Imagen is an Israeli startup that helps professional (wedding) photographers with editing and selecting the best photos for their clients by using AI.

Photographers can upload their own photos and the AI learns uses this to learn their specific editing style to apply to other images during the post-production process. With Imagen, photographers can save up to 95% of time they would normally spend selecting and editing photos. Imagen allows them to focus more on what they do best: Capturing special moments.

I supported them for about 2 years as a UX writer and worked on many flows and features of their core product. Here I’ll tell a bit more about the design of the culling settings page.

Challenge: Make the culling settings clearer

Culling is jargon for “selecting a set of pictures for your client”. It used to be a very time-consuming task, as photographers had to manually compare photos and select the best ones.

Imagen wants to make this task easier and faster, by having AI group and label the pictures to give photographers a quick overview of the best photos.

The first iteration of the culling settings (below) was confusing for users and we wanted to make it easier to digest.

First iteration of the culling settings page.

I was working closely with the product designer and product manager to discuss the mental models, terminology, and information architecture to come up with a new direction for the page.

We also had to bear in mind the preferences of photographers in terms of rating photos and defining a group of photos.

Goals

For this iteration, we had a few clear goals in mind:

• Increase user engagement with the different rating methods
• Find a balance between standing out from other photo editing tools and using familiar terminology

Process

Research

We started by doing a benchmark of rating methods in other photo editing software, such as Adobe Lightroom, Aftershoot, Photorefine,… This already gave us an idea of how photographers were approaching culling photos from their shoots.

We also had a chat with the customer success team to understand the main needs that the solution had to convince photographers to use the solution and stay engaged.

We understood that culling consisted of 2 main tasks:

• Rating or labeling the photos to make the final selection process easier. Here we had to work with a mix of rating the quality of the photo (e.g. the best of a group) and labeling them to stand out (e.g. a unique photo without similar alternatives)