Muze AI

@muskaan_m Congrats on the launch. I’ve overseen a lot of paid ads, so my honest first reaction was: how is this even possible? That curiosity is what made me click 🙂

A couple of genuine questions I’d love to understand better:

1. Creative sourcing — In my experience this is the hardest part. The search space is huge and good creatives are very non-obvious. How do you discover, generate, or converge on strong creatives rather than just many variants?

2. Learning toward high ROAS — Finding winners is usually slow and painful. What kinds of products have you tested this on so far? And do you see it consistently outperforming an above-average human optimizer, not just beginners?

Appreciate the ambition here — genuinely curious how far this can go.

@amitpsharma Great question- this is the #1 thing people want to understand.

Short answer: It can run solo, but most users don't start that way.

How it actually works:

1. Approval mode vs autonomous mode. When you start, you can run in "recommend only"- Muze suggests what to create, pause, scale, but YOU click approve. Once you trust it, you can flip to autonomous and let it run.

2. Creative direction comes from data + your brand kit. Muze doesn't wing it. It looks at what's working in your account, what competitors are running, what hooks are converting - and generates within YOUR brand parameters (logo, colors, fonts, tone).

3. You can always intervene. It's not a black box. You see everything it's doing and why. If you disagree with a recommendation, override it. Muze learns from that too.

The "too good to be true" concern:

Fair. Here's what it won't do:

• It won't magically fix a bad product or bad offer

• It won't outperform a world-class human creative team (yet)

What it WILL do:

• Execute faster than any human team

• Never sleep, never take weekends off

• Catch data anomalies humans miss

• Test more variations than you could afford to manually

Does that help clarify? Happy to answer follow-ups.

@juan_alvarez18 Let us know how it goes, will wait for your feedback! We are available to help you out if you need help in setting up and advice on ads.

@mykyta_semenov_ You're absolutely right - Google Analytics is powerful but basically unreadable for non-marketers. Even experienced marketers struggle to connect GA data to actual decisions.

Short answer: GA integration is on the roadmap. We already rn take all the data from Google TAG Manager.

What we're thinking:

• Pull GA data and correlate it with ad performance (not just clicks, but what happens AFTER the click)

• Plain-language alerts: "Traffic from Meta Campaign X has 40% lower time-on-site than Campaign Y - creative might be attracting wrong audience"

• Connect the dots between ad spend → site behavior → actual conversions

That is exactly what you described: AI that watches Analytics, understands what matters, and tells you in simple terms what's working and what's not.

Thanks for the suggestion - this bumps it up our priority list. 🇺🇦💪

@superbbob The optimisation is done on basis of the result the ads is giving. (we Analyse all the ad data coming from Meta for that ad, to make decisions). You can expect 30% boast in initial days. because it will fix all the things which are losing money. Later- winning ads generate more more revenue. Lets catchup and talk about it? https://calendly.com/muzecmo/30min

@taarak_gada Yes, giving back power and control in hands of founders. Thanks!