Sponsored Brands Custom Image: What It Is And Whether You Still Need One

What a custom image is in Sponsored Brands

Sponsored Brands is Amazon's ad unit for showing up as a video, a headline, or a group of products at the top of search results and on product pages. A custom image is the creative you supply yourself for that unit — a lifestyle photo, a styled product shot, a brand scene — instead of letting Amazon pull a plain product photo straight from your listing.

It matters most in the Product Collection format, where the ad shows several products under one headline. Until recently, that format required you to supply your own lifestyle image before it would run at all. If you didn't have the asset, you couldn't run the format.

That changed. Amazon added a wider Collections format — up to 10 products instead of 3 — and made the custom image optional. Skip it, and Amazon fills the space with catalog shots pulled automatically from your listings. Upload one, and it still runs alongside the products.

The rule that just changed, and what it actually does

Before the update, custom lifestyle imagery worked as a filter. Brands with a photographer, a designer, or an agency producing creative could run Product Collection ads; brands that didn't, generally couldn't. That gap was a real edge for anyone willing to invest in the asset.

Amazon's update removes that filter. Custom images are now optional on the new 10-product Collections format, and Amazon's own guidance is that any existing collection campaign should carry at least 3 products before making the switch — fewer than that, and the format underperforms.

The advantage of having good creative doesn't disappear. It just stops being the thing that decides who can enter the format at all. Every competitor who skipped Collections because they lacked the imagery can now run it with auto-pulled shots. The competition moves from who has creative to who tests creative properly, and who has the catalog depth to fill 10 slots with products worth showing.

Does a custom image still outperform the default — a worked example

The clearest test we can point to comes from Walmart Connect, not Amazon. Walmart runs its own Sponsored Brands product under the same name, with its own custom image capability, and it's a separate platform with separate specs. Treat the numbers as directional evidence about lifestyle imagery generally, not an Amazon-specific guarantee.

In that case, a food brand working with an agency partner ran a controlled test: identical keyword and item targeting, the only variable was creative. Logo-only ads versus lifestyle imagery showing the product in use. The lifestyle version delivered a 176% increase in click-through rate, a 49% increase in conversion rate, and a 150% increase in ROAS over the brand's prior logo-only campaigns.

That's the shape of the answer, even on Amazon: a generic or logo-only image usually loses to a well-made lifestyle shot, when everything else about the campaign is held constant. What changed isn't whether custom imagery helps — it's whether Amazon requires it before you're allowed to find out.

AI image generation as the middle option

Amazon's Sponsored Brands builder now includes AI image generation at no cost, aimed at brands that want more than the default catalog shot but don't have a photography budget. It sits between the two extremes: better than an auto-pulled product image, generally not as tailored as a real lifestyle shoot built around your actual product and audience.

Worth testing, not worth assuming. Run it against your auto-pulled default and, if you have one, against a real lifestyle image, before deciding which one earns the placement.

The common mistakes

• Assuming "Sponsored Brands" means the same product everywhere. Amazon and Walmart Connect both sell a Sponsored Brands unit with custom image capability. They are different platforms with different specs, different approval rules, and different image requirements. A creative built for one will not simply drop into the other.
• Reading "optional" as "worthless." The requirement went away; the performance gap didn't. Skipping custom imagery because it's no longer mandatory means defaulting to whatever catalog shot Amazon pulls, which is rarely your strongest asset.
• Transitioning under-stocked Collections campaigns. Amazon's own guidance is at least 3 products before moving an existing collection into the new format. Fewer than that and the ad underperforms regardless of the image.
• Not testing it properly. We've run accounts — ours included — where a Sponsored Brands ad sat on a logo-only image for a full quarter because nobody owned the follow-up test. The fix isn't complicated: hold keywords and bids constant, swap only the creative, and measure. Every proposed change we make now carries the evidence behind it, a measurement plan, and a rollback trigger — the same discipline that turned the Walmart test into a real answer instead of a guess.

Where this fits into a broader ad account

Creative is one lever. It sits next to bids, budgets, keyword and targeting decisions, and inventory — and a custom image test only means something if the rest of the campaign held still while you ran it. That's the harder part for most sellers: not building the image, but running the controlled test and acting on what it shows.

Dr. PPC is built to do that kind of testing across an account without waiting on a spare afternoon — it reads the account, proposes changes like creative swaps with the evidence and a rollback trigger attached, and lets the client set how much runs automatically. It's priced at $300 a month plus 3% of ad spend, capped, month-to-month, first 30 days free, and it includes Orbit, the analytics and tracking suite, at no extra cost. Whether or not that's the right fit, the creative test above is worth running either way.