What’s in this guide?
Five ways to use AI that you can start this week. Real examples, specific tools, clear next steps.
If you already have a list of things that have been pushed off in your business, this guide is the next step.
If you haven’t read the first guide yet, start there. It explains why things get pushed off and how AI takes the weight off them.
Before we start
Not every task works the same way
Before you bring AI into any process, it helps to know what type of task you’re dealing with. That decides how you approach it.
Built once, used constantly. A tool, a process, a document. The work is up front.
Happens every week or every month. Content, follow-up, summaries. AI slides into the process and takes the load off.
Writing, phrasing, breaking down ideas. AI is a thinking partner, not a replacement.
Most of the things that get pushed off are recurring, which is where AI makes the biggest difference.
Chapter 1 · Content from meetings
Every meeting is worth four deliverables
There’s real substance in every business meeting. Insights I share, things I say in the moment, things the client works out for themselves.
The problem: you don’t have the energy to sit down afterwards and pull them out.
The solution is a process that runs itself:
- Timely transcription tool joins the meeting, records video, and produces a full transcript. $19/month annual, $29/month monthly.
- The transcript goes into Claude Code.
- Claude analyzes your style, picks up the topics, and produces a social post, a newsletter piece, a video script, and a summary for the client.
Four deliverables out of one meeting. Without writing a word.
Not magic. A process you build once and then it runs on its own.
Do this: pick one meeting that already happened, run the process on it, and see what comes out. The full setup for Timely and Claude Code is coming in the install guide.
Chapter 2 · Clients and follow-through
Two things that always get pushed off, that don’t have to anymore
Client interviews that turn into content
The strongest posts aren’t the ones you write about yourself.
They’re the ones that tell a client’s story. Where they came from, what you did together, where they got to.
The problem: asking a client to write it usually gets you “It was great, you’re the best.” Nice, but useless.
What works: send a WhatsApp to a client you loved working with. Ask if they have a few minutes for a call. You want to hear their story.
The story has three parts:
- Before: What was hard? What was stuck? What had they tried that didn’t work? This is what the reader identifies with.
- During: What did you do together? What did they discover? What surprised them? This shows the value.
- After: What actually changed? What are they doing today they couldn’t before? This is what convinces.
Before the call, ask Claude: “Prepare interview questions that will pull out a client’s story in three parts, before, during, after, with an emphasis on specific moments and not generalizations.”
You get precise questions for every stage. Everything the client says turns into content.
Automated meeting summary for the client
After every client meeting, it’s worth sending a summary.
It builds trust, documents what was decided, and shows you’re professional.
The problem: there’s no time to write it. So it doesn’t get sent.
With transcription, that goes away. The meeting is transcribed automatically. As soon as it ends, Claude Code gets the transcript and produces a structured summary: what you talked about, what was decided, what the next steps are. You send it to the client within minutes.
The client gets a summary before they’ve even closed their laptop.
Fill out this short form and I’ll send you exactly how to build the AI agent team for your business. Fill out the form →
Prefer to talk directly? Book a 30-min call →
Do this: pick one of the two, not both. Build it until it works, then move to the second.
Chapter 3 · Research and scanning for businesses
Finding opportunities that are sitting there, and no one is looking
Some businesses grow on one thing: being in the right place at the right time.
But actually searching for those opportunities? That’s the work that never happens.
A coach for webinar hosts who turned AI into a lead machine
A coach who works with people who run webinars wanted to find new clients.
The problem: any cold outreach sounds like spam. Nobody wants to receive “I’d love to help you.”
Here’s the flow he built:
- Automatic scanner. Claude Code searches for publicized webinars on LinkedIn and event sites. Every day a fresh list comes in.
- AI attends the webinar. It signs up, joins, and transcribes the whole session automatically.
- Analysis and report. Claude analyzes the transcript, identifies strengths and weaknesses, and produces a short report.
- Personal email. Claude writes an email to the webinar host: “I sat in on your webinar. I watched the whole thing. Here are a few points that could sharpen it, and if you’d like, I’d be happy to spend half an hour with you going through them.”
The result: an outreach that feels personal, grounded in real work, not spam.
And it all runs on its own, every day, hands-free.
The same principle works for other things:
- Scanning conferences and events for speaking or attendance opportunities.
- Mapping complementary businesses for partnerships.
- Tracking what’s being published in your field.
Do this: what type of opportunity do you want to find? Write it down specifically, then let the scanner do the work.
Chapter 4 · Research and scanning for nonprofits
Finding grants without sinking hours into it
Nonprofits live on outside funding.
The reality: there are dozens of grants and foundations relevant to every nonprofit, and no one has time to hunt them down.
I sat with a nonprofit director who listed everything he wanted to do and couldn’t:
Search for open calls from foundations. Update his management system automatically. Map companies for partnerships.
A year ago, impossible without a big budget and a developer.
What you can build today:
- A scanner that searches weekly for open calls relevant to the nonprofit’s focus area.
- Claude filters by criteria: grant size, focus, submission deadline.
- You get a clean weekly list: what’s relevant, what the deadline is, what needs to be submitted.
- Claude helps draft the application itself, following each foundation’s guidelines.
We talked about it for an hour, broke it into steps, and now he can hand it to Claude to build himself.
No developer, no big budget.
Fill out this short form and I’ll send you exactly how to build the AI agent team for your business. Fill out the form →
Prefer to talk directly? Book a 30-min call →
Do this: write down the three types of grants that are most relevant for the nonprofit. That’s everything you need to start building the scanner.
Chapter 5 · Recurring processes
What happens every week, and still happens manually
Every business has processes that repeat. And still, they happen by hand, because there was never time to automate them.
A new client starting a program
Every time a new client fills out a contact form, the same thing happens manually.
You open a task, send a welcome email, put it in the calendar, remember to follow up.
With a connector like Make automation tool, all of that runs on its own:
- Client fills out the form.
- A task is automatically created in your project tool.
- A welcome email goes out with the first meeting details.
- A follow-up reminder is set for seven days out.
- Claude produces an initial client brief before the meeting.
Without touching a thing.
The same principle works for other things:
- Follow-up on proposals that haven’t gotten a reply.
- A reminder for a client who finished a program, one month later.
- Automatic invoice generation at the end of the month.
- Weekly team update on the status of every client.
The Make setup and how to connect it to your other tools is coming in the install guide.
Do this: write down the three things you do manually every week, and pick one to start with.
Chapter 6 · Infrastructure
Before AI can help, it needs to know you
Every example above worked because AI had material to work with.
When there’s no material, it guesses. And the results come out generic.
Infrastructure is what turns AI from a general tool into an assistant that knows your business.
Content infrastructure
Every meeting, every call, every transcript is raw material.
When you collect it in one place, AI has something to work with.
It’s not an archive. It’s a warehouse of ideas you draw from again and again.
Style infrastructure
AI can write in your voice, but only if it knows what your voice is.
You don’t need to write a formal style document. Just give it 5 to 10 posts you’ve already written and ask: “Analyze my style. How do I write, what’s the tone, what doesn’t show up, what repeats.” It’ll pull the style out on its own.
You build this once and then use it in every conversation. The output immediately sounds like you and not like AI.
Technical infrastructure
The places your information lives: transcripts, summaries, CRM, folders.
But it’s not just where to put things. It’s telling AI exactly how to work: first pull the transcript, then extract the key points, then write the summary in this format and not that one. AI doesn’t guess what you want. It does exactly what you explain to it.
It’s not complicated to set up. But you need to do it once, and do it right.
Fill out this short form and I’ll send you exactly how to build the AI agent team for your business. Fill out the form →
Prefer to talk directly? Book a 30-min call →
So what did we do?
- Content from meetings. Every meeting turns into four deliverables automatically.
- Clients and follow-through. Client interviews and meeting summaries, without the effort.
- Research for businesses. A scanner that finds opportunities without searching.
- Research for nonprofits. Grants and open calls you don’t miss.
- Automations. Recurring processes that run themselves.
- Infrastructure. Content, style, and technical setup, so AI actually knows your business.
So what now?
Go back to the list you made in the previous guide, AI for Business.
Pick one thing. Not the most impressive one, not the biggest one.
Pick the simplest one.
Something you can finish today. Not start. Finish.
Because one feeling of “done” is worth more than ten great plans.
Just one thing. Just today.
The first step: install the tools.
The full install guide, with every step, every click, every link, is coming soon.
Want to understand why this list matters in the first place?
AI for Business: How to Finally Do the Things You’ve Been Putting Off →
Fill out this short form and I’ll send you exactly how to build the AI agent team for your business. Fill out the form →
Prefer to talk directly? Book a 30-min call →