Google has quietly added something exciting to the Merchant API. While the principal announcement was about transitioning from the Shopping Content API to the new Merchant API, tucked away in the documentation is a hidden feature: AI-powered text suggestions in Product Studio.
At the time of writing, this is only available through the API. In the Google Merchant Center UI, Product Studio still focuses purely on images. But if it’s already baked into the API, it feels almost inevitable that Google will eventually release it into the UI as well.
What Is the New Text Suggestions Feature?
The feature allows you to generate product titles, descriptions & more using AI. Instead of relying solely on images (as Product Studio currently does in the UI), the API can take product attributes, such as titles, descriptions, or even images, and output suggested improvements.
Here’s what it can do:
- Suggest optimized product titles and descriptions.
- Generate SEO-friendly titles tailored for search visibility.
- Format product attributes in consistent ways (e.g., brand, color, size).
- Even create attribute-specific text, such as materials, colors, or sizes.
In short, it’s positioned as a tool to streamline feed management, speed up copywriting, and boost customer engagement and conversions.
Sounds Good, But What’s the Catch?
As with anything AI-generated, the output isn’t perfect. In my tests, titles generated from existing attributes often look neat, like “Nike Men’s Air Zoom Pegasus 38 Running Shoes Blue Size 12.” Not bad at all, but still requiring tweaks, especially around regional sizing (EU vs US), relevance, or missing details.
On the other hand, when the API generates text purely from an image, the results are much less reliable. For example, an output might be “Rustic ceramic and leather leaves necklace.” That sounds descriptive, but it lacks the brand, target audience, and key attributes that matter for shopping ads.
Descriptions suffer the same issue. AI often creates what I call “fluff content”, generic, flowery text like:
“This is a beautiful necklace made from high quality ceramic and leather. It features a unique design that is sure to turn heads.”
Yes, it fills space, but it’s not very useful for the shopper. It doesn’t answer the essential questions: What size is it? What style is it? Who is it for?
That’s why I never recommend publishing AI-generated titles or descriptions as-is. They must always be manually reviewed and improved.
How We Use This at FeedArmy
At FeedArmy, we’ve been building semi-automated systems for years using Google Sheets and custom scripts. The idea is the same: use automation and AI to speed up repetitive work, then apply manual editing to ensure quality.
We don’t treat AI as a replacement for humans, it’s a tool. The Merchant API text suggestions work best when you:
- Combine all available data (title, description, attributes, customer search data).
- Use AI to generate a starting point.
- Manually refine outputs for accuracy, readability, and SEO.
For example, instead of just letting the AI output a size or material, we cross-check if it’s correct for the target market. You don’t want a US shopper to see EU sizing or a description that misses a key selling point.
Why This Matters for Retailers
If Google does bring this feature into the Merchant Center UI (and they will), it could save retailers a lot of time, especially those with extensive catalogs. Imagine being able to select a product in Product Studio and have Google suggest optimized titles or descriptions on the fly.
For now, it’s only accessible via the API, so developers and apps need to implement it. But when it lands in the UI, it could be a real game-changer for merchants who want to improve their feeds without deep technical knowledge quickly.
Key Takeaways
- Google is moving from Shopping Content API → Merchant API.
- Hidden in this update: Product Studio can generate text suggestions, not just images.
- Titles and descriptions are decent starting points, but they need manual editing.
- Generating from images alone is weak, combine it with existing attributes.
- Always check for accuracy, SEO relevance, and customer intent.
- Expect this feature to arrive in the Merchant Center UI in the near future.
Final Thoughts
This is a fascinating development. Google clearly wants to integrate generative AI more deeply into Merchant Center. But like with all automation, it’s only as good as the human oversight behind it. Used correctly, it can save time and unlock opportunities for optimization. Used blindly, it can create fluff, errors, and even policy issues.
At FeedArmy, we’ll continue to treat AI as an assistant, not a replacement. It’s excellent at speeding up specific processes, but the real value comes from combining AI with your own knowledge, customer insights, and manual refinement.
If you’d like to discuss how we can help you optimize your product data using both AI tools and proven manual processes, feel free to get in touch, or drop a comment with your questions.