If you only check one highlight this week:

• B2B CMO? The evolution of the B2B marketing org

• Everyone else: how an AI-native works and thinks (bottom of the lead section)

-François

The New GTM org and the ChatGPT-native Talent

Sydney Sloan, the CMO at G2*, kindly invited me to co-lead a workshop with her at their Executive Advisory Board offsite recently. (see how I crated my presentation in the next section)

The topic: How B2B CMOs and CROs are rethinking their organizations to seize the opportunities of AI.

1. Are we witnessing the emergence of new org designs and roles? If so, which ones?

2. How are we accelerating the adoption and impact of AI on our teams?

I've addressed the 2nd question in prior editions (org charts, offsites, recruiting, and other tips), so let’s explore org design and new roles.

I reported that we are still seeing the same org designs, i.e. departments, in marketing and sales. I have not seen any CMO fundamentally break that mold yet. If you have, please let me know.

For marketing, that means, if I oversimplify (knowing that content can go anywhere):

• Product Marketing

• Growth / Demand generation (w or w/o Mkg Ops) /Field Marketing

• Brand/Corporate Marketing

However, we are already starting to see, especially in organizations that are ahead of the curve when it comes to adopting AI & agents:

• Flatter organizations - yes, that means streamlined teams, especially at AI-native startups

• Teams publishing org charts with humans who also manage “task agents” (hide these from your CFOs to preserve your headcount 😊)

• New AI-focused roles that equip these teams with AI tooling, workflows and knowledge bases. I've already written about the marketing intelligence engineer (renamed curator below).

Sydney then eloquently challenged the group to consider a redesign of the entire GTM group, around three AI-augmented teams:

1. Team 1: Brand Building & Signal Capture. Combines brand and demand functions. The lines are getting fuzzier anyway.

2. Team 2: Relationships & Engagement. Nurtures relationships across buying committees. Keeps champions engaged in coordination with account owners.

3. Team 3: Product Delight. Brings together PMM, customer marketing and CSM to connect customer challenges with product capabilities and best practices.

Is anyone game to try that?

Now, what type of talent will thrive in marketing going forward? Some argue that we will see two main types emerge:

1. Type 1: the artist types. Those with superb taste and talent for putting together great messaging, visuals, videos, and experiences. Assets that attract and move us.

2. Type 2: the builder types. Those who assemble different data and tools to deliver these unique - ideally ultra personalized - messages and experiences to customers in the right channels

It's a simplistic view, but we will definitely see some of that.

We're especially witnessing the rise of the junior talent that has always had AI as a partner since entering the workforce.

They are the ones who excel in these flatter org charts. They move fast, they learn even faster. To them, starting with AI is the default path.

Would you like to meet one of them and gain insight into how they work and think?

Check out this lightning talk, where Molisha, who currently fills three different roles at Augment Code (launch product marketer, social media manager, and now AI GTM engineer), explains how she uses AI to learn and automation to do more with less (including coordinating product launches and engaging ambassadors on X.com).

She shows how we also should work going forward.

*also an early supporter of these highlights. 🙏

My selection of tips, news and workflows

🌭 Sausage making: how I created my preso for the G2 Advisory Board

I had a 6-week head start, so I used ChatGPT to compile relevant intel weekly and store it in a project:

1. I first created a project folder in ChatGPT

2. Within that folder, I asked ChatGPT to run a deep research and create a detailed weekly report that surfaced the latest trends, the five most commented and liked articles of the last week about how AI is changing B2B marketing & Sales, from org design, to new roles, to new skill sets, etc.

3. I told ChatGPT to run that search weekly.

4. After a few weeks, I asked ChatGPT to identify the top trends, case studies, and new roles from all that information.

5. For each finding, I requested the most representative article, job posting, or case study and its link.

6. Still chatting in that Project, I then dictated a detailed prompt, including the context, my goals, the target audience, the title of my talk, plus shared all my initial thoughts.

7. I asked it to craft a detailed outline and iterated on it in a ChatGPT canvas, still using dictation.

8. Once happy with the outline, I specified for each slide: a title, body (text or prompt a desired graphic or both), speaker notes.

9. I asked ChatGPT to compile a detailed final prompt, including what’s above + what it knows of my existing writing style and brand slide style guide.

10. I then dropped that detailed prompt into 3 AI slide makers:

    • Manus created broken PPT slides.

    • Genspak.io created bland ones.

    • Gamma.app was the winner. Faster and more on-brand. Included images that were all in the same style. Most importantly, the conversion to PowerPoint was perfect and didn't need formatting adjustments.

11. I uploaded the PowerPoint to Google Drive and converted it to Google Slides.

12. I finalized the slides manually.

13. During my drive to the off-site, I initiated a discussion with advanced voice mode, from that project folder, to dive deeper into examples and case studies.
    Now… I then caught ChatGPT mentioning fake case studies and job postings. It apologized profusely when I caught it, promising it would never do that again. Only to do just that a few seconds later. At least now, we all have a pretty good spidey sense for when it makes sh*t up…

All in all, I spent very little time designing the slides. But more than I thought, working on the outline. Please, email me back if you have a faster workflow. 🙏

🛠️ When should you create a custom GPT for your team vs a shared project?

Another good ChatGPT tip from Nicole Leffer:

“The rule of thumb that covers most situations 👇

• If you’re building an AI TOOL for your team members (meaning you want them to have access to the functionality, but you don’t need to see - or build on - each other’s individual conversations with the AI) then you should create and share a GPT.
  A GPT lets everyone use the pre-prompted functionality that (…) helps less experienced users get better results, without sharing the chats with your team.

• If you want your team to COLLABORATE within shared conversations, and want to be able to see each other’s messages, (…) create a shared Project.
  Projects allow you to not only share the custom instructions and files with specific team members, but also actively see each other's chat threads. It also allows you to continue new chats from where other people left off.”

📝 9 GTM use cases with ChatGPT

This one dates a little (July), but includes good use cases, with their prompts, especially the advanced ones (I covered graders in a recent edition).

💡 It's time to clarify swim lanes when these lanes shift with AI

Don't let marketers, designers, webdevs, and product leaders get angry at each other because they can start encroaching on each other's turf:

“An underdiscussed phenomenon that is starting to happen in organizations: AI-driven role conflict.

The roles in charge of design, product management, coding, marketing, etc. in a project used to have relatively clear lines. AI lets entrepreneurial workers extend into other roles and accelerate work, but there are no good templates to follow. Everyone can now code a little, design a little, market a little - what does that mean?

The result so far has been a mix of acceleration and retrenchment. Some cases I hear about, this gets bogged down in defense of existing roles ("how dare the marketer show me a product prototype when they can't even code and are just using AI") and in others, organizations start to rethink old teams and experiment.” - Ethan Mollick

💡🛠️ What it takes to vibe code right

As you will read in Marcel's post, it's not just about using a vibecoding tool (e.g. Lovable, V0, Base44, Bubble, etc.) but understanding how to put together various tools to scope your app, plan your build, code, and iterate.

“A friend of mine came over that had a lot of knowledge of Loan Officers. He told me he had an app idea.

I immediately pulled up my computer and said STOP, just talk again.

- Step 1: got a Notion doc and had him dictate it using Wispr Flow

- Step 2: used a shaping doc prompt and started to come up with a plan

- Step 3: kickoff a ton of deep researchers to plan out the knowledge and data

- Step 4: took a high level plan and asked Lovable to build initial screen

- Step 5: started to build out the views, data model, and functionality bit by bit

- Step 6: bought domain and built landing page

- Step 7: built advanced features with AI (simple prompts in Lovable that used edge functions, just gave OpenAI API key)

- Step 8: Created a quick logo in Figma and polished SEO stuff

Used about $100 of credits and used tools that cost me about $200/mo in max subscriptions.” - Marcel Santili (more intel, including a video, in this post)

💡 LLMs love FAQs

From Josh Grant, VP of Growth at Webflow:

“We just ran an experiment adding FAQ content across Webflow’s /templates/ subdirectory and the results were surprisingly strong for something so old school.

Here’s what happened:

• 📈 +6% overall traffic growth to the /templates/ folder in just 3 weeks

• ⚙️ AI Search visits up 2% PoP (early but promising signal)

• 🔍 +640 clicks from non-branded queries, while branded clicks actually declined

• 💬 Pages with FAQs: +3,303 clicks PoP

• ❌ Pages without FAQs: –968 clicks PoP

Even after removing outliers like free-website-templates, the lift held steady:

• ➡️ +8% growth on FAQ-treated pages

• ⬇️ –6% decline on non-FAQ pages” - Josh Grant

🖥️ And for fans of the Claude desktop app…

Final Words

Thanks for sharing these highlights with busy marketing execs around you.🙏 

Someone forwarded you this email? You can subscribe here.

François | LinkedIn 

I'm a CMO, advisor, and "CMO Wingman". Yes, that's a thing :-). Ask my clients: in this AI era, CMOs need a strategic proactive advisor more than ever. I’m a former CMO at Twilio, Augment Code, Apollo GraphQL, Decibel, Udacity, and Head of Marketing for LinkedIn Talent Solutions.