I built a Claude marketing & sales OS. Here's how it works.

A simple, practical way for go-to-market teams to start using Claude

After a week off for a vacation with my family, I’m back at it and this time showing a very simple Claude operating system for marketing and sales teams. This is one of the exact solutions that I’ve used with actual customers to get them started with Claude.

I’ve been working with a couple small to mid-sized customers over the last few months to implement a simple operating system in Claude for their sales and marketing teams.

I feel I’m at a spot with it where I can share it a little more broadly to see if it can spark any ideas for my readers.

While there’s a lot of buzz about Claude, as well as almost weekly releases of new features (a bit overwhelming even for me), it is still a bit of a jump from standard prompting. Enough of a jump that I think some people are a little hesitant to explore further.

I want to bridge that gap!

This operating system I built gives people a manageable, approachable way to actually get started building an operation in Claude for sales and marketing.

These four objectives were the pillars of what I was trying to accomplish with this:

SIMPLICITY

There is a ton of power in Claude Cowork, and I wanted a simple first step and way to get started for people that just don’t have the time or maybe even the interest in exploring all there is within the Claude ecosystem.

SCALABLE

My intention was to set up something that is a solid base of operations that, once customers are comfortable with the system, can easily begin to scale.

CUSTOMIZABLE

This is not a one-size-fits-all Claude operating system or a bunch of standard downloaded Skills. From the minute it starts up, it’s beginning to fully customize towards your environment, goals, and objectives. The system is built to adapt and create based on the needs of the user.

AFFORDABLE

I built this in a way that is an affordable entry point into using and exploring Claude from a setup standpoint, ongoing support standpoint, and Claude fees standpoint.

COLLABORATIVE

I wanted to make this into something that is used across teams and can be maintained by the team. This whole system is easily shareable and set up across a team.

---

Glad you’re here and thank you for reading Approachable AI by Pat Schaber! In case you missed some recent articles, here’s a recap:

AI on the resume means nothing. Here's what does.

I put Claude in PowerPoint against Gamma. Here's who won.

---

The Why

Before we jump into what it is and I actually show you how it works, I want to describe a little bit more about the “why” behind it.

What I see right now is that people are individually using AI and getting individual results within the same team. Take the marketing person that wants to post on social. They create all their social assets and post. The next day they do the same thing but get a different voice and tone. Then the person next to them does it and gets a totally new set of results.

Then there is the salesperson that wants to do call prep with AI. Their prep and positioning is going to look a little bit different than the person that sits next to them because Claude doesn’t know the brand. It doesn’t know the positioning. It doesn’t know how you stack up against competitors. It doesn’t know your value proposition, so it’s only natural they’re going to get different results from their AI tool.

It’s not the fault of the AI or the fault of the person. It’s just that there’s not a system in place for everything and everyone to work cohesively. This operating system solves that.

If time savings and efficiency are your goals with AI, this can be designed for that. If increased output and scale are the goals, it can be designed for that as well.

The OS System

Essentially, it’s a pre-built file system that you put on your computer and open in Claude Cowork. It looks as simple as this:

While this looks simple, there is a “brain” component in here that runs the operation. The CLAUDE.md file in the folder is the trigger that makes everything start and work.

Once the folder structure is set up on your computer, you simply open Claude Cowork and select the folder that has the files in it.

Then you simply type a command like “Let’s get started” and the CLAUDE.md file will take over. Here is the first thing customers see. This is essentially what I refer to as the “onboarding” process or where the foundation gets built.

The Foundation

The onboarding process will walk you through the creation of three main components of your foundation. Many teams have these elements to put in already so the process is quick. If you’re starting from scratch it will still take less than an hour.

Brand

It asks about your company’s personality, voice, and tone. Not in a fill-in-the-blank way. It’s built for a real back-and-forth dialogue so it’s customizing for your team. How do you want to sound? What words do you avoid? Who’s your primary reader and what do they care about? It asks about colors and logo too, so visual work stays on-brand.

Market Analysis

It asks who your top competitors are and gets specific about each one including what they do well, where they fall short, why customers pick you over them. It asks where the biggest opening in your market is. This becomes the competitive context that lives in every piece of content and every sales brief going forward. If you’re unsure about some of these details, the system will help you.

Product Marketing

This is the most important section, in my opinion. It asks what your product actually does (in plain English, not pitch language), what problem it solves, and who your ideal customer is which includes both the company and the person inside the company who buys. It asks why your best customers chose you, and what they say when they refer you. When this section’s done, you have a positioning file grounded in how customers actually talk, not how your marketing team wishes they would.

The whole onboarding takes about 20–30 minutes. At the end, you have three files that represent your brand, your market, and your product written in your voice, specific to your business.

If you want additional setup information, that can be done as well.

(Beacon is a fictional company I made up for the purpose of this demonstration)

Right now, most small GTM teams have everyone using AI independently. The VP of Marketing has her prompts. The SDR has his. The content person has theirs. Everyone’s briefing Claude or ChatGPT from scratch, every time, with whatever context they happen to include. The outputs are all over the place including different tone, different messaging, and different assumptions about who the customer is.

The foundation files fix this. When every person on your team is working from the same brand.md, market-analysis.md, and product-marketing.md, they’re all working from the same context. The outputs are consistent. The messaging is aligned. And when something changes such as new positioning, a new competitor, an updated ICP, you update one file and the whole system reflects it. You don’t re-brief anyone.

Your New AI Assistants

This is where the fun stuff starts to happen!

Once we have the foundation built, we can now move on to creating your new AI Assistants. In the folder structure, I call these Skills. Skills are a main component to Claude and Claude Cowork and how it operates.

To get started, I’ve already built out two skills that are often asked for:

1. A Content Creator which helps build blog posts, social media posts, etc.

2. A Prospect Research skill that helps salespeople research prospects before reaching out.

So the next time you go in and open up Cowork, select your folder with your files, the CLAUDE.md file or “The Brain” will take over. It reviews all of your foundational documents and skills and lets you know what we can get started working on. In this case, it’s the two skills that I’ve already created.

When I work with customers on this particular setup, we’ll take that Content Creator and the Prospect Research assistant, and we will customize those skills towards the customer’s operations. But they’re definitely good enough to start even without further customization.

Then we’ll work on building out all the other assistants the teams need and simply plug those skills into that folder, and you’re starting to build a working team of AI assistants. They all work together and they all work from the same context. A few examples of what is possible:

• Marketing

  • Content Creator

  • Content Repurposing

  • Paid Ad Planner

  • Campaign Planner

  • Competitive Briefs

  • SEO/GEO Audit

  • Market research assistant

• Sales

  • Prospect research assistant

  • Outreach assistant

  • Call prep

  • Call review coach

  • Prospect list creation

The idea behind all of these assistants is not to replace people on your team, but it’s to give them the assistance they need to be more efficient and free up time for other things the team and company needs.

I used my Beacon company example below and gave it that super simple prompt. Remember, I don’t need to add in context or what it needs to be or what the output is. This has already been trained to do that.

It knows what market I’m in and what topics might be good. You can see it went ahead and suggested topics. The output was a great blog post written in Beacon’s tone and voice and aimed at their target persona.

Now, the person that writes blog posts can take that, edit it, and make it even more about the company’s voice and positioning.

Conclusion

There are way more sophisticated operating systems that can be built with Claude Cowork and Claude code, but sometimes just starting with something simple like this can get the ball rolling for most teams.

Imagine what this could look like after a few months of using it, where additional skills have been built out, working together and sometimes working at the same time to make the marketing and sales teams even more efficient in their operation.

Hopefully this gives you some ideas, thoughts, and inspiration on how to build your own operating system with Claude!

---

---

Ways to Work with Me

Quick Win Session
A focused, high-impact working session designed to identify one immediate AI opportunity and give you a practical roadmap you can implement right away.

AI Readiness Assessment
A structured evaluation of your people, processes, data, and tools to determine where AI will create the most value and what needs to happen before you scale.

Workshops & Projects
Hands-on, team-based engagements that turn AI from theory into execution through tailored workshops, use-case design, and guided implementation.

Fractional AI & Marketing Leadership
Ongoing executive-level support that embeds AI strategy, operational rigor, and revenue-focused marketing leadership directly into your business.

---

AI Tech I’m Currently Using

I get quite a few questions on Substack, LinkedIn, and through my website on what AI tools I use for what use cases. I’ll try to share a few in each newsletter so it will give you some ideas of tools you can try for specific purposes. Here are a few for this week:

Everyday LLM - ChatGPT and Claude: I cover a lot of this above so won’t dive in too far here.

Call and Meeting Transcriptions - Granola: I like Granola for a couple of reasons. It takes the transcription and rolls it up into very good notes and action items but doesn’t need to be added to a meeting as an attendee. Also, it syncs great with my Hubspot CRM which makes it very easy to send notes to the contact record.

Presentations / Branded Documents - Gamma.app: I use this 3-4 times a week for anything I do with client presentations or professional document creation. I cover why I’m a heavy Gamma user in this post:

The 2026 Playbook for Managing AI Operations

Video Recording / Editing - Descript: I’m trying to work in time to my weekly routine for video creation. I feel it’s important to give people a more personal connection. Descript is a huge timesaver. Imagine being able to erase or add text from a script and have it be reflected in the actual video in seconds. Unbelievable.

Sales Prospecting - Apollo.io and Clay: I don’t do a ton of outbound prospecting for my business but I’ve been using the free versions of Apollo.io and Clay to find companies that I could get to know through LinkedIn that may be interested in my services. I’m only touching on a sliver of the capabilities of these tools.

---