Bannerbear Image Collections Share Layer Names, but Every Layout Still Needs Review

For teams continuing to use Bannerbear V3, Image Collections send one data set to multiple templates. Mapping identically named layers reduces repeated input but can amplify naming mistakes; text and cropping still need checking at every size.

Source: www.bannerbear.com
November 15, 2026
3 min

When creating square, landscape and portrait images for the same campaign, Bannerbear’s Image Collections can reduce repeated data entry. The core mechanism groups templates into a Template Set and generates a collection using the same text and images. Each size’s layout still depends on its template.

This article concerns existing V3 workflows. The official Image Collections help page is explicitly marked as older documentation, while the V5 migration guidance confirms that existing V3 features and APIs remain fully supported. Both versions run concurrently and do not share project data. Teams maintaining older workflows therefore have a basis for continuing to use them, but should not treat these instructions as a guarantee of V5 API compatibility. This is a documentation analysis, with no actual generation API calls performed.

Layer Names Determine Where Data Goes

The official help explains that identically named layers within a Template Set receive the same data; the API does not distinguish between them. Suppose all three sizes contain title and photo layers. Consistent naming sends a campaign’s headline and main image to the corresponding positions. This mapping saves repeated input; it does not decide how a designer should arrange content for each aspect ratio.

Conversely, if title means the campaign theme in the landscape template but a call to action in the portrait template, the same text goes into two semantically different positions. Before assembling a template set, list layer names and their purposes. Share names only where content should be shared; name fields separately when they need different content, then check each in a sample request. This is a design recommendation derived from the documented mapping rule, not an additional automatic validation feature.

Successful Generation Still Requires Review

Use one ordinary campaign record and one long-headline record to inspect the entire output set: is every title complete, is the font still readable, does cropping retain the important subject, and is the call to action appropriately positioned? The official product page mentions automatic text resizing, but fitting text into a space and making it readable are different matters. An acceptable square image cannot stand in for checking the portrait version.

Record each template’s target channel, dimensions and corresponding output, then review them individually to avoid mixing up images from the same collection. After adding a template or changing a layer name, reuse the same samples to check mapping and layout. Do not merely count returned images.

This feature suits teams with a stable information structure and a recurring need for multiple layouts. If each channel needs a different theme or entirely different content, shared data saves less work. Set two separate acceptance criteria: correct field mapping and deliverability at every size. The former lends itself to repeated checks; the latter still requires design judgment.

Content note:This article is based on publicly available information and is intended as a reference for evaluating tools. Features, pricing, and related policies may change at any time; please refer to the latest official information.

Original source:www.bannerbear.com