The process of making an average T-shirt involves a trip around the world and more than a dozen steps from seed to shirt. How do you capture this entire process on video within a multimedia narrative? In April, NPR’s Planet Money launched a Kickstarter campaign to do exactly this. Six and a half months after the project was funded with more than 10 times its original goal by more than 20,000 backers, Planet Money finally launched its multi-part narrative. Here’s the intro video:

The story consists of five video chapters filmed across three continents as well as an accompanying text story rich with charts and illustrations, all encapsulated in an interactive and responsive website. Furthermore, the project was funded entirely by a Kickstarter and most of the marketing was crowdfunded as well.

The T-shirt project was the first time Planet Money ventured into video-centric territory. It was also the inaugural project of the new NPR visuals team, which merges the multimedia and news apps team together. Now that we know how a T-shirt is made, I talked to managing producers Brian Boyer and Josh Davis, who also served as video director, to find out how exactly the team executed the project itself.

Q&A with Brian Boyer, managing producer

You guys got an overwhelming amount of support on Kickstarter. What was that like, and did that change the way you planned the project at all?

Brian Boyer: The observation that I’ve made since coming to NPR is that we have a remarkable amount of love in public media. Our audience really likes what we do. When they ‘like’ us on Facebook, they actually like us on Facebook. They mean it. I’ve been working on projects that try to harness that love. The scope of the project definitely increased because of the support we got from our audience.

You also used Instagram as a way to let the crowd market the story for you. How do you think that worked out?

Boyer: Our team has done a lot of projects with people sending us photography, whether that’s through Instagram or submitting a post through Tumblr. They’ve been surprisingly successful. That again relates to how our audience loves to participate. I think it’s because they have a personal connection to NPR. We’re in your kitchen in the morning. We’re with you in the car.

This particular piece was different than a lot of our other pieces in that we had a direct line to the people who would work with us. Usually, these projects are kind of a cold call just to the Internet. And then hopefully some people on the Internet notice. This scenario, since we used Kickstarter, we had a direct line to everyone in the world who owned a Planet Money T-shirt. We were able to email them directly and say, “Hey guys, we have this project launching, we’d love it if you would send us a picture of yourself in the T-shirt so you could be a participant in this project.” I feel that this definitely raised awareness of our product. We had a built-in marketing team. We had a group of people who were interested in promoting the project that they helped with.

The project itself is very engaging with multimedia and data built into the text. What was the process like behind organizing the components of the stories themselves?

Boyer: It started before we had done all the reporting. We did an inventory of ideas of what kind of stuff we wanted to talk about and an inventory of places we were going and who we thought we were going to talk to. We also, more importantly, thought through an exercise that we do with every project where we ask, ‘Who are our users? Who cares about this? What questions do they have about the subject?’ After we brainstorm those and prioritize those, we decide which we care about more, which questions we care about more. Finally, we talk about what features we’re going to build.

It’s very easy to think, ‘We should take photos and have some videos.’ I feel like if you do that before you really think about your audience, you’re missing the point. This is sort of a user-centric design process we like to follow. The problem with this is that, when we’re doing the reporting for this, we don’t really know what we’re going to get. We know we’re sending people to Bangladesh, but I don’t know what they’re going to come home with. We can plan our motivation and structure, but a lot of it really came together as we were figuring out what we had and what stories we could tell.

NPR t-shirt machines
Planet Money’s T-shirt project site included full page images and video

Our teammate, Wes Lindamood, is specifically a user-experience expert. He spent a lot of time touring other people’s structures for how they combine video and text and charts. We looked at a lot of people’s work and thought of the guiding principles. We wanted to use each medium we had available to us to its best. Let’s use video to tell the part of the story that video is really engaging. Let’s not cram the text story into the video. Let’s use a chart when it furthers the story. We arrived at this novel interface where the narrative of the video was the heartbeat of the piece.

We realized one of our strengths as an organization is narrative storytelling. Planet Money’s sound, the way we explain things. Why throw that out? We’ve got something very special here, we should use it. At the end of the project there’s a lot of user-testing, where we failed and putting it back in front of people. We did a lot of ad-hoc testing with people in the building. That part of the process was really essential; we changed a lot of the project structurally.

Whose work did you look to for inspiration? What were some key takeaways you learned from these other projects?

Boyer: We looked at the New York Times. National Geographic had a great piece about lions in the Serengeti that was really beautiful. The big part of the exercise was thinking about the structure of the piece.

If you have just one rule when making software, a good rule is don’t invent something new. People have learned to use a website, they’ve learned to use a door, they’ve learned to use a telephone. When you change the interface, that’s a really difficult thing to get through. A lot of these journalism projects are giving people a totally novel way to interact with them. That can be very difficult for someone to use. It’s very easy for us — after spending months building it — to completely internalize how it works. That’s the importance of user-testing. There’s a lot of things we realized through user-testing: Certain things need to be more obvious, a lot of things we need to throw out until we stripped it down to just the essential bits.

I feel like the project is very minimalist. It has a definite structure, and there aren’t a lot of choices for the reader in terms of navigation. Was this intentional?

Boyer: That was definitely intentional. This story had a definite narrative thread where we start with cotton and we end up with a T-shirt. We felt that the steps in the process were stories we wanted to tell; there was an obvious way to tell them one after the next. The trick is, how do you get people to watch a video, read some text and then watch another video? Or let them watch videos if they just want to watch videos? How do you give people enough choice and enough way finding points so that they understand where they are in the process, they don’t feel lost?

We did lots of little things. Right before the video is done playing — while we’re still talking to you — we scroll you down to the text. We make it obvious that you can click to the next video, but we’ve also made it really obvious there’s a bunch of interesting stuff here that you might want to read.

Boyer says that he often used the architecture of Frank Lloyd Wright as a metaphor to explain the structure of the project to his team.

Do you think you guys made something that you could replicate in the future? Would you change anything about the project?

Boyer: I generally shy away from the idea of calling a project a template. With each story, you have something different to say. That thing, the next time, might need a different treatment. So it’s not something that we can copy and paste as a template. But the thing we do have is that we now have a team that knows how to work together on a project like this, which is a huge step. This was the first project of the unified NPR visuals team, which we just recently merged the multimedia team with the news applications team.

If a story similar to this came along the way, we might crib a whole lot of code. But I think it’s going to be awhile before we do another half-million Kickstarter that’s process-oriented. One of my goals for the team is to produce projects like this more frequently on a shorter deadline.

Q&A with Josh Davis, video director and managing producer

What was the basic timeline for the project? How far ahead did the team have to plan to get coverage from all these different locations?

Josh Davis: The timeline centered around when our shirts were in production. What’s fascinating about this project is that it started with people pre-ordering T-shirts on Kickstarter. We’re working with Jockey to make the shirts. So a lot of it was arranging with Jockey.

I was brought in at the end of May, beginning of June. We initially thought that the project might wrap in September, October. But we started realizing that this project was going to be bigger than we expected. We gave ourselves a six-month timeline between June and December.

We wanted to tell the story in a couple different ways. We wanted to show the process of the shirts being made. We’re getting factory footage, we’re seeing how seeds are manufactured, how the shipping is executed. We’re also trying to meet the people along the way. We started running into some great characters, people who had great stories. When we found that, our instincts were we want to go and we want to go deep on that. We want to make sure we get all the assets that we need for that. I had everyone approach every shoot as if we were shooting almost for a feature documentary so that when we came back after a couple months of shooting on the road, we would have the assets that we need for multiple ways of telling the story.

When we reassembled the team in September, I wanted to be sure that whatever avenue we decided to go, whether it was going to be one big video or a chapterized project, we’d have what we needed.

I noticed as well that most of the videos cap in under two minutes. Was it hard condensing all of the footage and info into something so bite-sized?

Davis: We have a team and everyone has different ideas. We brainstorm and synthesize our ideas down until we started coming up with something that worked. One of the challenges was — at the beginning, before we shot footage — you have a lot of conversations in the abstract. Everyone has a vision for the project but, until some of the visual assets in place, no one is quite sure what you’re going to make. I think the same is true for documentary filmmaking too. You go out and you shoot your video, but you come back and the story changes in the edit.

Here, the same principle holds, but you have the extra ability to change the delivery system or architecture to tell the story. We shot about a hundred hours of footage, which is an insane amount, but it was kind of neat. It all found a place. If you wanted time-lapse photography, we had that.

Once we started finding focus, it was really easy for us to sit down as a group and figure out ways to chapterize it. That’s the main thing: You want to find a focus with the story you’re telling. We wanted to be very deliberate about the fact that we’re showing the process of how this shirt is being made, but at the same time sharing the stories of the people who are making them and explaining the economic concepts along the way. As we found a further focus, it was pretty easy to chop some stuff out.

Davis explains the decision to break the project into short video chapters.

How did you keep a consistent tone and narrative between all the different locations?

Davis: Every single shooter we used, I would have a conversation with them. I would talk to them specifically about what the shot list is and show them some visual examples of how we’re shooting. One concrete example is that the final chapter, the “You” section, we knew that we wanted to have a portrait of each character. Sometimes I think video portraits can be cheesy or overused. But I thought that if any project should have one, it should be this one. There’s so many people involved in the making of a simple T-shirt. One thing I told them is to get a video portrait of the person. I sent a shooter to Monsanto, and one of the most intense portraits that we got was this lab scientist who worked in this really visual lab and he was a seed engineer. We just found out details about him. He likes to brew mead. That’s kind of neat, let’s get that in there. It’s fun and interesting.

What were some challenges you faced and obstacles that you overcame in terms of getting all the multimedia necessary for this project?

Davis: One thing about this project is that it works on multiple delivery devices. It works on your phone, it works on your desktop. That was a goal that we wanted to have at the beginning. As a video-driven project, we knew that was going to be difficult. When you look at a lot of other large multimedia projects that have video, a lot of stuff that’s come out in the last year, you go look at it on your phone and it says “this experience is optimized for desktop.” We didn’t want to do that. We wanted to be sure that each user could access this. We wanted to be sure that anywhere you are, whatever device you’re using you can access this website. That was a big challenge for us in trying to figure out how that would work.

Did you have any difficulty with access for any of the reporting?

Davis: Especially with international reporting, there’s going to be some level of access issues. Some of it is just the traditional challenges of international reporting: Their language is different, their customs are different. And sometimes there’s just a lot of traffic, stuff you don’t think about ahead of time. In Bangladesh, it took us a really long time to get around.

Working with Jockey, there’s a sensitivity to places like Bangladesh. We’re reporting on a country on the heels of Rana Plaza where the factory collapsed. It’s evident that they want to put their best foot forward. As reporters, we understand what’s going on there. We understand it’s going to be a little difficult for us to get access to the workers. I don’t think they wanted us necessarily talking to workers for this project, but we had to do our jobs.

Once we really introduced ourselves to Jasmine and once she really understood what we were making and why we were making it, we got in. For us, we come from a documentary, photojournalism background where we’re not necessarily out to expose something scandalous or find a smoking gun. We really wanted to know what her life is like, we wanted to get a solid insight into what it’s like being a garment worker in Bangladesh and why someone chooses this life over another life. We went back with her to her village and saw what her life was like there. She really opened up to us once she realized what we were making and why.

How did you aggregate all the information and multimedia that you had and decide on the final flow of the piece?

Davis: We’re making a project for Planet Money, and Planet Money has a lot of personality. We don’t want to tell a traditional type of narrative where you have your character-driven story and sometimes they’re very heavily produced. We knew that we wanted to use the voice of Planet Money in some way. What we ended up doing is having Alex Blumberg do the narration. We wanted to have the project the fun and conversational tone that Planet Money brings to its storytelling.

We wanted to also play to our strengths. We knew we have the Planet Money voice, we had some really good footage. But we needed to find that third thing to make it right, and that was the delivery system and that was where news apps came in.

How can we best tell the story? This is “seed to shirt,” what are the phases between seed to shirt? And that’s when it became so apparent that we were going to tell the story this way. We started realizing that as we got more focused that we were going to chapterize it.

We wanted to make something that people were going to watch, so we were going to play to the strengths to what we had. We have Alex, we have good video and having the news apps team being able to break that down in a way that gives people a way to access the information without having to make too many choices. We didn’t want to overwhelm people, we wanted to herd people through a little bit. It gives you a way to easily navigate without getting lost.

Looking at analytics, close to 60 percent of the video plays are being finished. It goes higher for some of the longer videos. The “people” video is about six and a half minutes long and the finish rate on that is somewhere near 60 percent. For web video, that’s a big deal. To anyone who wants to say, “web video can’t be longer than 2:30,” I would say that’s wrong. Good content is going to deliver results.

Denise Lu is a former editorial intern at PBS MediaShift. A journalist focusing on culture, tech and online media, she has contributed for Mashable, Pretty Much Amazing, Evolver.fm and other publications. Denise is currently a senior at the Medill School of Journalism at Northwestern University. She recently joined the design team at the Washington Post as part of her journalism residency.