Amazon Responsive Ecommerce Creative: How AI Automates Product Ads

Amazon has been pushing generative AI deeper into its advertising tools, but Responsive eCommerce Creative is more than another image generator. REC can take the products you choose, build responsive display ads in different sizes, pull live commerce information from Amazon, and continuously test combinations of images, headlines, product titles and layouts.

For sellers, that changes the job from designing every version of an advert to deciding what source material Amazon is allowed to work with. That can save production time, but it also makes product-detail accuracy and brand control more important.

What Amazon Responsive eCommerce Creative does

Responsive eCommerce Creative, or REC, is Amazon Ads' automated display format for products sold on Amazon. You can submit products and creative assets once, then Amazon adapts the advert for different placements across Amazon properties, devices and third-party sites.

The commerce section can pull live information such as price, star rating, Prime eligibility, deals and vouchers from the product listing. The visual section can use the default product image, your own lifestyle creative, video or AI-generated material.

Amazon's current specification says a single REC can include up to 20 ASINs. The system can rotate products, skip items that go out of stock and automatically stop bidding if every product in the creative is unavailable.

The important 2026 change is how much more of the creative itself can now be automated.

What happens when you enable AI enhancements

Amazon's "Enhance creatives with AI features" option allows the system to go beyond resizing an advert. It can select assets from your product detail pages and existing campaigns, create new backgrounds or compositions, modify headlines and product titles, and test different layouts.

Amazon says the newer REC system can also generate lifestyle images and product-in-use content. In the US, REC has expanded further into in-banner video and conversational prompts that send shoppers into an Alexa for Shopping experience. Static lifestyle imagery and AI enhancements are currently documented as available globally, including the UK, while REC video and prompts remain US-only.

This distinction matters. A UK seller can use AI-enhanced REC today, but should not assume every feature shown in a US announcement is available in their account.

Your product page is becoming part of the creative brief

When an AI system can source material from a product detail page, the listing itself becomes more than a destination after the click. It becomes input.

That means weak source content can spread further. If the product title is awkward, the lifestyle images are inconsistent, or an old graphic is still sitting in the asset library, automated creative has more poor material to choose from.

Before turning on AI enhancements, clean the inputs first. Make sure the ASIN has an accurate title, current product imagery, correct variation information and approved brand assets. Remove outdated campaign creatives you would not want resurfaced.

For jewellery, pay particular attention to the images Amazon may reuse. A bracelet clasp, engraving, chain thickness or metal finish that is unclear in the source image can become harder to inspect once the image has been adapted into a small display placement.

Automation does not remove the need for product accuracy

Amazon explicitly notes that when AI enhancement is enabled, advertisers may see asset and copy variations they did not create. That is the point of the feature, but it is also the main reason sellers should treat automation differently from a simple resize tool.

A generated seasonal background is usually lower risk if the product remains untouched. A generated product-in-use scene is higher risk because the model may need to infer how the item looks from a different angle or in a different context.

For a jewellery seller, check whether a necklace still has the correct pendant, whether a bracelet has gained or lost links, whether an engraving is readable and whether gold, silver or rose-gold finishes still resemble the actual SKU. For packaged goods, check labels, quantities and colours. For electronics, check ports, controls and proportions.

The useful rule is simple: automate context aggressively, but verify product identity conservatively.

Give Amazon better image options

REC can work from default product imagery, but sellers can upload custom images. Amazon currently recommends three responsive formats when you want more control: square at 1:1, tall at 9:16 and wide at 1.91:1.

Those custom images should be high quality and uncluttered. Amazon's current REC rules also say responsive custom images should not contain added text, logos, calls to action or white or transparent backgrounds. Branding and headlines can be handled by the ad components instead.

This is a good place for an AI product-imaging workflow to help. A seller can create several accurate lifestyle scenes from an approved product image, check them manually, and then provide the strongest versions to Amazon rather than letting the ad system invent every visual from scratch.

What Amazon's performance numbers do and do not tell you

Amazon says REC ads with AI optimisation enabled saw an average 14% lift in click-through rate in an internal 2025 study. In a separate June 2026 announcement, Amazon reported up to 37% higher CTR and 26% higher conversion for REC when adapting creatives to context, shopper intent and campaign objectives, based on a US study.

Those figures are useful signals, but they are Amazon's own studies. They do not mean an individual seller should expect the same lift, and they do not tell you whether AI generation itself caused all of the improvement.

The practical reason to test REC is simpler: it can create and optimise many placement-specific versions that a small seller would probably never produce manually.

A sensible workflow for sellers

Start with one product group that already has strong imagery and a clean detail page. Review the assets Amazon can source, then upload a small set of approved lifestyle images in the recommended aspect ratios if brand or product fidelity matters.

Enable AI enhancements only after that source material is ready. During the first campaigns, inspect the ads you can preview and watch performance rather than assuming more automation is automatically better.

If you sell products where tiny details matter, keep your real product photography as the reference. Tools such as Lustra Studio can help create additional product-led scenes and variations, but the strongest workflow is still to approve the product image before it enters an automated advertising system.

The takeaway

Amazon's newer Responsive eCommerce Creative turns a product listing into something closer to a living creative system. It can resize, compose, select, generate and test ad elements continuously while keeping live commerce information such as pricing and promotions up to date.

That is valuable for small sellers who do not have a design team creating dozens of display formats. The trade-off is that you are giving the platform more creative freedom.

Use that freedom where it saves repetitive work. Keep tighter control where the customer needs to see exactly what they are buying. In ecommerce, the best automated advert is not the one that looks most imaginative. It is the one that scales without quietly changing the product.

Sources: Amazon Ads, “The new creative advantage: Tools that easily scale, optimize, and drive creative performance,” 22 June 2026; Amazon Ads, “Responsive e-commerce creatives” specifications, accessed 26 August 2026.