
Idea ads on Pinterest appear as a set of multiple videos and images in a single pin. For this summer-themed campaign, Starbucks asked us to create two Idea ads to promote their ready-to-drink chilled coffee collection. We were provided with product shots, key visuals, and video content to repurpose into the Idea ad format. For this project, we created 2x videos and 2x static images for each idea pin, with the videos acting as the intro and outro of each.
Idea Ad 1 – Frappuccino
The first Idea pin spotlights Starbucks Frappuccino Chilled Coffee Drink. The video starts with the campaign’s key visual, with the frappuccino in the centre, then transitions smoothly into footage supplied by the client. The two static images then display one after another in a similar manner to an Instagram story. The outro video shows a short, dynamic part of the footage featuring the frappuccino before transitioning back into the key visual to form a nice bookend, with the addition of a bold CTA.


The two static images were comped together entirely manually, with no use of AI tools. Stock was used for the background imagery, and the bottles were edited in with meticulous attention paid to the lighting, shadows, and focus (blur) to ensure they looked realistically in-situ.
Idea Ad 2 – Collection
The second Idea ad promotes other items from the full Starbucks Chilled Coffee range. The static images in this Idea ad provide some contrast, an unsweetened black nitro cold brew next to a ‘sweet & buttery’ iced caramel macchiato – showing there’s something for everyone. The videos start and end in a similar way to the first Idea ad, however the footage incorporated here is more people-focused rather than product.


As with the static images in the first Idea ad, these were comped together entirely manually. The challenge with stock in projects like these is always finding something that doesn’t look too “stocky”, while also appearing cohesive with the product. In the second image, a table surface was comped into the foreground, so the product could more realistically appear in focus in comparison to the background of the image.