Podcast Personalization Research

Podcast Personalization Research

Research a prospect's podcast appearances and generate deep personalization angles with specific quotes, topics, and draft openers for outreach.

Instructions

When a user wants to personalize outreach based on a prospect's podcast appearances, search for episodes, extract key insights, and synthesize them into actionable personalization fields and openers.

Steps

1. Identify the prospect. Extract any available identifiers from the user's request:
   • LinkedIn URL
   • Email address
   • Name + company If the user provides minimal info (e.g., first name only), ask for at least a full name and company or a LinkedIn URL before proceeding.
2. Enrich the person by calling mcp__claude_ai_Amplemarket__enrich_person with all available identifiers. Key data points to extract:
   • Full name
   • Current title and role
   • Company name and domain
   • Career history and previous companies
   • Industry and areas of expertise These details are essential for constructing accurate podcast search queries in the next step.
3. Search for podcast appearances using WebSearch with multiple query patterns to maximize coverage. Run all of these queries:
   • "[full name]" podcast
   • "[full name]" episode guest
   • "[full name]" interview [industry topic]
   • "[full name]" "[company name]" podcast
   • "[full name]" webinar OR panel OR fireside Scan the results for podcast directories (Apple Podcasts, Spotify, YouTube), show notes pages, and episode listings. Deduplicate results across queries and rank by relevance and recency.
4. Extract podcast content by calling WebFetch on the top 3-5 most promising results. For each episode page, extract:
   • Podcast name and host
   • Episode title and date
   • Key topics discussed
   • Direct quotes or paraphrased insights from the prospect
   • The prospect's stated opinions, predictions, or challenges
   • Whether the prospect was the host, a guest, or merely mentioned If a page is behind a paywall or lacks useful content, move to the next result.
5. Enrich the company by calling mcp__claude_ai_Amplemarket__enrich_company with the prospect's company domain or linkedin_url. Key data points to extract:
   • Industry and sub-industry
   • Company size and growth trajectory
   • Description and value proposition
   • Recent funding or milestones
   • Tech stack Use this context to connect podcast insights back to the prospect's current business reality.
6. Synthesize findings. Analyze all extracted podcast content and identify the strongest personalization angles based on:
   • Specific quotes or opinions they stated on the podcast. These are the most powerful hooks
   • Topics they are passionate about. Recurring themes across multiple appearances signal genuine interest
   • Predictions they made, especially ones you can validate, reference, or ask about
   • Challenges they described. Pain points they openly discussed are ideal outreach entry points
   • Contrarian or unique takes. Bold stances that set them apart from conventional thinking Rank angles by specificity and relevance to the outreach goal.
7. Generate dynamic fields for each podcast appearance found. Populate the {{podcast_*}} fields listed in the Dynamic Fields Generated section below. If multiple appearances are found, generate a set of fields for each episode, numbered sequentially (e.g., {{podcast_name_1}}, {{podcast_name_2}}).
8. Draft 3-5 outreach openers that naturally reference podcast content. Each opener should:
   • Reference a specific episode, quote, or topic, not generic podcast mentions
   • Connect the podcast insight to a relevant business challenge or opportunity
   • Feel conversational and authentic, not like a templated mail merge
   • Be 1-3 sentences long
   • Vary in angle: one quote-based, one topic-based, one prediction-based, one challenge-based, one shared-interest-based (where applicable)

Important Notes

• Never fabricate quotes. Only use direct quotes that were actually extracted from podcast content via WebFetch. If you cannot verify a quote, paraphrase and note it as paraphrased.
• Distinguish between host, guest, and mention. The prospect's role in the podcast matters. Being a regular host signals thought leadership differently than a one-time guest appearance.
• Quality over quantity. One deeply researched episode with real quotes beats five shallow mentions. Prioritize depth of extraction over breadth of appearances.
• Recency matters. Prioritize recent appearances (last 12-18 months) over older ones, as the prospect's views and role may have evolved.
• Respect content boundaries. If podcast content is behind a paywall or requires authentication, note the episode metadata and skip content extraction rather than attempting to bypass access controls.

Dynamic Fields Generated

Examples

Example 1: Prospect Who Was a Podcast Guest

User prompt: "Find podcast appearances for linkedin.com/in/maria-torres-cro"

What the skill does:

1. Calls mcp__claude_ai_Amplemarket__enrich_person with linkedin_url: "https://linkedin.com/in/maria-torres-cro".
2. Gets back: Maria Torres, CRO at ScaleFlow (scaleflow.io), former VP Sales at Gong.
3. Runs WebSearch with:

• "Maria Torres" podcast
• "Maria Torres" episode guest
• "Maria Torres" interview sales leadership
• "Maria Torres" "ScaleFlow" podcast
• "Maria Torres" webinar OR panel OR fireside

4. Calls WebFetch on the top 4 results.
5. Calls mcp__claude_ai_Amplemarket__enrich_company with domain: "scaleflow.io".
6. Synthesizes findings and generates output.

Example output:

PODCAST PERSONALIZATION BRIEF: Maria Torres

Prospect Summary Maria Torres is the CRO at ScaleFlow, a Series C revenue operations platform with 320 employees. She joined 14 months ago from Gong where she was VP of Sales. She has appeared on at least 3 podcasts in the past year.

Podcast Appearances Found

1. Revenue Architects Podcast - "Scaling Revenue Without Scaling Headcount"

• Host: David Kim
• Date: 2026-01-18
• Role: Guest
• Key topics: Revenue efficiency, automation in sales workflows, the myth of "more reps = more revenue"
• Quote: "We cut our SDR team by 30% and actually increased pipeline by 40%. The secret was not hiring more people. It was giving the right people the right signals at the right time."
• Stance: Strongly believes that AI-driven signal routing will replace traditional SDR qualification within 3 years
• Prediction: "By 2028, the SDR role as we know it will be unrecognizable. The companies that figure out signal-based selling first will have a 2-year head start."

2. SaaStr Sessions - "The CRO Playbook for PLG Companies"

• Host: Jason Lemkin
• Date: 2025-11-04
• Role: Guest
• Key topics: Product-led growth, sales-assist models, when to hire a CRO
• Quote: "The biggest mistake PLG companies make is waiting too long to layer in sales. By the time they realize they need it, they have left 18 months of enterprise revenue on the table."
• Challenge mentioned: Bridging the gap between self-serve users and enterprise sales motion at ScaleFlow

3. Women in Revenue - "Breaking Through the CRO Ceiling"

• Host: Rachel Greenfield
• Date: 2025-08-22
• Role: Guest
• Key topics: Career progression for women in revenue, building executive presence, negotiating CRO roles
• Insight: Shared that she almost turned down the ScaleFlow CRO role because the company had never had a CRO before, so she had to define the role from scratch

Dynamic Fields

Draft Openers

1. Quote-based: "Maria, I caught your episode on Revenue Architects. The stat about cutting SDRs by 30% while growing pipeline 40% stopped me in my tracks. We are seeing similar patterns with our customers and I would love to share what is working."
2. Prediction-based: "Your prediction on the SaaStr podcast about the SDR role being unrecognizable by 2028 really resonated. We are building exactly the kind of signal-based selling layer you described. Would love your take on our approach."
3. Challenge-based: "On the SaaStr episode, you mentioned the challenge of bridging PLG self-serve with enterprise sales at ScaleFlow. That is a problem we have helped 3 other Series C companies solve. Happy to share what worked."
4. Topic-based: "Loved your take on revenue efficiency over headcount growth on the Revenue Architects pod. Most CROs I talk to are still defaulting to 'hire more reps'. Your signal-based approach is ahead of the curve."
5. Personal-based: "Your story on Women in Revenue about defining the CRO role from scratch at ScaleFlow was impressive. Building a revenue org without a blueprint takes a different kind of leadership. Curious how the first year has gone."

Example 2: Prospect Who Hosts Their Own Podcast

User prompt: "Research their podcast - Jake Randall, CEO of DataForge"

What the skill does:

1. Calls mcp__claude_ai_Amplemarket__enrich_person with name: "Jake Randall", company_name: "DataForge".
2. Runs WebSearch queries and discovers Jake is the host of "The Data Stack Show," a podcast with 80+ episodes.
3. Calls WebFetch on the podcast landing page and 3 recent episodes to understand his interviewing focus and recurring themes.
4. Calls mcp__claude_ai_Amplemarket__enrich_company with domain: "dataforge.io".
5. Synthesizes findings.

Example output (abbreviated):

PODCAST PERSONALIZATION BRIEF: Jake Randall

Prospect Summary Jake Randall is the CEO and co-founder of DataForge (dataforge.io), a 150-person Series B data infrastructure company. He hosts "The Data Stack Show," a weekly podcast with 80+ episodes focused on data engineering, modern data stack architecture, and interviews with data leaders.

Key Finding: Jake is a podcast HOST, not just a guest. This signals strong thought leadership investment and means he is deeply familiar with the podcast format. Referencing his show directly is more personal than referencing a guest appearance.

Recurring Themes Across Episodes:

• Data mesh vs. data warehouse architecture debates
• The "unbundling of Snowflake" thesis
• Real-time data pipelines as competitive advantage
• Hiring and retaining data engineers

Draft Openers

1. "Jake, I have been following The Data Stack Show. Your conversation with [recent guest] on real-time pipelines was exactly the debate we are seeing play out with our customers."
2. "As someone who has interviewed 80+ data leaders on your podcast, you have probably heard every take on the modern data stack. Curious if you have seen anyone solving [specific problem] the way we are approaching it."
3. "Your 'unbundling of Snowflake' thesis from The Data Stack Show keeps coming up in conversations I am having with data leaders. Would love to share what we are building in that space. Might make for an interesting episode too."

Example 3: Sparse Results - Only One Appearance Found

User prompt: "What podcasts has this person been on - Priya Sharma, VP Eng at NovaTech?"

What the skill does:

1. Calls mcp__claude_ai_Amplemarket__enrich_person with name: "Priya Sharma", company_name: "NovaTech".
2. Runs all 5 WebSearch query patterns.
3. Only finds 1 relevant result, a webinar panel from 6 months ago.
4. Calls WebFetch on that result.
5. Calls mcp__claude_ai_Amplemarket__enrich_company with NovaTech's domain.
6. Delivers findings with a transparency note about limited results.

Example output (abbreviated):

PODCAST PERSONALIZATION BRIEF: Priya Sharma

Prospect Summary Priya Sharma is the VP of Engineering at NovaTech (novatech.dev), a 90-person Series A developer tools company. Limited podcast presence found. 1 appearance identified.

Note: Only one podcast or media appearance was found. Personalization angles from this single source can still be effective but should be used carefully to avoid appearing overly fixated on a single data point.

1. DevOps Dialogues Webinar Panel - "Scaling Engineering Teams in a Remote-First World"

• Date: 2025-09-10
• Role: Panelist (1 of 4)
• Key topics: Remote hiring, async communication, engineering culture at scale
• Paraphrased insight: Discussed NovaTech's approach of hiring senior engineers in non-traditional tech hubs and investing heavily in documentation culture instead of synchronous meetings
• Challenge mentioned: Maintaining engineering velocity while scaling from 20 to 60 engineers

Draft Openers

1. "Priya, your point on the DevOps Dialogues panel about documentation culture over sync meetings really stood out. We have seen similar results with our engineering customers who made that shift."
2. "Scaling from 20 to 60 engineers while maintaining velocity is no small feat. You mentioned that challenge at the DevOps Dialogues panel. Curious how that journey has gone since then."

Dynamic Fields

Recommendation: Given the sparse podcast results, consider supplementing with other personalization approaches. The prospect may be more active on LinkedIn, conference speaking circuits, or in written content (blog posts, technical articles).

Troubleshooting