Goose Ads Remixer

Congrats on the launch. One thing I'm curious about, the Meta Ad Library shows you which ads are running and for how long, but not spend or performance. So how do you decide which ones are actually winning vs just live?

when Goose learns the pattern from a winning ad in my niche, where's the line between remixing the structure and just recreating that competitor's ad with my logo swapped in - is that boundary a manual review step or does the model handle it?

the nano banana / gpt-image-2 / opus checking-each-other's-work loop is the interesting bit - when the three disagree on whether an output passes the brand check, is there a tiebreaker or does it just regenerate until they converge?

I think if everybody is learning from the same winning ads then the meta feeds would really look similar over time.

the brand-drift question above is the UX risk, but there's a legal one underneath it too: "remixing patterns" is fine, but if the tool is looking at a specific competitor's ad to generate from, how do you make sure it's not echoing their actual copy phrasing rather than just the structure? a hook and a layout aren't protectable, but a distinctive line of ad copy can edge into IP territory if it comes out too close to verbatim

Just signed up for Goose Ads Remixer and tried it for my website ToolJunction. I’ve used a few AI ad creative tools before, but goose feels more polished and practical at this price point. Starting with ads that are already working in your niche, instead of generating from a blank prompt, makes a noticeable difference. And, the outputs felt much closer to something I could actually run, and the overall experience was smooth.The only issue I noticed was that it picked the wrong color palette for my website. Great work, team!

Most AI ad tools generate noise.This actually helps teams iterate on what's already working. Thats a much stronger approach. Congrats on the launch

Congrats! Big fan of the product

Congrats on your launch! Excited to try this out.

This seems especially useful for small teams that don't have dedicated designers or copywriters. The first 10 free ads make it easy to test before committing. Curious whether it supports generating multiple variations for A/B testing across Meta and Google Ads?

Love the concept of learning from proven patterns! Does Goose provide any performance benchmarking after the ads go live, so users can see how the AI-generated versions actually perform? A feedback loop where live ad performance retrains the model for that specific brand over time.

So I've been using ProductHunt for a couple of weeks now. Some products get a cursory glance, some I play with for a little while. This one properly sucked me in and I have really enjoyed using it. The product feels professional and fun - and it definitely delivers results. I suspect I will need to keep tweaking my settings and improving my prompts to generate output that is the right balance between the original source and my own brand. And a couple of times a "work in progress" was better than the final output (would be good to know if I could revert back to a "work in progress" stage of the process). But overall this looks great - and I will definitely revisit. My one comment... pricing was hard to find. I discovered it in the documentation, but it certainly took me a while! Which seems a shame because (starting at $29 per month) it feels very fair pricing.

I'd love a side-by-side explanation showing what changed from the inspiration ad to the generated version. Does that happen already?

Congrats on shipping. As someone running Meta campaigns every week, ad fatigue is probably my biggest problem right now. Goose seems like a promising solution.

I am a performance marketer. I like this product because I was doing all of this research manually. Great ads leave clues, and this feels like it's decoding those clues automatically.