Podcast Summary: Using AI at Work — Episode 63

"Using AI for Content, Analytics, and SEO"

Host: Chris Daigle
Guest: Debra Andrews (President, Marketree)
Release Date: July 28, 2025

Episode Overview

In this insightful episode, host Chris Daigle interviews Debra Andrews, President and owner of Marketree, about the real-world applications of generative AI in marketing, analytics, and SEO. The pair dive deep into how marketing was among the first business functions to adopt AI and discuss the strategies, successes, and pitfalls encountered in leveraging AI tools. The conversation offers tactical advice for leaders and teams exploring or expanding AI use in content creation, analytics, and optimizing for generative search engines.

Key Discussion Points & Insights

Debra's Background and the Early Adoption of AI in Marketing

• Debra Andrews introduces Marketree as a strategic marketing consultancy operating on a fractional model, serving as outsourced marketing departments for businesses.

  • "[Marketing] was one of the first areas to be affected by AI, generative AI in particular. So we’ve been actively learning and operationalizing AI for the last 18 months to two years." (02:09)

• Clients don’t ask Marketree about specific tools but are increasingly curious about AI’s impact on marketing strategy and outcomes.

  • "They’re more interested holistically in whether we are embracing AI best practices. They’re relieved that, yes, we’ve heard about it. We’re following the trends." (03:48)

Client Attitudes, Savviness, and Advice for Beginners

• Clients display a range of AI awareness: some are curious, others feel behind or paralyzed by rapid changes.

  • "It’s almost like a paralysis—feeling like, 'Oh, we should be doing something. We don’t know what to do, so we’re doing nothing...'" (04:55)

• Most clients do not yet have a holistic "AI roadmap" for their businesses outside of marketing functions.

• Debra’s primary advice: Gain a foundational understanding of AI through reputable courses (e.g., MIT executive class) and track influential resources (e.g., Marketing AI Institute, Paul Roetzer's podcast and book).

  • "Just having that basic level of understanding was very, very helpful... That I would recommend to anyone to take some sort of a course." (08:39)

• Use templates to map business processes and identify high-impact, easy-to-implement AI use cases before scaling efforts.

  • "You start getting baby steps and getting momentum because you’re like, 'Oh, I just kind of infused AI into a use case. I see I’m saving like 10% of my time.'" (09:51)

Content Generation: Wins, Cautions, and Best Practices

• Content creation was the first big win—starting with internal use, not client work, to perfect workflows and avoid risks.

  • "The first thing people noticed is, wow, it can write blogs, it can write social posts... Our phase one was in the content use cases." (11:00)

• Marketree waited about a year before adopting AI-generated content for client deliverables, prioritizing internal experimentation and quality assurance.

  • "If something’s going to go wrong, I want it to go wrong for my own company and not one of my clients." (12:39)

• Top practical advice:

  • Don’t roll out AI-generated content carelessly—form a committee, develop best practices, and train staff.
  • Subject matter experts must provide the original ideas; never rely on models to generate unique insights.
  • "The thoughts need to be yours... Don’t shortcut the thought process. The thought process still needs to be uniquely human or it’s going to sound like everybody else." (14:34)

• On overcoming the blank page: Use prompts, dictation, or handwritten notes to gather ideas, then let the LLM organize and polish.

  • "I love the fact that I don’t have to sit there and agonize over, 'Does this sentence sound good?'... Large language models are much better at that than I am." (16:00)

AI for Image Generation and Analytics — Trials and Tribulations

• AI Image Generation:

  • Despite experimentation with Gemini's Imagen, ChatGPT's advanced models, and others, output still isn't at a “production quality” level for marketing.

  • "Maybe we need to have a designer putting in prompts... but just as everyday people... we haven’t had a tremendous amount of success with that yet." (18:02)

  • Host’s tip: Try Ideogram and explore new capabilities in ChatGPT 4O for improved results.

• Analytics Integration:

  • The team struggled to unify data sources (HubSpot, Google Analytics, SEMrush, LinkedIn) and generate actionable, AI-driven insights.

  • Past tools like Qlik Data were "clunky," but new adoption of DataBox has streamlined reporting and enabled natural language querying.

  • "We want that kind of level insight into the data... and after carefully vetting... DataBox... does give you a really nice AI insight overview." (21:18)

  • Lesson: Don’t be discouraged by early failures—improved tools are constantly emerging.

Generative Engine Optimization (AIO/LLMO) and the SEO Paradigm Shift

• Early on, Debra and team manually queried LLMs (ChatGPT, Gemini, Claude) to see what surfaced for their business and clients—discovering that reviews and external proof points fed directly into models’ answers.

  • "They were pulling information directly from reviews on Glassdoor, reviews on Google... So, note to self: we need more reviews." (22:45)

• Hired a full-time PR person to ensure every client developed third-party credibility—case studies, reviews, awards for citation and visibility.

• New tools for AIO/LLMO like Scrunch AI and Revere help audit and improve how businesses appear in LLM responses.

  • "Scrunch AI... allows you to put in various prompts related to your business and see how you’re faring... and gives you recommendations." (24:06)

• SEMrush has introduced an "AIO" platform, but is presently expensive and positioned for enterprise users.

  • "But for now for Marketree, we've been pretty happy with Scrunch and we're going to move in that direction." (26:36)

• Content Strategy Changes:

  • Less focus on generic, "top-of-funnel" educational material—LLMs satisfy those queries instantly.
  • Shift toward original case studies, reviews, testimonials, and middle-funnel content unique to the client.
  • "Large language models really like those kind of proof points. So think about case studies, reviews, storytelling." (27:55)

Traffic Shifts and Generative AI's Real Impact

• Dramatic drop in organic search traffic over the last few months (~20%) as users shift queries to LLMs (e.g., ChatGPT, Gemini) instead of Google.

  • "We have seen across our clients like a 20% drop off in organic search. In HubSpot, it'll say referred by ChatGPT, referred by Gemini." (29:42–30:15)

• Clients are being discovered via listicles created by LLMs, reinforcing the importance of reputation, reviews, and third-party citations.

  • "Every single case... they’ve already done their homework. Now they're asking ChatGPT to give the top sources and they're going to those sites." (31:10)

  • "That's why your review strategy is so critical." (31:30)

Change Management and AI Adoption Internally & With Clients

• In regulated industries, clients want clarity on safe AI use but more often express curiosity and hunger for learning over outright fear.

  • "They really want to understand like what we’re doing to make sure that we’re not putting them at risk." (33:09)

• Most internal resistance is rooted in lack of understanding; early company efforts focused on playful, non-critical tool exploration to build comfort.

  • "We had a ChatGPT team, a Perplexity team... asked that they would just spend a week experimenting, then share their findings. Just play." (35:46)

• Debra advocates for AI literacy at every level, forming councils or committees to champion education, discussion, and incremental experimentation.

Notable Quotes & Memorable Moments

• On Foundational AI Education:

  • "It's not ChatGPT, right?… Understand what generative AI is. How does it work?" (Debra, 08:10)

• On the Human Element in Content:

  • "Don’t shortcut the thought process. The thought process still needs to be uniquely human or it’s going to sound like everybody else." (Debra, 14:34)

• On Analytics Frustrations:

  • "We want that kind of level insight into the data… and after carefully vetting… DataBox… does give you a really nice AI insight overview." (Debra, 21:18)

• On Reviews as a Ranking Factor:

  • "Note to self, we need more reviews, we need more third-party proof points because it's drawing out to draw conclusions." (Debra, 22:45)

• On SEO vs. AIO:

  • "If you’re going to write educational [content], make it wow… Talk about something that you cannot find on a large language model and that’s only going to happen because you have the experience." (Debra, 28:00)

• On Generative AI Traffic:

  • "Every single case... it’s been a listicle. They’ve asked for the top companies... then they're going to those sites." (Debra, 31:05)

Timestamps of Key Segments

• [01:56] Debra’s Background & Marketree’s Approach
• [04:21] How Clients Perceive and Approach AI
• [07:04] Debra’s Foundational Advice on Getting Started with AI
• [10:46] Early Use Cases and Mapping Business Functions
• [11:53] Content Creation Wins and Model “Best Practices”
• [13:48] Warnings: Don’t Automate Without Training; Original Insights Needed
• [17:44] AI Image Generation – Roadblocks and Recommendations
• [19:40] Analytics Integration—Failures and Eventual Success with DataBox
• [22:20] Discovering How LLMs Source Company Info
• [24:06] New AIO/LLMO Tools: Scrunch AI, Revere, and SEMrush AIO
• [27:06] Changing Content Strategy for the Generative Era
• [29:37] Measurable Drop in Organic Search and the Impact of Generative Search
• [31:30] Review Strategies and Third-Party Proof Points
• [33:05] Handling AI Change Management—Industry Differences
• [35:27] Internal Team AI Exploration and Literacy Building
• [37:10] How to Access Marketree’s AI Resources and Webinars

Further Engagement

Resources:

• marketree.com — For AI content, webinars, and learning resources
• Email Debra: dandrews@marketree.com for webinar recordings
• Check out tools mentioned: DataBox, Scrunch AI, Ideogram

Overall Takeaway:
Debra’s experience underlines that AI is not a passing fad or mere tool, but a generational business shift. Starting with foundational AI education, experimenting in low-risk environments, and focusing on uniquely human insights are key to lasting success. Meanwhile, reputation signals such as reviews are becoming crucial not just for search engines, but for generative AI discovery and recommendations as well.