DIRECTV Stream

In compliance with my non-disclosure agreement, I’ve obscured confidential information. This case study presents my own ideas and work and may have been summarized, re-created or altered to maintain confidentiality.

One of the biggest challenges I faced was integrating this new tech within the existing, limited infrastructure (which predated the project by a few decades). Creating new products was a difficult match for older, proprietary systems. The languages and syntax were substantially different.

Deadlines were already approaching, and I needed to test and quickly iterate upon several new ideas. The legacy codebase and fussy, older satellites made it tough to flush out new ideas quickly. I didn’t have the luxury of time to get tangled in complex programming languages of yesteryear.

As a designer, it’s necessary to fully-understand what I’m building. After joining the design team at AT&T, I learned that I had an incredible opportunity with a seemingly infinite pool of knowledge resources.

My team consisted of design veterans, satellite wizards, technical geniuses, industrial pioneers and an army of top notch developers from all over the world. Needless to say, I had a lot to figure out.

Where should I begin? What was the vision and goal of the overall product? Since I was hired to bring a prototype to life, I began by reviewing existing design assets and writing down a list of initial questions.

It was my first experience with an OTT (over-the-top) platform. It was especially unique because a majority of the content was being pulled from outer space using the existing satellite infrastructure (similar to the diagram below)

As a warm up to develop my understanding, my first task was to re-think what the experience of using the guide would be on a streaming platform. Created as a way to choose programs in a linear fashion as maybe it was no longer relevant. After all, we were moving from a physical remote to touch devices.

To further understand how the guide worked, I experimented for hours with different environments, game consoles, smart televisions, etc. As a self-proclaimed tech nerd, I was like a kid in a candy store.

Mostly, these devices all worked the same way, but the challenge would be to learn how the remote functioned on a very technical level and what role that played in designing for touch screens. How would the guide look and respond on mobile devices?

I started sketching.

What apps were already doing this well? I was on my quest to learn. The commands were simple (Up, Down, Left, Right, Select), but there were certain engineering complications that I needed to learn in order to solve for them.

It became a fascinating crash course in industrial design. Seeing a big picture of how everything worked together was crucial in moving ahead.

After a few weeks of onboarding, asking a lot of questions and a little trial and error, I learned details of what I was building. I was creating an interactive prototype utilizing available assets for testing.

The trick was that it had to be functional and agile in the sense that updates needed to be made quickly, sometimes even for same-day testing.

The next few days were spent researching tools and software to accomplish this.

I found that a combination of Sketch and Principle would best suit my needs, mostly because the software is lightweight and has the ability to integrate live video into visual design. Plus, they work incredibly well together (as intended). I encountered a small problem in that I had never used Principle and didn’t know anything about how it worked.

Luckily, the Principle website is great and had a lot of resources available for designers looking to learn. I watched every video and was tirelessly dedicated to the cause. I found that teaching my colleagues some tricks I had learned really helped me understand the workflow. In a little under two weeks, I had it figured out.

I gathered assets for testing and started to build my prototype. I experimented with different video formats and content to create a themed user journey. One was made with kids content, one for sports fans, one for the movie buff and one for the binge-watcher. I was ready for the first wave of user testing.

In order to preserve trade secrets, I won’t disclose details of the testing method. However, I will say that customers enjoyed the new experience of what we were building. The feedback we received was that the product was simple, clean and modern.

We discussed new features and jotted down usability notes. The prototype was already bringing several different elements together quickly, which was what we set out to do.

The path was now essentially paved and I was excited to continue.

It’s worth noting that this prototype was also gaining traction with management. In a large corporation, sometimes it can be difficult to see things implemented quickly.

Since I was more familiar with Principle by this point and confident the direction we were going, I started to experiment with conceptual changes, on-the-fly.

This enabled me to really push the limitations of what was possible. I was designing a living, malleable software experience that could immediately change with the needs of the company – which added great value.

But, it was important to understand those limitations so we could manage expectations and know how large we could scale this. The next task was to undo everything I had built and try to make it come crashing down. Of course, I made a backup first.

It took a while, but I kept layering and layering until we reached maximum capacity. The prototype started falling apart and we had our finish line – an interesting and worthy experiment, indeed.

Now that we had a prototype, the team decided to look at solving some bigger problems. Specifically, we wanted to tackle some ideas of how to stream over cell service and what it would mean to enable auto-play. 

Remember, this was a linear format using older satellite infrastructure, being converted into radio waves with the ultimate goal of watching anywhere at any time. The FCC has different rules on different wavelengths and merging them had never previously been attempted. 

It was a new era, ambitious and worthwhile to explore. I was forward-thinking and motivated.

The diagram shows the complexity of a sample signal flow and how it pertains to what we were building. It has been modified and redacted for security. Needless to say, it was extremely complex.

One of the biggest takeaways from testing that we learned was that users still wanted access to their recordings. Content was still mostly linear and DVR was alive and well.

As you could imagine, this posed monumental engineering and usage rights challenges, especially if families were viewing content separately on different devices and even on-the-go. This was pre-Covid, so staying at home and watching wasn’t yet a big part of everyday life for most people.

They wanted their content and they were accustomed to having it in their pockets, ready to go. If we were to remain competitive, we needed our own cloud-based content delivery system.

As the release date was just a few short months away, it was ultimately deemed out of scope for launch. However, several weeks of conceptualizing followed as to what this experience may look like.

The prototype had become a pivotal piece of the testing process. It was working very well. One of the last pieces of my puzzle was to explore ideas as to how we could bring everything together into one ecosystem.

Meeting with a team of industrial designers was a remarkable experience. Since I had built the prototype, I offered my feedback from months of testing and started to understand the level of detail that goes into something like a modem, router or cable box.

Sorting through the hours of research on remote shapes and sizes even just barely tapped the surface. We started asking some very important questions to really understand the vast scope of this project moving forward.

This process takes years to perfect, but helping to develop the roadmap for future products is one of the most memorable collaborations of my career.

My contract had come to an end. The prototype was fully up to date, scalable and consisted of different versions for different tests. The testing continued.

The finishing touches included partnering with other design teams and adapt the designs to other formats and tech stacks. The first version of the product was launched and we were on to new features and iterations.

I gained invaluable experience that has made a lasting impact on design decisions, heuristics, trends and testing methods.

DIRECTV Stream is available on all major platforms and consists of millions of users worldwide.