Designing Adobe Lightroom’s MAX ’23 headlining feature
For those who don't know me, my name is Jessica Kende. I am the Lightroom Mobile Growth designer. But for the past year, since last October give or take. In addition to my growth role, I have been working on implementing a new feature — one of Lightroom’s most exciting features releasing this October 2023 at Adobe MAX!
This was completely out of scope for my role, my position, and frankly out of my comfort zone and the task before me was daunting and filled with more unknowns than anyone on the team anticipated or warned me about.
For the past year, I have been the lead designer on a feature across our entire Lightroom ecosystem, meaning ACR, Classic, Desktop, Web, Android, and iOS. This was my first time implementing a feature, my first time working across the ecosystem, my first time designing an AI/ML feature, my first time getting designs obliterated in critique, my first time working with Adobe Research, and the list of firsts could truly go on and on.
Going into this feature there were a few things that I felt were very important to implement with this feature, or as we like to say on the Lightroom design team, hills that we will die on.
Firstly, I wanted to make this feature freemium for mobile users.
Secondly, and hand in hand with the first one is that it is a One-Tap Success.
And finally, we simultaneously ship this feature across the entire Lightroom ecosystem.
Before I explain why I felt those were essential. I want to share what I have learned that I will bring to my next feature project in AI/ML.
- Set a quality/ performance goal
- Never anticipate AI/ML will work properly
(also its a lot of **bleeping** work, get a design partner if you are working across 6 surfaces)
Set a quality/ performance goal
We knew our users wanted this feature since 2021 when it was the #1 voted innovation feature from a user survey. However, I was unsure about their quality expectations for this innovative feature once it was released.
In hindsight, I wish I had created a quality and performance rubric before I started designing. It seems easier said than done — it’s like grading art. This is a challenging task — measuring subjectivity objectively.
I think if I had defined my expectations for output, quality, and processing time, then shared this with the cross-functional team, we could have come to a consensus on what we believed was the most likely and best-case scenario of the tech behind this tool. Not only could this have helped engineers prioritize, but it would have helped start the conversation on where we believe the quality will actually be at release.
And I am still thinking about how I would do this for the next project. Likely having a few yes or no questions in each category, helping me become objective.
Thinking about the quality of the AI/ML model made me realize earlier on that we needed to design a contingency plan when the computation doesn’t work properly. A clear fail-safe.
This leads me to the second learning;
Never Anticipate AI/ML will work properly.
I learned this the hard way. As a growth designer, I strongly believe in the scientific method of design, test, iterate — we were completing multiple cycles of this throughout the whole year for this feature.
It wasn’t until the final two months of design that I realized the tech may not be what I originally anticipated. And when this was discovered, I was a deer in headlights.
I could have anticipated it, and I should have, but I didn’t..
To be honest, solving this was a scramble — it was a buzzer-beater. This feature is the first time we are working in the Z-axis on a Lightroom user’s image. When the AI/ML model fails, we needed to come up with a way for a user to manually fix it — but there were challenging and new concepts involved, and it wasn’t as straightforward as we would have hoped.
In Hindsight — I wish I had prepared a design for AI/ML miscalculation. And when I design my next AI/ML feature I will be sure to do this and keep it in my back pocket for when the day comes that we need it.
I want to give you all one more lesson I learned, a bonus lesson
Those three things that I would die on the hill for, had never been done before in Lightroom.
And I want to explain why I still believe these were essential to this feature, and I hope will become a pattern for future releases of Lightroom. In my mind, they all kind of go hand-in-hand together, but I will do my best to separate them.
Firstly, As a growth designer, I am acutely aware of the power of Product Lead Growth. Releasing tools that allow users to all users to access (freemium and premium) has the ability to create a much larger impact across the ecosystem. It allows users to perceive the value before paying — comparable to trying on clothes in the fitting room. This provides trust between the user and the tool and the ability for it to affect the business more dramatically. Moreover, by providing to a larger audience, more users have access to it, which allows us to harness more data around the usage of the tool — which helps us give more priority to the optimizations that make sense.
I also want to touch on One-Tap Success. I define it as a tool that provides exportable results with one tap or one action. And this is already done in Lightroom with some of our highest usage features across the ecosystem: Auto and Presets. Now, this doesn’t mean we don't provide tinkering and customization for users who want to play around. It means quality results with one tap as a starting point, giving all users simple success.
Thirdly, I wanted an ecosystem sim-ship
Firstly, it’s really cool!
But more importantly, it will allow us to understand usage patterns across all surfaces, further understand user segmentation, and iterate and optimize more effectively per surface. It also gives a golden key to our Product Marketing Partners to make a big splash about the feature to our entire audience — which you will be able to see this October at Adobe Max 2023.
Final Thoughts
While this project has taken my whole heart and more, I plan to continue taking risks, working outside of my comfort zone, and pushing myself. This project has led me to build my community within Adobe, Lightroom, and Adobe Design. It has also given me the opportunity to speak to Senior Leadership and all of Adobe Design at our yearly Design Summit. I am so excited to see this very tangible feature in our ecosystem, and looking forward to the feedback from you all.
Is there anything that I missed ?— let me know in the comments (:
