Making Gen AI copilots your marketing superpower

The following is a conversation with Rebecca Schimmoeller – Demand Gen Content Strategist, formerly @ VMWare & Co-Founder of Think AI Collective.

No one gets up thinking, "I’m going to use Google today," they just use it.?

Ohh tell me more.

AI will soon be something we use without thinking twice. To some extent, we already do, since it’s integrated into much of our day-to-day tech. From a business perspective, this is the nut I want to crack for team leads and individual contributors. It’s not hard to get started and it’s such an enabler, on so many levels.

How do you recommend people use their time wisely to invest in AI tools, considering how busy they are?

Like anything, it’s really getting out of your own way. I just started creating my own custom GPT for a use-case I’ve had in mind for forever, for example. It’s faster to do things the way you know rather than learning something new, right? Even though, in the long run, you know it’ll save time and money.?

A lot of people suggest just starting with an AI use-case (which isn’t a bad idea), but I think a misstep is not also taking the time to understand the basics - how does it work?

My training is in content strategy, for example. I’ve been a content marketer since before it was a term. So for me, understanding AI is like getting to know your audience. It’s foundational and critical.

Taking a course at Oxford University last fall helped me understand the technology better. They balanced the geeky stuff with practical insights. However you do it, take the time to understand what AI can and can’t do, and you’ll get better results.

When do you start seeing compounding results from AI?

Million-dollar question, right? It depends on what you're measuring and what you want to achieve. And it varies on an individual level, as a team player, or if you’re managing multiple teams.?

If you’re asking what AI skills you need to use AI effectively, it’s about asking good questions and defining what ‘good’ looks like.?

Nuance is the biggest thing humans bring to the equation and the hardest for AI to understand. Anyone doing content marketing for over a decade, or even three to five years, understands nuance.?

As a writer, we don’t think about how long it takes to draft something or get feedback from a subject matter expert. It’s just a process you do instinctively. How long does it take to write a first paragraph versus the body copy that flows once you’ve figured out your thesis??

Time is money, so knowing your baseline time investments help benchmark, before engaging with AI. Think about all the steps in your day – how long it takes to email someone for feedback, read that feedback, and apply it. There’s AI for that, too… You can ask it to summarize feedback in simple terms, which helps you understand and apply it quickly.

100%. I was thinking about this a few months ago. I haven't abandoned older processes of getting feedback from someone I trust and then using the model with lots of context for iterative back-and-forth. It’s like querying the entire internet about an idea and then waiting for someone to reply. It’s a transition process of understanding the tech to enable it. Many people ask ChatGPT or Claude a generic question and get frustrated with a generic output.

It’s like, "Help me, help you." Pro tip: Tell your AI to act like it's had two cups of coffee before tackling a task. It hyper-focuses the output.

Really? I’m gonna try that.

It’s really funny – Claude is always “taking sip of coffee”.

Oh my God, Claude is crazy. I love it.

The other important thing is knowing which tool to use for what. ChatGPT is known for research and can write, but Claude is better at writing right out of the gate. Using both together is fascinating.?

I co-founded an organization called Think AI Collective with some folks from the Oxford AI Programme. Our mission is really to help unlock the potential of AI for senior leaders, showing them how AI can help them deliver and that it’s not a scary thing. We made it through cloud transformation. We'll make it through AI transformation too!

For that group, I was working on copy for an article using a ChatGPT prompt. It was a start, but it was clunky. So, I tried using Claude to tighten it up. The result was much better.

It's almost recursive. You can have one model check the work of another, strengthening and tightening it. When you multiply their capabilities, it’s not just additive; it’s multiplicative in terms of quality.

Exactly. You don’t have to operate within a tool silo. You can ask the tools to help create a better prompt. You can ask one tool to analyze feedback from another, and then incorporate that into the next draft. This way, you keep the ball moving without losing time waiting for feedback.?

AI has also been a great enabler for me as a thought partner while waiting for others to wake up on the east coast. It helps me maintain momentum when I’m hours ahead in my day.

It’s crazy to think about the leverage people will get with AI. With more tools, better models, better data, and more experienced humans steering them, the potential is immense. In two to five years, there will be a massive gap.

Absolutely. Critical thinkers who ask great questions will excel. Anyone not investing in these skills will be at a disadvantage. Spend time with experienced professionals to understand nuances they take for granted. Build up your own database of knowledge and learn to ask better questions. It’s about problem-solving and accelerating your ability to leverage the tool. AI is an enabler that complements your skills, but you need to appreciate it and invest in your own development.

What future plans do you have to build things with AI?

We talked about it a little bit…I want to work on custom GPTs for scalability. I have a lot of similar use cases, and it's nice to have templates for success. (I recommend starting a OneDrive document to track your prompts, by the way. Keep what works handy.) I'm used to working with designers and video production teams, so AI in the creative space is another fun use-case too, but I haven’t gotten that far yet.?

On my wish list is a dashboard that distills insights and analytics, showing how content performs and how to optimize it with specific recommendations. This would help refine assets and also identify what to abandon.

Are you talking about a core interface to access different models and integrate all your tools?

That would be very cool, but I'm talking about content performance. How is it doing in the market? Why is something underperforming, and what can I tweak to improve it? There are analytic AI tools out there, but a single interface would be ideal. The Martech stack in any business is complicated, and tools don’t always talk to each other. A single software interface, powered by AI, that brings it all together and offers optimization recommendations would be magical.

I'm sure that's on a lot of people's wish lists. Well Rebecca, it's been a pleasure. I really enjoyed getting to know you.

Right back at you.?

Remember, folks: Work with AI, not against it. Your future self will thank you.