical Applications

Professional Context

  • Email writing improvement
  • PowerPoint presentation enhancement
  • Meeting communication
  • Customer/prospect interactions
  • Colleague relationship building

Personal Context

  • Family communication
  • Social connection deepening
  • Creative problem solving
  • Goal motivation techniques
  • Self-talk for anxiety reduction and performance

Cultural and Contextual Considerations

Cross-Cultural Applicability

  • Core language types translate across cultures
  • Agency/identity words universal when translated
  • Confidence expression varies by cultural context
  • US culture values independence and certainty
  • East Asian cultures emphasize group harmony
  • Deployment strategies must consider cultural norms

Contextual Deployment

  • Subcultures within organizations
  • Status-dependent communication strategies
  • Situation-appropriate language choices
  • Understanding when to hedge vs. when to be certain

Future Implications and AI

Large Language Models Impact

  • LLMs excel at producing realistic language
  • Current focus on "reasonable" vs "persuasive" answers
  • May eventually be tuned for persuasion effectiveness
  • Understanding human behavior remains crucial
  • Behavioral science knowledge still valuable for optimization

Brand Communication Research

  • Social media post engagement analysis
  • Customer support page effectiveness
  • Word "you" - grabs attention but can assign blame
  • Context determines language effectiveness
  • Attention-grabbing vs. helpful language differences

Broader Research Insights

Societal Applications

  • Quarter million songs analyzed for misogyny over 60 years
  • Music bias decreased over time, mainly due to male artist changes
  • Police language bias research (Stanford)
  • Different language patterns with different community members
  • Movie success prediction based on language
  • Song hit prediction using linguistic analysis

Ethical Considerations

  • Persuasion vs manipulation distinction
  • Intent and outcome determine ethical nature
  • Environmental influence occurs constantly (intentional/unintentional)
  • Understanding influence helps make informed choices
  • Behavioral science should be used ethically
  • Knowledge empowers better decision-making

Key Takeaways

Understanding Language Power

  • Words used constantly but not optimally
  • Subtle shifts create significant impact
  • Everyone is a speaker and writer in daily life
  • Communication skills can be learned and improved

Practical Implementation

  • Six types of magic words framework
  • Context-appropriate deployment strategies
  • Recording and analyzing own speech patterns
  • Free resources available at jonahberger.com
  • Guides for impactful content and persuasion techniques

TwinMind

Get summaries like this for your own meetings

TwinMind takes notes in the background during meetings, lectures, and calls, then writes the summary for you.

Free plan to get started

No bot joins your calls

Phone, desktop & Chrome apps

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome