Posts

Some marketing teams are stuck in a documentation trap. They won’t build AI teammates because they think they need perfect internal knowledge first.

You can build custom versions of ChatGPT, Claude, Gemini, or Copilot trained for specific tasks. Upload your best practices, templates, and examples of good work, and the AI gets better at handling that task for your team.

But many teams never start building custom GPTs, projects, gems, or agents because of what they don’t have.

Some have nothing written down. Others have scattered or outdated playbooks. And those with established best practices assume it’s still the best way forward because it worked before.

You don’t need perfect internal knowledge to build effective AI teammates. You can use external research to fill gaps, validate what you have, or challenge assumptions you didn’t know you were making.

Deep research features in tools like ChatGPT, Claude, Gemini, and Perplexity gather industry benchmarks, best practices, and frameworks in minutes. You can use that research as knowledge for your custom AI teammate, then refine it based on your team’s situation.

I used this approach for my own digital twin. I asked the AI to research publicly available information about me, my work, and my frameworks. The research report became part of the knowledge base, giving my digital twin context about how others see my work and what’s being said about my ideas.

Below are examples of what marketing teams can research to build stronger AI teammates, whether starting from scratch or questioning what they already have.

Build your AI teammates by blending external research with your situation. Start with deep research, test it in practice, and let your team improve it over time. That becomes knowledge worth keeping.

See original post here

If you’re only using AI to get faster at old work, you could be training it to replace you unintentionally.

A global cybersecurity leader’s marketing team uses AI beyond improving productivity. They are reimagining work with AI to drive growth. See their case study, reimagined daily workflows, and early results in the comments.

See original post here

The average piece of marketing content gets passed around multiple times before it ships. All that time coordinating could have gone to creating.

I joined Brandon Redlinger on Stack & Scale to walk through exactly how I’ve fixed this. This one’s worth watching while you’re on the treadmill, drinking your morning coffee, or folding laundry because I do a live demo of chaining GPTs together. See link to the video in the comments.

You’ll see one GPT brainstorm content angles, pass it to another that builds the webinar structure, which feeds the email writer, which flows into social posts. All in one conversation.

Here’s what you’ll walk away with:

• ➡︎ How to build and chain specialized GPTs in one thread so each one gets full context instead of starting fresh every time

• ➡︎ Why small, focused GPTs beat one big GPT that tries to do everything (plus when you need both)

• ➡︎ The prompt frameworks that make AI show you its thinking before giving answers, so you stay in control

• ➡︎ Guardrails that keep GPTs on task and get them to flag what they don’t know instead of making stuff up

• ➡︎ How to write prompts that push AI past generic ideas into stuff that actually stands out

Thanks to Brandon for going deep on what actually works instead of surface AI tips. His focus on real implementation makes Stack & Scale required listening for anyone rebuilding how GTM teams operate.

The future of marketing is about better systems, not bigger teams. This conversation shows you how to build them.

See original post here

No AI here. Just Amazon boxes, glue, paint, fabric, and trinkets. And a lot of great memories from when my kids and I turned them into these costumes.

AI might replicate them one day, but it can’t copy us building them together or their faces lighting up on trick-or-treat night.

Which one’s your fave? I can rock the Jurassic Park Jeep too. 🤣 Photo in comments.

Have a safe and Happy Halloween! 🎃

P.S. Fun bonus: The army tank and astronaut with a jetpack even made national TV. Truly special moments that we’ll remember.

See original post here

Most companies buy AI tools and hope for transformation. A global leader in cybersecurity redesigned work itself.

In less than 6 months, their marketing team trained 75 “trailblazers” and built 57 custom AI teammates (from 211 created during experimentation) now embedded into daily workflows. Now they’re scaling it to more than 150 marketers.

The big difference is they didn’t bolt AI onto existing processes. They architected a layered “AI Network” where humans and AI collaborate by design.

The results speak for themselves. The SDR team is saving 1-3 hours per day per rep with 2-3x better open rates (15% to 40%). Marketing team members are getting back 5-7 hours per week for higher-impact work.

Four key insights that separated their success from companies that struggle:

• ➡︎ Partner with security from day one → They worked closely with their CISO to ensure every workflow was deployed safely. Security wasn’t a blocker; it was foundational.

• ➡︎ Build infrastructure, not just use cases → GPT Tracker, AI Navigator, AI Triage System. Simple tools that prevent chaos as you scale.

• ➡︎ Systems beat tools → As their VP of Marketing Operations and AI transformation champion put it: “We don’t just hand them an AI tool. We give them our decision tree, governance model, and proven workflows.”

• ➡︎ Measure transformation, not just efficiency → Their SDR operations leader: “We’re not just saving time. We’re doing work that would never get done otherwise. That’s the difference between efficiency and transformation.”

The company built 8 workflow layers (Foundation, Story/Campaign, Domain, Asset, QA, Human-in-the-Loop, Automation) with specialized AI teammates at each level—Audience GPT, Campaign GPT, Product GPT, SEO GPT, and more.

AI tools give you short-term help. Systems give you long-term advantage.

See the full breakdown in this week’s newsletter below.

For those who prefer audio, I’ve also created a 16-minute AI podcast version with two AI hosts (link in comments.)

If you found this helpful, share it with other GTM leaders navigating their AI journey.

See original post here