Hey good folks and happy Thursday! We’re in the last days of July––and this Substack post captures my feelings perfectly.

I hope you’re enjoying it. Now, let’s dive in!

I love Emily Kramer’s MKT1 newsletter. It is packed full of helpful insight and tactical tips to put AI to use in your role, right now. And this week’s newsletter headline was really validating for me: “Marketing teams are stuck in single-player Claude mode. Here's how to go multiplayer.”

This is my #1 complaint about Claude, and something I’m sure Anthropic and OpenAI and all of the others will soon fix. But it is this: there is significant knowledge sharing required to ensure that everything you create and build can be used and built upon by your team.

And significant knowledge sharing is difficult. I’d argue it’s one of the bigger benefits of working in-person. Knowledge sharing happens in passing more easily when in person, and info that you might not otherwise share gets shared…and that info turns out to be really helpful for the other person.

In a remote environment, you have to rely on Slack channels, documentation, and things that humans just aren’t always really great at maintaining. We’re social creatures! We talk, a lot. We write, a lot. But we are not very good at maintaining a central source of regularly updated knowledge-sharing.

Here’s an example of what I mean in practice:

• Say I create several Claude skills that I know to chain in specific prompts to produce the outcome I want.

• I test the process, and think it works great.

• I share those skills with my org.

• I create a document that outlines the use cases, how to call the skills, in which order (and typically, why, so that the person builds a muscle memory around this), what information to input into the prompt, and what to do with what it returns.

• I share this out with the team.

Great!

But then, our style guide changes, and we add in several thought leadership pillars we’ve had as an organization before.

OK, so:

• I update the style guide skill.

• I add in the thought leadership skill (or better yet, could I add it to something existing?)

• I test the process, and am generally happy.

• I update the document, and reshare that with my organization.

• I share it over Slack as well as email.

But, keep seeing the same error style error in blogs that come from this one specific person.

Well, turns out––that person cloned the style skill so they could edit it, and were never using the company one. Now, we need to update their process, and explain why they should not do that. We need to add this information to the document, and give folks a path to documented resolution when they don’t like something in a skill.

Also, I keep getting blogs that don’t map back to our thought leadership pillars.

Yikes. Turns out, a lot of folks didn’t go back and read the new documentation. They assumed the update was simple, or didn’t have time to go and look at it, and never knew to call that skill into the chain.

Now, I need to reeducate, and better understand how to share this information, because even when change is documented, humans are creatures of habit.

This is why I built an AI content production tool on top of Claude. It solves for these issues, removing the need for prompting, and chaining skills for everyone, with the right skill, at the right time, so that outputs are the 60-70% fidelity we need to call AI content production a success.

Claude, and most of the LLMs today, are single player tools right now. For them to really scale and work well for the enterprise, they need to be multiplayer tools, with account roles and hierarchies, skill management, product management system integration, CMS integration, and so much more.

Claude and the other LLMs are powerful, but their best days are ahead of them, and it won’t be a new AI model itself. No, it’ll be a relaunch of LLMs for business teams, not just individual contributors.

A NOTE ON THIS ADVICE

You do you!

One content marketer’s best practices aren’t always right for another one, though I do try to distill out the main concepts and core practices I believe everyone can benefit from. That said, you must use good judgment when deciding whether to take advice given from folks on the internet. I am an expert, and this advice comes from my direct experience, but I am not smarter than you, and I have nothing to gain or lose because of what you do.

THINGS KEEPING ME CONTENT

• What I’m watching: I just finished The Perfect Couple. I realize that I found that sooooo late. But it was sooo good!

• What I’m buying: Last night, I went on a friend date (we gotta do this more!) and got a green negroni. It was wonderful!

• What I’m eating: Wagyu dumplings and so much more from Uchiba!

• What I’m reading: WHOA, late addition but this essay is awesome.

Thank you so much for reading. Let me know what you think by replying to this email.

Very excited to be here with y’all. 

Tracey