Hi there,

Recently, I’ve been heads down in Claude Code. I wanted to share a practical workflow you could implement in your business today.

Let’s dive in!

Today, in 10 minutes or less, you’ll learn:

• 💎 Why drowning in client feedback is actually a goldmine waiting to be unlocked

• ⚡️ Why Claude Code beats regular Claude Projects for serious customer development

• ⏱️ How a non-technical person can build a client research workflow in 15 minutes

SPONSORED

🤩 Join our Portfolio Career Community

Ready to grow your portfolio career with like-minded peers?

If you have 5+ years of working experience, join our invite-only community.

🚀 Why join the community: Accelerate your learning, grow your revenue, and build genuine relationships with an exclusive group of portfolio professionals

🎯 What you get:

• Lifetime access to Portfolio Path Community

• Expert fireside chats, peer workshops, & community calls

• 30-day challenges and group accountability

• Shared resource library (AI workflows, offer templates, etc.)

• Full access to PT Consulting Launchpad lessons + workbook

• Full access to Consulting Offer Blueprint

Apply now - we'll notify accepted applicants within the next couple weeks:

To get your brand in front of 7k+ professionals, founders, and consultants, fill this out and let's chat.

💡 How I built a client research analyst in 15 minutes (using Claude Code)

You’ve gotten your first clients. You’re finally scaling up.

But now you’re drowning in service delivery. Overwhelmed.

You absolutely don’t have time for customer research, when you barely had time to do your own laundry yesterday.

However, you just know your 14 recent client calls must contain nuggets of insight that can help level up your service quality or grow your business faster.

So how do you unlock these insights without overburdening yourself?

Previously, I showed you how to use AI prompts for customer research.

An obvious solution: Dumping your client meeting notes into a Claude Project, then running an effective prompt over them. I still do this a lot. It’s fantastic.

But sometimes it’s not enough.

In these cases, I bring out the heavy artillery - Claude Code.

While it looks intimidating at first, I’ve found it extremely powerful even as a non-technical person.

Today, I’m going to show you how I leverage Claude Code to accelerate my client research workflow in 15 minutes.

1/ Why Claude Code?

I still use Claude Projects a lot, but have run into several pains:

• Too much copy-pasting between Claude + docs

• Repetitively uploading files into projects

• Difficult to move context across chats

• Losing track of my prompts

• Inability to update docs

Not to mention - that feeling of horror when I see my Claude Desktop app freezing yet again.

Claude Code emerged onto the scene this spring.

At first I was skeptical. But then I tried it after hearing so much hoopla about it - and was blown away. Using it was the first time I felt a visceral “agentic AI” moment.

Now I’m leveraging the tool not only for coding, but also business workflows.

Hopefully this client research walkthrough can show you how Claude Code can help you too.

2/ How to setup Claude Code as a non-technical user

First things first. To get Claude Code running, I whipped up the ol’ Terminal:

Next, I downloaded Claude Code by entering this command into terminal (Anthropic’s full instructions). Here it is for Mac:

curl -fsSL claude.ai/install.sh | bash

And for Windows:

irm https://claude.ai/install.ps1 | iex

Then, I created a “customer_research” folder in my Finder and navigated to it using the command “cd customer_research” (quick instructions).

Once I was there, I typed “claude” in the terminal to open Claude Code:

Boom. You’re in.

Now for the exciting part - building your own analyst.

3/ Building a client research analyst

Now that we’re in Claude Code, I’m going to show you how I setup my research analyst.

I created a sub-folder meeting_notes and uploaded all my meeting notes, transcripts, and documents as markdown files.

Afterward, to avoid having to copy-paste prompts, I created a Claude Slash Command. To do this, I created two prompt files in the folder .claude/commands:

• topic_analysis.md: Analyzing customer feedback and bucketing them into 4 topics: pain points, blockers, customer requests, and solution feedback.

• insight_generation.md: taking the output and grouping similar snippets into common insights with actionable recommendations.

This basically saved my prompts as markdown files in a folder, which is pretty convenient. No more losing track!

Next, I prompted Claude Code to run the topic analysis and save the output in a markdown file called topic_analysis_output.md

Then, I fed this output into the insight generation prompt to created the final synthesized insights.

Voila! I got a research summary of client insights, supporting quotes, actionable recommendations grouped by the topic.

The actual research output was a great starting point:

I’ve been impressed by how Claude Code manages context more smoothly than Claude Projects, producing higher-quality outputs IMO.

In particular, these insights helped me answer questions like:

• What are the common pain points, requests, and feedback?

• What are the actual quotes clients have said related to this insight?

• What actions should I take in response to this feedback?

Separately, I’ve also built a web dashboard to visualize this data with charts and a structured UI. (Not covering this today, but so much easier to create than before with Claude Code.)

4/ Wrapping up

From here onwards, I can see tons of possibilities.

I can use this research output to do things like:

• Craft ideal client profile descriptions

• Create a dashboard UI (like above)

• Generate sales landing page copy

• Build the actual landing page

• Create a discovery call script

• Write sales emails copy

All within Claude Code. And that’s just the tip of the iceberg.

So would I recommend Claude Code?

It might seem intimidating as a non-dev, but I’d recommend trying it out if you’re feeling inspired and not afraid to get your hands dirty.

And if you do, building your own customer research analyst could be the perfect weekend project. Let me know how it goes.

💎 Last Week’s Gems

🤯 Shorting the AI jobs apocalypse (David Deming)

🧠 On the Reverse Flynn Effect (Cal Newport)

💵 How to Keep This Hot Stock Market From Melting Your Retirement Dreams (WSJ)