Making memories worth coming back to.

As OneDrive moved toward a more Photos-first consumer experience, we needed to understand what people wanted beyond safe storage.

Foundational research pointed toward reminiscing as an important job. People wanted memories to find them rather than always having to search. Story-like experiences from Instagram, WhatsApp, and Google Photos had also created familiar expectations around how memories could unfold.

But copying that format wasn’t enough.

A memory feels personal because of what it contains: the people, places, relationships, and moments inside it. No transition or visual effect could substitute for that connection.

That became the foundation for how I approached the experience.

I worked with product to shape the ambition into a crawl, walk, run plan.

The useful part wasn’t the terminology. It was agreeing on what each phase needed to accomplish and being disciplined about what could wait.

This helped us keep the first release focused without designing it into a corner.

Web gave us significantly more real estate to work with.

The temptation with a larger canvas is to fill it. More navigation. More actions. More persistent controls.

We went the other way.

I wanted the active photo to command the experience, using the additional space to create presence and immersion while keeping orientation and control within reach.

That led to a simple hierarchy:

• The photo is the hero.
• Playback controls support the immediate experience.
• The filmstrip reveals the broader memory.
• Collection-level actions sit outside the playback hierarchy.

This gave Slideshow a very different rhythm from the surrounding photo library.

It could feel cinematic without becoming a black box. Someone could lean back and watch, but the tools to pause, explore, or change direction were never far away.

A slideshow naturally asks people to move forward.

Memories don’t always work like that.

You might want to go back to the expression you noticed two photos ago. Jump ahead. Pause on a particular image. Or simply understand how much more of the memory there is to see.

The filmstrip gave the sequence shape.

It showed where someone was, hinted at what was coming, and let them move directly through the collection without pulling them out of the experience.

That meant carefully balancing its visual weight.

Make it too prominent and Slideshow becomes another gallery. Hide it too aggressively and people lose orientation and agency.

So I treated the filmstrip as a secondary layer of the memory: available when useful, visually connected to playback, but never competing with the active photo.

The larger web canvas made that balance possible. We could expose more of the sequence while still allowing the current image to feel expansive.

The player needed to accommodate people who wanted to press play and lean back, as well as people who wanted to actively navigate the memory.

That meant I didn’t approach controls as one generic toolbar.

I organised them around the intention they served.

The goal wasn’t to hide functionality.

It was to make the interface understandable through placement, hierarchy, and consequence, so people didn’t have to decode what each control was going to change.

Playback is only one way of experiencing photos.

Sometimes something catches your attention and you want to stop. Sometimes you want to see everything that was included. Sometimes you want to spend longer with one photograph.

So I designed a static collection as another view of the same memory, rather than a route back into the general photo library.

The title and context remain. The relationship to the slideshow remains. And crucially, the path back to watching remains.

That distinction sounds subtle, but it mattered.

I didn’t want someone to feel like they had exited Slideshow just because their intention had changed from watch to look.

Moving from playback to the collection was only useful if people could comfortably return.

The filmstrip, static collection, and player therefore weren’t designed as separate destinations.

They were different ways of engaging with the same memory:

• Watch it.
• Explore it.
• Change it.
• Return to it.

That continuity allowed someone to shift their attention without losing context or returning to Moments to rediscover the same memory.

We spent a lot of time thinking about what happens after the last photo.

Because ending a slideshow with a replay button would have solved playback, but missed the larger opportunity.

At that moment, someone has just spent time in a memory. They may want to respond to it, replay it, share it, or continue.

I separated those intentions into an end experience with its own pacing:

Just watched → Up next → Playback

Each stage has a different job.

That separation stopped the experience from feeling like one endless feed of photographs.

One memory gets to end before another one begins.

Play next gave us a way to turn one memory into the discovery of another.

But that came with an important design tension.

Automatically jumping to another memory could increase engagement, but memories can carry very different emotional weight. Switching abruptly from one to the next could feel less like discovery and more like the product changing the subject.

So I designed the transition to be visible.

A preview tells you what is coming.

A countdown tells you when.

The viewer has an opportunity to continue or choose something else.

Play next becomes an invitation rather than an invisible product decision.

And because the transition happens within the experience, people don’t need to return to the Photos landing page and begin discovery again.

The end of one memory can naturally become the beginning of another.

A slideshow is a sequence, not a stack of screens.

Pacing, music, transitions, cropping, timing, and photo treatment all influence how the memory feels.

We explored dynamically assembling effects through heuristics. That created variety, but it also introduced unpredictability.

Returning to the same memory could result in a different story.

So we moved toward authored presets.

Each preset defined the behaviour of the sequence, giving us variety across different memories while keeping an individual memory consistent.

The categories also gave design, product, motion, and engineering a shared vocabulary for the feeling we were trying to create.

Static mockups couldn’t fully communicate Slideshow.

A transition beginning 100 milliseconds earlier or an image disappearing before the next was ready could materially change the experience.

At the same time, a polished prototype could disguise the mechanics engineering needed to reproduce.

As we moved toward delivery, I contributed beyond the main interaction flows into the edges and dependencies connecting Slideshow, Moments, and Albums.

These questions are easy to dismiss as specification details.

But collectively, they determine whether an experience feels considered or merely designed for the demo.

My role was to keep connecting those details back to the outcome, clarify trade-offs across design, PM, and engineering, and keep the experience coherent as it moved from concept to implementation.

The initial engagement signals were encouraging:

A separate usage snapshot gave us another view into scale and response:

Slideshow events represent usage events, not unique people. Structured feedback represents submissions, not the total viewing audience.

These signals helped us look beyond whether people entered Slideshow and begin asking better questions about completion, sentiment, performance, and what should evolve next.

Shipping wasn’t the end of the design loop. It gave us a new way to learn.

It’s that the pieces work as one experience.

• The invitation to begin.
• The space given to the photographs.
• The filmstrip that turns watching into exploration.
• The controls that appear around intent.
• The ability to remove something that doesn’t belong.
• Motion that remains predictable and resilient.
• A way back when you stop to look.
• And an ending that can gently lead somewhere else.

Together, those decisions changed Slideshow from a playback feature into a way of spending time with the memories already inside OneDrive.

Because ultimately, we didn’t need to make the photos more meaningful.

They already were.

We needed to design the space around them so people could feel that meaning again.

The photos were already worth keeping.
We designed a reason to come back to them.