Podcast Summary: Marketing Against The Grain

Episode Title: The AI Skill Ladder (Beginner → Workflow Builder)
Hosts: Kipp Bodnar (HubSpot CMO), Kieran Flanagan (SVP of Marketing)
Guest: Kevin Hudson (Futurepedia)
Release Date: April 8, 2026

Episode Overview

This episode takes listeners on a practical journey through the “AI Skill Ladder”—from AI novice (question asker) to advanced “workflow builder.” Guest Kevin Hudson (Futurepedia) shares a structured approach to building AI fluency, discusses the tools and mindsets essential for layer-by-layer advancement, and showcases detailed real-world use cases (with a special emphasis on workflow automation and content creation). The tone is actionable, energetic, and cutting-edge, offering listeners both concrete strategies and a mindset shift to succeed with AI in marketing.

Key Discussion Points & Insights

1. Understanding the AI Skill Ladder

[01:15 – 03:16]

• Level 1: Question Asker
  • Most people begin their AI journey by using language models (LLMs) like ChatGPT or Claude as search engines: “You've just replaced Google with ChatGPT or Claude.” – Kevin Hudson [01:15]
• Level 2: Prompt Designer
  • Users discover that adding instructions, context, examples, and constraints significantly improves results: “You start putting in clear instructions, add context, examples and constraints, it just changes everything.” – Kevin Hudson [01:23]
• Level 3: Power User
  • Unlocking tool-specific features (like Claude’s Projects and Skills).
  • Projects serve as “AI second brains”—organizing instructions, files, and retaining context across sessions:
    “I have a cloud project for every singular kind of project...it has the ability to kind of talk through those different projects with you in a really intelligent way.” – Host 2 [02:55]
• Level 4+: Workflow Builder/Weaver
  • Combining multiple tools into interconnected workflows; automating multi-step processes.

2. Tool Deep Dive & Use Cases

[03:42 – 17:20]

a. Customization in Claude: Projects and Skills

• Projects: Organize contexts, guidelines, and SOPs for each major activity or channel.
• Skills: Reusable, multi-step routines (e.g., content repurposer) that can draw context, customize output, and automate repetitive creative tasks.
  “This one’s a content repurposer…[Claude] drafts posts for X and LinkedIn.” – Kevin Hudson [03:42]

b. Manus: Autonomous AI Agents

• What is Manus?
  • An AI agent platform that acts as a digital employee, able to autonomously execute complex, multi-step tasks by deciding which models or tools to use for each subtask.
    “Manus is autonomous…it executes complex multi step tasks all on its own.” – Kevin Hudson [05:17]
• Key Advantages
  • Entry-level friendly compared to code-heavy platforms.
  • Model switching: For example, Manus uses Gemini for YouTube consumption (transcripts and visual understanding), image generation, PDF creation, web scraping and data entry, even building sheets of leads complete with info.
  • Full end-to-end execution: “It just figures out what tools it needs and gets the job done.” – Kevin Hudson [05:17]
• Concrete Use Case:
  • Creating a lead magnet PDF from a YouTube video:
    “So it'll go out and actually watch that video right on YouTube, then it'll develop all the copy it needs to put into that PDF…and design the whole thing on its own using those colors [you upload].” – Kevin Hudson [06:08]

c. Workflow Automation & Content Research at Scale

• Using Manus for research and content mapping:
  • Ingesting inputs (YouTube links, Reddit threads, comments) and outputting fully researched reports, content angles, B-roll recommendations, and audience pain points:
    “I have it…read a bunch of different subreddits, finds YouTube comments…and isolates pain points from all of those.” – Kevin Hudson [10:17]

d. Landing Page & App Generation with Lovable and Google AI Studio

• Lovable:
  • Instantly generates landing pages to host downloadable offers, connect to email collection (via API), all from a single prompt.
    "If I have this prompt here to generate a landing page...I can drop in that PDF that Manus just created." – Kevin Hudson [13:08]
• Google AI Studio:
  • Rapid prototyping of interactive internal tools and software, affordable for experimentation:
    “You can do a lot just for free…create any sort of internal tool completely for free and…give stuff like that out for free.” – Kevin Hudson [14:02]

e. The Impact of "Show, Don't Tell" in Lead Magnets

[15:05 – 16:35]

• Instead of traditional static content, users can now give away functional software/apps as lead magnets:
  “Software code is essentially becoming disposable...You don’t have to maintain it, don’t have to do all of the old software work that you used to have to do.” – Host 1 [15:05]

3. Mindset: Moving Up the Ladder

[12:43 – 13:05]

• The leap from prompt engineering to workflow-building is less about technical skill and more about curiosity and willingness to learn:
  “Really, most of it is just curiosity and the willingness to do it…You could learn any of these tools in a weekend.” – Kevin Hudson [12:43]

• Iteration matters: skills and flows should be updated with feedback or when things don’t “sound right.”
  “Anytime it just does something you don't want, you can just ask it right in the chat: update this part of it.” – Kevin Hudson [12:18]

4. Standout Quotes & Memorable Moments

• "I think of it as my AI second brain." – Host 2 [02:55]
• “Manus...can switch between models for whatever task is relevant to that model...that is a real advantage.” – Host 2 [07:25]
• "You’re finding where the arbitrage in the market is for what to create. And then you’re also using the same tool to kind of repurpose that content once you’ve built it.” – Host 1 [10:41]
• “Software code is essentially becoming disposable...it’s a good reminder of the show don’t tell.” – Host 1 [15:05]
• “Once you see the power of it, pretty much anything you want to do, you just add in a little bit of context and it’ll just go off and get it done for you.” – Kevin Hudson [18:01]

Timestamps for Important Segments

• Defining the AI Skill Ladder: [01:15–03:16]
• Tool Deep Dive (Claude, Projects, Skills): [02:30–04:18]
• Transition to Workflow Building: [04:25–05:17]
• Demonstration of Manus for Multi-Step Content Creation: [05:17–07:11]
• Use Case: Creating PDFs, Research and B-roll with Manus: [07:53–09:35]
• Extracting Insights from YouTube and Reddit: [10:17–10:41]
• Lead Generation, Data Research Automation: [07:53–08:37]
• Automating Landing Pages with Lovable: [13:08–13:44]
• Show, Don’t Tell—Giving Away Software: [15:02–15:26]
• Building Custom Apps on Google AI Studio: [14:02–16:10]
• Mindset: The Core Difference in Leveling Up: [12:43–13:05]

Key Takeaways for Marketers & Builders

• Expand Your Tools: Don’t get stuck at prompting—explore Projects, Skills, and autonomous agents.
• Iterate & Customize: Building workflows is about trying, tweaking, and updating. The biggest barrier is willingness, not skill.
• Think Like a Builder: Go beyond text outputs—use AI to assemble apps, automate research, and deliver value in new ways.
• Leverage Content “Arbitrage”: Use AI to find gaps, differentiate, and repurpose winning ideas at scale.
• Lead Magnets, Upgraded: The most compelling offers are interactive or functional—not just PDFs, but mini-apps or software.

Practical Next Steps

• Try Manus or similar agent tools for multi-step content projects (e.g., research + writing + design).
• Use Claude’s Projects & Skills or GPT-4’s custom instructions for persistent workflows.
• Experiment with generating (and giving away) apps or interactive lead magnets using Google AI Studio or Lovable.
• Foster curiosity—pick a weekend, pick a tool, and explore hands-on.

For further insights, check out the “AI Skill Ladder” free guide referenced in this episode and explore the Marketing Against the Grain newsletter for ongoing frameworks and tutorials.