Deep Market Researcher AI Agent for Product Managers

Things just got serious. This is probably the most specialized and fastest AI research agent for Product Managers.

Hey, Paweł here. Welcome to the free edition of The Product Compass Newsletter!

Every week, I share actionable insights and resources for PMs. Here’s what you might have recently missed:

• The Remote Product Manager's Playbook

• Product Manager Onboarding: Your First 30/60/90 Days

• AI Prototyping: The Ultimate Guide For Product Managers

• The Ultimate ChatGPT Prompts Library for Product Managers

• The Ultimate List of 147 Product Manager Interview Questions

Consider subscribing and upgrading your account for the full experience:

---

As promised, agents from our aigents.pm have become autonomous.

Today, I’m sharing a fully autonomous AI agent, Deep Market Researcher, that can help you perform deep research when working on product strategy, crafting a PRD, or preparing for a PM interview.

Deep Market Researcher

Soon, it will be plugged into other AI agents.

Important: A standard post will be published on Thursday. And from March 1, I’ll start working full-time for this newsletter.

---

Examples of Queries and Results

Here’s what you can use it for, with example results.

Example 1: Market Research for Amazon AWS

Research without a specific scope:

• Researched: 60 websites

• Time: ~30 sec

• Result: PDF 22 pages (Google Drive)

Example 2: Market Research for Nvidia

Focus on key insights, market share, and revenue over time. Includes competitive landscape, SWOT analysis, and market trends:

• Researched: 42 websites

• Time: ~30 sec

• Result: PDF 18 pages (Google Drive)

Example 3: Perform market research for Netflix

Focus on revenue, market share, competitors, and MAU in 2020-2024:

• Researched: 36 websites

• Time: ~20 sec

• Result: PDF 14 pages (Google Drive)

Example 4: Prepare for a PM Interview at Microsoft

Get key insights about the company, common interview questions, and answers:

• Researched: 46 websites

• Time: ~25 sec

• Result: PDF 19 pages (Google Drive)

---

How It Works

In the first version:

• First, the agent browses the web to get additional context about user research. For example, if you ask about Netflix, it will list competitors, key market and industry trends, etc.

• Next, it uses the user request and additional context to plan the work for other agents (up to 12 “researchers”).

• Each researcher gets a specific area with key research questions, like “MAU trends 2020-2025 for Netflix and competitors X, Y, Z” and starts browsing the web. All researchers work in parallel.

• Finally, an LLM combines all findings and creates the final report.

• Everything was created with Lovable. Here, you can learn more.

In the current version, the number of websites is usually limited to ~55. This is an artificial limitation designed to better control costs.

---

Can I Use It for Free?

Yes! I’m currently testing the platform to see what infrastructure costs it will generate.

So, in the early days:

• Everyone gets 2 deep research questions/day.

• Free and premium subscribers get 5 deep research questions/day.

• Limits will be adjusted after understanding the usage. They will be much higher for premium subscribers.

• Other agents, like the PRD Generator, Product Strategist, or PM Resume Reviewer, remain free (fair use policy applies).

• There is no dedicated paid tier. It’s a tool to support my subscribers best, not a standalone product.

---

Known Limitations

Several areas keep improving:

• It works for standard products like Amazon AWS, Microsoft 365, or X (Twitter). At the same time, it might struggle with something less typical, like market research for a VC fund. Your query should be more specific to avoid a generic approach in those cases.

• Occasionally, it provides general information about the methodology rather than answering the question. This is due to an artificial limit on the number of sites it's allowed to research.

It will only get better.

And soon, it will fuel our other agents to support you best.

I’d love to get your feedback: pawel@theproductcompass.tech 🙏

---

Thanks for Reading The Product Compass!

It’s great to explore, learn, and grow together.

A standard post will be published on Thursday.

Coming soon:

• [Premium] Fine-tuning and RAG for PMs

• [Premium] Step-by-step video of building a full stack app without coding, no cuts

• [Premium] Managing portfolio of products

Starting March 1, I’ll be working full-time on this newsletter! 😊

Have a fantastic weekend and a fabulous week ahead,

Paweł