Google now has multiple ~free “studios”.

I spent 10 days testing them.

Turns out, the real problem isn’t the tools.

It’s starting at the wrong layer too early.

𝗦𝘁𝗮𝗿𝘁 𝘄𝗶𝘁𝗵 𝘁𝗵𝗲 𝗹𝗮𝘆𝗲𝗿 𝘁𝗵𝗮𝘁 𝗺𝗮𝘁𝗰𝗵𝗲𝘀 𝘆𝗼𝘂𝗿 𝗷𝗼𝗯:

1. Use Google Stitch when you want screens and flows in minutes, without waiting days for a design. I lets you explore the solution space with the team or run an ad-hoc prototype and record a quick Loom.

2. Use Google AI Studio when you want an app-like prototype that behaves, can be shared, and used in experiments. A great option to prototype AI features (native capabilities).

3. Use Firebase Studio to build production apps only if you’re ready for developer workflows like the terminal, dependencies, and ports. Cursor is great too, but both come with a steeper learning curve for PMs.

4. If you want to build something that feels like a real product without touching a terminal and can become much more, use Lovable or Dyad. Just keep in mind that engineering teams in large organizations won’t use them. They don’t fit traditional workflows.

𝗧𝗵𝗲𝗻 𝗯𝗿𝗶𝗻𝗴 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗶𝗻 𝘄𝗵𝗲𝗻 𝘁𝗵𝗲 𝗲𝘅𝗽𝗲𝗿𝗶𝗺𝗲𝗻𝘁 𝗲𝗮𝗿𝗻𝘀 𝗶𝘁 (𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻):

1. Vibe engineering (no terminal, lowest friction, fast iteration): Lovable, Dyad.

2. A team that can code (production foundations): Firebase Studio, Cursor.

Most teams don’t move slowly because of bad tools.

They move slowly because they start building before they’ve finished learning.

Hope that helps!

—

P.S. Want to learn more?

In my new post:

• Stitch vs. Google AI Studio vs. Firebase step-by-step

• Real-life workflows, mental models, and tradeoffs

• When exactly I switch to Lovable/Dyad

• When Firebase Studio can feel like a wall

• Figma vs. Stitch & how to stop waiting days to see the first flow

You can read it here: productcompass.pm/p/sti…

——

Want to break into AI PM?

I recommend the AI PM Certification by Miqdad Jaffer (Product Lead at OpenAI). I lead the AI Build Labs. We start on Jan 26, 2026.

$500 off: bit.ly/aipmcohort