Start with one task you already hate doing. Not the most important thing in your business — the most annoying thing. The thing that eats 45 minutes every week and makes you feel like a robot.

For most solopreneurs, that’s one of three things: writing follow-up emails after calls, repurposing content across platforms, or drafting proposals that say roughly the same thing every time.

Pick one. Open ChatGPT or Claude. Give it context about your business and what you need. See what comes back. Edit it. Send it.

That’s it. That’s where you start.

The mistake most people make is trying to “learn AI” as a subject. AI is not a subject. It’s a tool. You don’t need to understand how it works any more than you need to understand how your dishwasher works. You need to know what to put in it and what to expect when it’s done.

You need two tools. Maybe three. Not twelve.

One AI platform: Pick one general-purpose AI and get good at it. As of right now (2026) I recommend Claude. With the right foundation, it will handle writing, brainstorming, editing, summarizing, and thinking through strategy. Learn how to use it and it will cover 80% of what you need.

A transcription tool – Fathom is great if you’re ok with a bot on the call or you want a tool that can go to the meeting without you. If you’re on team “no bots on the call” go with Granola. If you don’t mind an extra step you can also go with the built-in transcription from Zoom or Google Meet. These tools turn your calls into searchable text that AI can then summarize, extract action items from, or turn into follow-up emails.

That’s probably enough for the first 90 days. Most solopreneurs add tools before they’re clear on the business case. These two will get you started and give you perspective on next steps.

Instead of trying to “learn AI” I recommend that you find a good use case. For most solopreneurs and small businesses, creating marketing content is a great place to start.

Think of something you need to write this week – could be a follow up email, a LinkedIn post, an RFP….

Go to Claude and give it some background. Who the audience is, who you are and what you’re trying to accomplish.

Will you have to edit it? Yes. Will it get you there faster than staring at a blank page? Yes. Fix it and send it.

Do that three times and you’ll start to get a feel for what AI can do for you and what it can’t. And you’ll have a better idea about what you need to learn in order to help it help you (HT: Jerry McGuire).

AI fluency isn’t about understanding the technology. It’s about learning what to ask for, how to evaluate the output, and when to override it.

This is all about experiential learning. The real time investment isn’t really “learning AI” from a tech perspective. It’s really around integrating it into your tool flow and the figuring out how to incrementally improve the way you work together.  

Incorporate your experience and expertise. That’s the whole answer.

If your prompt is “write something about marketing,” you’ll get the combined average of what the internet thinks the answer is.

But if your prompt includes your specific framework, a real client example, and the exact audience you’re writing for, and your personal opinion the output will sound nothing like what anyone else is producing.

AI is good at structure and formatting. But new ideas based on years of lived experience? Not really AI’s thing.

I touch everything I publish. I give it the ideas, the frameworks, the stories. Those are mine. Then I let it give it structure and flesh it out. And then I take it back and make sure it sounds like me. 

Think of it like 10 | 80 |10.  The AI does 80% of the work – you put in 10% on the front end and 10% on the back end.

End of the day – your clients hire you because you have experience and opinions. That’s gotta stay all you.

This is the wrong question — but it’s the one everyone is asking, so let’s get into it.

AI is free. You time is not. If you’re competing on the same playing field as a free tool, your fees were already hard to justify before AI showed up. The issue isn’t AI. This issue is the way you’re positioning your value.

You need to focus on outcomes – what changes as a result of your work – rather than on deliverables – the work product.

Deliverables can be commoditized. Outcomes not so much.

A client can get ChatGPT or Claude to write a marketing plan. They cannot get it to tell them which parts of that plan are wrong for their specific market, their budget, and their team. AI can’t sit across from their board and defend a recommendation. 

The fee justification is: “You’re not paying me to produce something. You’re paying me to know what to produce, what to ignore, and what it means.” 

Start with your last three proposals. Feed them into Claude or ChatGPT along with context about the client, the problem, and what you recommended. Ask it to identify patterns: What do you always include? Where do you tend to be vague? What sections take the most time?

That audit alone will save you hours, because you’ll see where you’re reinventing the wheel every time and what you can reuse.

Then build a proposal template inside your AI tool. Give it your standard structure, your brand voice notes, and examples of your best work. When a new opportunity comes in, feed it the client details and let it generate a first draft.

You can’t expect AI to take you as far with proposal writing as it can for marketing content – but it should be able to get you 60-70% of the way there.

Don’t forget to also feed in your meeting transcripts that led up to the proposal. They will uncover hidden wants and objections that you can use that will speed up your time to close.

It’s not always one thing that saves 5 hours at once. Instead it’s often about handing off the repetitive items that give you those hours back over a week. Here are some examples:

• Meeting follow-ups: AI-generated recaps and action items from call transcripts. Saves 30–60 minutes per week.
• Content repurposing: Turn one blog post or video into social posts, email content, and a newsletter section. Saves 60–90 minutes per week. (Even more impact if you weren’t visible on social platforms before.)
• Email drafting: First drafts of outreach, follow-ups, and responses to common questions. Saves 30–45 minutes per week.
• Proposal drafting: Template-based first drafts with client-specific details filled in. Saves 45–60 minutes per week.
• Research and prep:  Summarize articles, competitor info, or industry updates before calls. Saves 1-2 hours per week.

None of these even require special tools. At most you need to do some one-time set up for your main AI platform and hello Friday afternoons off!

Don’t start with a workflow. Start with what you’re already creating.

Do you already create content? A blog, podcast, youtube or newsletter? The AI workflow is: take that one piece of content and use AI to turn it into everything else.

Let’s say you write a weekly or monthly newsletter for your list. Use AI to turn that into a blog post, three LinkedIn posts and a carousel. One becomes five and all of a sudden you have marketing content 5 days a week.

But TBH creating more content is not going to be useful if you’re not crystal clear on what problem you solve, who you solve it for and why it matters.