#33- TeenAIge Dream: How to Build Custom GPT that saves HOURS, Agents Galore, The Intelligence Age, and Notebook LLM.

The newsletter and GTM AI Podcast are sponsored by GTM AI Academy AND AI Powered GTM .

Business Impact> Learning Tools

1. We have the SLACK COMMUNITY around AI , come hang with your fellow GTM professionals, get insights, and content free to help in your teams and roles.
2. The GTM AI Demo Tools Library is a free resource for you to view AI tools and demos to see what is out there to help your team or to help you in your own workflow
3. We have a Weekly AI Collab starting on Oct 3rd for free and anyone is welcome! Hit me up or comment with your email to be invited.

Now with all that being said, lets move forward with todays newsletter which is:

1. We have #33 Podcast episode and have the pleasure of collaborating with my cohost Jonathan Moss where I pick his brain on custom GPTs, his thinking and strategy around using them, and how it saves him hours of time.
2. Agentforce breakdown from Dreamforce and Salesforce (I know its a lot of forces in one sentence)
3. Agent AI, Hubspots response to the above
4. Sam Altman talks about the coming Intelligence Age
5. AI Tool of the week: Notebook LLM

Some AI posts from this last week in case you missed it:

Sam Altman and the 5 levels of AI, hint, we are at level 2

Sample Custom Instructions for ChatGPT

BBVA Open Mind 2025 AI Trends and my GTM Breakdown

Step 1 of 7 in AI Enablement

Now with that, lets goooooo

You can go to Youtube , Apple , Spotify , or here on Linkedin as well as a whole other host of locations to hear the podcast or see the video interview.

How to Build a Custom GPT That Saves Hours

I had my man Jonathan Moss on the show again, and we dove deep into the world of custom GPTs for go-to-market strategies. Jonathan's been talking about this stuff for months, and he finally showed us the goods. He built this badass GPT that helps create comprehensive GTM plans, and let me tell you, it's impressive. We got into the nitty-gritty of how it works, why it's so useful, and the thought process behind building it. Jonathan even gave us a live demo, showing how this thing generates product descriptions, ICPs, buyer personas, and a whole lot more. It was like watching a master craftsman at work.

You can find his Custom GPT and process here

What really blew my mind was how much time this tool saves compared to the old-school manual way of doing things. We're talking weeks or months of work condensed into a couple of hours or even minutes. Jonathan and I got into a great discussion about the importance of domain expertise when using AI - you can't just expect the machine to do all the work, you gotta bring your A-game too. We also talked about making these GPTs user-friendly, because what's the point of a tool if no one can figure out how to use it? The conversation got me fired up about the potential of this tech, especially when we started talking about using GPTs for multilingual team enablement. It's crazy to think about how far we've come and where this tech could take us in the future.

Bullet highlights:

• Jonathan sees GPTs as starter kits for AI agents
• His GTM GPT incorporates 12 key steps, from product description to risk analysis
• The GPT uses a knowledge base of trusted GTM frameworks and resources
• It generates tailored content like ICPs, personas, and sales messaging
• They discuss how this drastically cuts down strategy development time
• Importance of proper context and prompting in getting good AI outputs
• Using GPTs for multilingual team enablement

Key quotes:

Coach:

"They used to say that Content is king, but with AI Context is king. Because if people know context then - Trade and mark that. Trade and mark that! Seriously, because every time I prompt and I get something wrong, I'm like, I didn't give enough context. It's my fault. I'm the problem. It's not the AI's fault."

Jonathan:

"Think about putting like a go to market strategy like this, or even a component of that... You would sit, you would think you would ask people, you'd try to figure it out. You'd have to pull all the resources in and stuff. It would take people, it would take weeks or months to figure this out. And then now you can just have a starting point."

Jonathan:

"AI is not just going to go and do things for you. It's no silver bullet, but those that have domain expertise... use AI to just catapult that to another level."

Jonathan:

"If I go through this process, it already has so much information and insight about my go to market. It can provide me a better result on what it is that I'm looking for. And that's really the reason that I come back and use it on and on again."

Check out the episode, tell me what you think!

Before we dive into Agentforce and Agent.ai from hubspot, consider this where we are heading according to Sam Altman of OpenAI:

Now with you quickly looking over that, lets dive into Salesforce 's new Agentforce:

AgentForce is Salesforce's ambitious new AI platform designed to create autonomous agents that can take actions, not just answer questions. It's positioned as the next evolution of CRM, combining AI, human oversight, data, and existing Salesforce capabilities.

Key points:

- Aims to close the productivity gap by automating complex tasks

- Integrates with existing Salesforce data and workflows

- Promises to be more flexible and capable than traditional chatbots or co-pilots

Critical observation: While the concept is compelling, the real test will be in its practical implementation across diverse business environments. The complexity of real-world scenarios may challenge the system's ability to make nuanced decisions.

Core Components and Features:

a) Reasoning Engine:

- Central to AgentForce's functionality

- Makes decisions based on context, data, and available actions

- Adapts to various scenarios without extensive pre-programming

b) Agent Builder:

- Tool for customizing out-of-the-box agents or creating new ones

- Uses natural language instructions to define agent behaviors

- Integrates with existing Salesforce tools (Flows, Apex, MuleSoft APIs)

c) Data Integration:

- Leverages Salesforce's Data Cloud for structured and unstructured data

- Uses vector search and embedding models for efficient data retrieval

- Aims to provide agents with real-time, relevant information

d) Action Framework:

- Allows agents to execute tasks using existing Salesforce automations

- Integrates with partner ecosystems for extended capabilities

Critical observation: The success of AgentForce will heavily depend on the quality of data and the accuracy of its reasoning engine. There's potential for errors if the data is incomplete or the reasoning isn't sufficiently sophisticated for complex business scenarios.

Use Cases and Customer Examples:

a) Wiley (Educational Resources):

- Expanded self-service topics from most common to comprehensive coverage

- Provided personalized, dynamic responses

- Achieved a 40% increase in case resolution compared to their old chatbot

- Notably simplified the conversation design process

b) EasyCater (Food Ordering Platform):

- Implemented AI for personalized catering recommendations

- Integrated order history and customer preferences into decision-making

- Automated order placement while maintaining the option for human oversight

Critical observation: While these examples are impressive, they represent controlled environments with specific use cases. The true test will be how AgentForce performs across various industries and more complex scenarios.

Implications for Go-To-Market (GTM) Professionals:

a) Potential Benefits:

- Automation of repetitive tasks, freeing up time for strategic work

- 24/7 customer support with personalized interactions

- Scalable operations without proportional increase in human resources

- Leveraging of existing Salesforce investments and data

b) Challenges to Consider:

- Potential need for significant data cleanup and organization

- Training requirements for teams to effectively use and manage AI agents

- Balancing automation with the human touch in customer interactions

- Ensuring compliance and ethical use of AI in customer-facing roles

c) Strategic Considerations:

- Identifying high-impact areas for AI implementation in sales and marketing processes

- Developing a phased approach to adoption, starting with pilot programs

- Planning for change management and potential role evolutions within teams

- Assessing the ROI of AgentForce against other potential investments

Technical Aspects and Integration:

a) Platform Integration:

- Built on existing Salesforce architecture, leveraging familiar tools

- Integrates with Salesforce Data Cloud for comprehensive data access

- Uses Salesforce Einstein for AI-driven insights

b) Customization and Flexibility:

- Allows for extensive customization through natural language instructions

- Supports integration of external APIs and partner solutions

- Provides tools for testing and refining agent behaviors

c) Security and Compliance:

- Built on Salesforce's trust layer, emphasizing data safety

- Allows for setting guardrails and limits on agent actions

Critical observation: While the integration with existing Salesforce tools is a strength, it may also limit adoption for companies not already deeply invested in the Salesforce ecosystem. The learning curve for effective customization could be steep.

Rollout and Availability:

- Most features announced are set to be Generally Available (GA) in October 2024

- Spring release (February 2025) will introduce advanced features like batch testing, improved search, and multimodality support

- Currently in pilot phase with select customers

Recommendations for GTM Professionals:

a) Immediate Actions:

- Whether you use Salesforce or not, consider the implications of what is happening generally as we move forward and how it will impact both internal and external workflows.

- Watch Dreamforce sessions for deeper insights (MuleSoft, Platform keynotes) above

- Visit the Launch Zone to get hands-on experience with agent building

- Begin internal discussions about potential use cases and impact

b) Short-term Planning:

- Conduct a thorough assessment of current sales, marketing, and service processes

- Identify areas where AI automation could have the most significant impact

- Develop a business case for AgentForce adoption, including potential ROI and risks

c) Long-term Strategy:

- Plan for a phased implementation approach, starting with pilot programs

- Develop a data strategy to ensure AI agents have access to high-quality, relevant information

- Consider the long-term implications for team structures and skill requirements

- Stay informed about AgentForce updates and best practices as they evolve

Final Thoughts:

AgentForce represents a significant leap in AI capabilities for CRM and customer engagement. However, its success will depend on factors beyond just technological capabilities. GTM professionals should approach it with both excitement and caution, carefully considering how it fits into their overall strategy and customer engagement philosophy. The potential for improved efficiency and customer experience is substantial, but so are the challenges of implementation and ensuring that the human element of customer relationships isn't lost in the process of automation.

Agent.ai is Hubspots answer to the new AI AGENT driven world.

HubSpot's Agent.AI is a new platform designed as a "professional network for AI agents." It allows users to create, customize, and deploy autonomous AI agents for various business tasks without extensive coding skills.

The platform supports integration with multiple AI models (like GPT-4 and Claude 3.5) and enables "agent composition" for complex tasks. Agent.AI uses a credit-based system, with users receiving 100 free credits upon signup. It's positioned as a flexible, community-driven ecosystem where users can potentially share and monetize their custom agents.

The platform aims to democratize AI capabilities, allowing marketers, salespeople, and other professionals to automate tasks, enhance productivity, and unlock new capabilities in areas such as content creation, research, and customer support. While still in its early stages, Agent.AI represents HubSpot's vision of a future where teams consist of both human and AI members, potentially reshaping how businesses approach software tools and automation.

1. Overview of Agent.AI :

- HubSpot's new platform for autonomous AI agents

- Positioned as a "professional network for AI agents"

- Allows users to create, customize, and deploy AI agents for various tasks

- Will eventually include a marketplace-like system for sharing and monetizing agents

2. Key Features and Capabilities:

- Users can build agents without coding skills

- Agents can perform multi-step tasks and reason through complex problems

- Supports integration with multiple AI models (e.g., GPT-4, Claude 3.5)

- Allows for "agent composition" - combining multiple agents for more complex tasks

- Includes a credit system for usage (100 free credits to start)

3. Implications for GTM Professionals:

- Potential to automate and enhance various marketing, sales, and customer service tasks

- Opportunity to create and monetize specialized agents based on domain expertise

- New skill set emerging around "AI management" and effective prompting

- Potential to significantly increase productivity and unlock new capabilities

4. Example Use Cases:

- Content creation (e.g., LinkedIn posts from YouTube videos)

- Company research and lead generation

- Automated customer support

- Data analysis and insights generation

5. Potential Impact on Software Industry:

- Democratization of software development

- Potential disruption of traditional SaaS business models

- Increased focus on problem-solving and domain expertise rather than coding skills

6. Challenges and Considerations:

- Balancing sharing vs. protecting intellectual property in agent design

- Ethical considerations around AI-generated content and automation

- Need for humans to develop new skills in AI management and prompt engineering

7. Future Outlook:

- Prediction of hybrid teams consisting of humans and AI agents

- Potential for AI to create new job opportunities rather than just replace existing roles

- Emphasis on curiosity and continuous learning to stay competitive

Critical Analysis:

Agent.AI represents a significant step towards democratizing AI capabilities for businesses. However, its success will depend on several factors:

1. Quality and reliability of agents: As the platform grows, ensuring consistent quality across user-generated agents will be crucial.

2. Ethical and legal considerations: HubSpot will need to address potential issues around AI-generated content, data privacy, and intellectual property.

3. User adoption and learning curve: While the platform aims to be user-friendly, there's still a learning curve for effective AI management.

4. Impact on existing HubSpot ecosystem: It's unclear how Agent.AI will integrate with or potentially disrupt HubSpot's existing product suite.

5. Competition: As other major players (e.g., Salesforce) are also entering this space, HubSpot will need to differentiate its offering.

For GTM professionals, Agent.AI presents both opportunities and challenges. It offers the potential to significantly enhance productivity and capabilities, but also requires developing new skills and rethinking existing processes. Those who can effectively leverage this technology may gain a significant competitive advantage, but it's important to approach it strategically and consider the broader implications for your business and industry.

So here is the question, which one do you use? I want to be clear that there are many options of CRMs, but since these are the two top options and because of their respective Agents launch, lets compare:

1. Platform Approach:

Salesforce AgentForce: Integrated into the Salesforce ecosystem, focusing on enhancing existing CRM functionalities.

HubSpot Agent.AI : Positioned as a standalone "professional network for AI agents," potentially more open and flexible.

2. Target Users:

Salesforce: Primarily aimed at existing Salesforce customers, especially larger enterprises.

HubSpot: Seems to target a broader range of users, including small businesses and individual professionals.

3. Development Model:

Salesforce: Offers pre-built agents for specific functions (sales, service, marketing) and a builder for customization.

HubSpot: Emphasizes user-generated agents, positioning itself as a platform for creators and developers.

4. Integration:

Salesforce: Deeply integrated with Salesforce's Customer 360 data and existing workflows.

HubSpot: Integrates with HubSpot's CRM but also seems designed to work independently.

5. AI Model Usage:

Salesforce: Primarily uses their own AI models (Einstein) and reasoning engine.

HubSpot: Allows integration with multiple AI models (GPT-4, Claude, etc.), offering more flexibility.

6. Monetization:

Salesforce: Likely to be part of broader Salesforce licensing.

HubSpot: Introduces a credit system, hinting at a more flexible, usage-based model.

7. Focus Areas:

Salesforce: Heavily focused on customer-facing operations (sales, service, marketing).

HubSpot: Seems to have a broader scope, including content creation and various business tasks.

8. Community Aspect:

Salesforce: Less emphasis on community-driven development.

HubSpot: Positions itself as a network, suggesting a stronger focus on community and sharing.

9. Customization:

Salesforce: Offers a structured approach with pre-built components and customization options.

HubSpot: Appears to offer more open-ended customization, allowing users to build agents from scratch.

10. Deployment:

Salesforce: Integrated deployment within Salesforce apps and workflows.

HubSpot: Seems to offer more flexible deployment options, potentially beyond just HubSpot's ecosystem.

Critical Analysis:

Both platforms are leveraging their strengths: Salesforce is capitalizing on its dominant position in enterprise CRM, offering a more structured and integrated approach. This may appeal to larger organizations that prioritize consistency and compliance.

HubSpot, true to its roots, is taking a more open, community-driven approach. This could be more attractive to smaller businesses, startups, and individual professionals who value flexibility and creativity.

The success of each platform will likely depend on:

1. Ease of use and learning curve

2. Quality and reliability of the AI agents

3. Integration capabilities with existing workflows

4. Pricing and accessibility

5. The ecosystem of developers and users they can attract

For GTM professionals, the choice between these platforms may depend on:

1. Your existing CRM ecosystem

2. The level of customization and flexibility you need

3. Your team's technical capabilities

4. The specific use cases you're targeting

Ultimately, both platforms represent significant advancements in bringing AI capabilities to CRM and marketing operations. The competition between them is likely to drive rapid innovation in this space, benefiting users regardless of which platform they choose.

Sam Altman released his view of where are heading with the INTELLIGENCE age yesterday , I wanted to give a few thoughts about the article. I wanted to take a critical or observation stance and give a few thoughts:

The article paints a pretty rosy picture of AI's potential. And sure, there's plenty to be excited about. The idea of having a personal AI team to supercharge our work is tantalizing. For GTM pros, the prospect of hyper-personalized campaigns and real-time market insights could be a game-changer. Imagine being able to predict and respond to market shifts before they even happen, which is mind blowing in and of itself.

But let's pump the brakes a bit. We've heard grand promises about technology before, and reality often falls short. The author's vision of AI solving all our problems - from climate change to space colonization - feels a tad optimistic. It's easy to get caught up in the hype, but GTM professionals need to approach this with a healthy dose of skepticism.

One thing that gives me pause is the potential for AI to widen the gap between companies that can afford cutting-edge AI and those that can't. Could we see a world where only the big players have access to these game-changing tools? That could seriously shake up competitive landscapes across industries. To his credit, Sam wants everyone to have access to AI like most do to the internet, but even then, it seems there could develop a divide.

The article glosses over some pretty significant challenges too. Data privacy concerns, algorithmic bias, and the ethical implications of AI-driven decision making are huge hurdles we'll need to overcome. GTM teams might find themselves navigating a minefield of regulatory and ethical issues as they try to leverage AI.

And let's talk about jobs. While the author is optimistic about new opportunities emerging, the reality is that AI could displace a lot of traditional GTM roles. The transition period could be rough, and not everyone will be able to adapt quickly enough. We might see a significant reshuffling of skills and roles in the GTM world.

That said, I can't help but feel a bit of excitement about the possibilities. The potential for AI to handle repetitive tasks and free up GTM pros to focus on strategy and creativity is genuinely thrilling. And if AI can help us understand and serve our customers better, that's a win for everyone.

The author's point about the pace of innovation is spot on. GTM teams will need to become incredibly agile, constantly learning and adapting to new AI-driven tools and strategies. It's both exhilarating and exhausting to contemplate.

Ultimately, while the article's vision of the future feels a bit too utopian, it does highlight some important trends GTM professionals need to watch. AI is coming, whether we're ready or not. The smart move is to start preparing now - experimenting with AI tools, upskilling teams, and thinking critically about how to blend human expertise with AI capabilities.

The "Intelligence Age" might not be the unmitigated boon the article suggests, but it's certainly going to shake things up. GTM pros who can navigate this new landscape - balancing the potential of AI with its pitfalls - will be the ones who thrive. It's going to be a challenging, exciting, and probably pretty bumpy ride. Buckle up, folks.

GTM AI Tool of the week, NotebookLM

A ton of people have been talking about this tool since its release and its built all on Gemini 1.5 pro, Googles best AI so far. It is like having a few tools in one all for free and just recently developed the ability for an AI to create a conversational podcast style conversation about a blog, PDF, website, you name it, it will talk about it, a sample I uploaded is below where all I put in was my www.gtmaiacademy.com website and this is what it produced, (please note, it produced the entire thing by itself, I had no control over the specific content, so any weirdness know it was not me, but still pretty cool):

Now lets talk about what you can do with this tool, again for free:

Google's NotebookLM is an innovative AI-powered research and analysis tool designed to revolutionize how we interact with and extract insights from large volumes of information. This tool allows users to upload up to 50 documents, including PDFs, Google Docs, and web pages, creating a personalized knowledge base. NotebookLM then employs advanced AI to help users navigate, understand, and synthesize this information through a chat-like interface.

What sets NotebookLM apart is its ability to provide context-aware responses, always citing its sources from within the uploaded documents. This ensures that the AI's answers are grounded in the specific information you've provided, rather than pulling from a broader, potentially less relevant knowledge base. The tool also offers features like suggested follow-up questions, note-taking capabilities, and even the ability to generate audio summaries of your research, making it a versatile assistant for various professional applications.

Key Features:

? Document upload (up to 50 documents)

? AI-powered chat interface for document interaction

? Source citation for all AI responses

? Suggested follow-up questions

? Note-taking and saving capabilities

? Audio summary generation

? Integration with Google Docs and Slides

Possible Use Cases:

For Sales:

? Quickly digest and summarize lengthy client briefs or RFPs

? Create personalized sales collateral by analyzing multiple product documents

? Prepare for client meetings by synthesizing account history and industry trends

? Generate tailored FAQs for specific products or services

For Customer Success:

? Analyze customer feedback across multiple channels to identify trends

? Create comprehensive onboarding materials by combining various product guides

? Quickly find solutions to customer issues by querying a database of support documents

? Generate personalized check-in scripts based on account history and product usage

For Enablement:

? Develop training materials by synthesizing product documentation, market research, and best practices

? Create interactive learning modules using the chat interface for Q&A sessions

? Generate audio summaries of key policies or procedures for on-the-go learning

? Curate a knowledge base of frequently asked questions and their answers

For Marketing:

? Conduct in-depth competitor analysis by uploading and analyzing multiple sources

? Create content briefs by synthesizing industry reports, customer feedback, and internal data

? Generate ideas for blog posts, social media content, or email campaigns based on trending topics in uploaded documents

? Develop comprehensive buyer personas by analyzing customer interviews, survey results, and market research

NotebookLM presents an exciting opportunity for GTM professionals to enhance their research capabilities, streamline information processing, and generate insights more efficiently. By leveraging this tool, teams can potentially save time on routine research tasks, allowing them to focus more on strategic thinking and creative problem-solving. As AI continues to evolve, tools like NotebookLM are likely to become increasingly integral to the GTM workflow, offering new ways to handle information overload and extract valuable insights from complex data sets.

PHEW that was a lot, tell me what your thoughts are and any questions you have, until next time adios amigos.