Time Line: August 2016 – January 2017
Contributors: Mondo Robot in Boulder, CO
Visit website: globalxplorer.org
As designers, we are always daydreaming about working on that dream project. Most designers might say that this dream has something to do with alcohol or sea turtles. My passions and daydreams have always been about science – I hoped one day Radiolab or NASA would ask me to solve all their problems. GlobalXplorer isn’t NASA or Radiolab, but it was the winner of the 2016 TED prize, so I was extremely excited to be picked for the small team to work on this.
Sarah’s project was a tall order: A website that would leverage citizen scientists to help find every archeological site in the world by allowing them to review satellite imagery of the earth, marking what they see. Once images were marked by users, they would go back to the Parcak team where they would decide what part of the world they should pay attention to and possibly start an archeological dig.
Would this work?
Finding every archaeological site in the history of the world is a big ambition but you have to start somewhere – and that place was the country of Peru. The goal was clear – collect votes on every square mile of Peru, over 5 million image tiles, from the moment the website launched to only 3 months later, when Parcak was scheduled to deliver her next TED talk.
We did the math, we did the math again. and then we did the math one more time. The goal seemed impossible. From my knowledge, I knew of no one that would spend more than 10 minutes voting on endless images of the ground. I was really into science and I couldn’t imagine myself doing this for more than half an hour. But this where I learned something very valuable – what I think I know about my users is useless without doing the research.
It turns out that citizen scientist projects, like Tomnod or Zooniverse, are extremely popular among the disabled, senior citizens, extremely bored, and clearly obsessed people of the world. Tomnod provided the research and the validation that we could easily make it through 5 million tiles in 3 months. The Tomnod crew actually thought we might have the problem of running out of tiles before the time was up.
Ok, let’s make a list.
The best thing to do at the start of a project is to list the design challenges. This allows us to evaluate everything we’re up against, prioritize them, and double check we are solving these problems throughout the life of the project. Th design challenges of GlobalXplorer were:
- How do we train users how to vote on images in the correct way?
- How do we make users care about this project?
- How do we encourage users to participate for as long as possible?
- How do we get users to share it?
- How do we make it fun and enjoyable?
- How do assure our users that they’re being helpful?
- How do we help our users when they’re stuck?
- How do we reward users for participating?
- How do we prevent the GX data landing in the wrong hands?
These questions were developed into a core challenges statement:
The challenge of GlobalXplorer° will be to create an experience that is engaging by making it both fun to use and informative. The communication must be timely and transparent but also protect information that could be helpful to looters. It needs to keep the integrity of the research so that it is as effective and helpful as possible. To do this, users will need to be trained properly. To keep and create new users, it will need to be responsive, easy to use, accessible and always up to date. Effective marketing and secure funding are essential to making all this happen and to keep it going.
Who are these really bored and very obsessed users?
Our target personas included two people who we knew who would use it based on our user research: Ester and Karl. This older generation personas have lots of free time and are motivated by activities that make them feel needed and productive. Our other two personas were people who we wanted to attract to the site: Matt and Olivia. These younger people might have less time to dedicate to the project but were more likely to share what they have learned to the friends, family, and colleges.
The Journey of the User
The third exercise after our design challenges and our personas was the user journey. This exercise is extremely helpful and I try to go through it no matter what challenge I am dealing with.
A user journey is participated with the designers and the stake holders of project. We use sticky notes on a whiteboard to define the overall user path and list out all the pain points they might encounter during their journey. Out of this path and pain points we discover opportunities to alleviate pain points and make a better experience.
For example, a pain point might be that our user gets bored while reviewing tiles. We then discussed possible gamification tactics that could help with this. We could create a progress bar or progress map. We could play meditative music. Whether or not we ended up using these ideas, the user journey exercise gave us an opportunity to throw around different ideas.
Like most projects, there is the site that is ideal for your users and then there is the site that is ideal for your client. This project was fortunate enough to have the perfect balance. We didn’t have the time and budget to add our so-called “brain music” but we were able to sneak in every feature that mattered.
The features that we found to be essential were:
- an introduction that makes the user care about the project
- a tutorial that is easy to understand
- a voting interface that is simple
- pages about the project and the country, Peru
- a library to hold the National Geographic assets and articles
- a place to donate
Creating a simple game is not simple at all.
Although the basics and necessary elements were figured out and ready to be wireframed and designed, the client and our team found that we had a little extra to take on the challenge of figuring out how to keep our users motivated and engaged. While I am very aware of avoiding gamification for the sake of gamification, we felt this project could benefit from something simple.
Creating a game became more complicated than we originally thought. Everyone on the team had a different idea of what made a fun game and or what would work for this project and everyone had a different experience with gaming.
The problem wasn’t creating a game that would work, our problem was that our team was disconnected and not communicating efficiently. I saw this problem and took it upon myself to solve it. I created a visual for every version we had purposed so far and put them in front of everyone. Now that the concepts had a common visual language, we could see the strengths and differences between them. I had the sketch file open and was able to build the final wireframe with the whole team in the room so we could agree on something before anyone left the room (final wireframe above).
The ideas I sketched or wireframed to help the team understand the options: (click to expand)
The final wireframe (left) and what was designed and developed (right):
And it all comes together.
With the well developed process we put this project through, wireframes were simple to build. We were solid on the pages and functionality and we had a clear idea about how it all connected.
The secret is to have excellent engineers.
I can’t say I had much to do with the development of the project, which says a lot about the development team at Mondo Robot. They built what we designed on time and with pixel perfect precision. They went above and beyond what was necessary and I couldn’t be happier with how it turned out.
Putting it to the test with Beta testers.
We had our first taste of the user base and excitement for this project when we received 3,000 user requests to participate in the beta test. The GX team was able to narrow that list down to 300 of the closest friends and colleges. The usertest became my project, which was the creation of a private online survey.
Here are the testing objectives were were after:
- Test the user’s understanding of the purpose of GlobalXplorer and they’re part in the project.
- Test the tutorials as to it’s effectiveness to teach the user how to search for looting and encroachment sites.
- Understand the user’s thoughts or frustrations around participating in exploring/participating.
- Test the user’s understanding of the gaming mechanics and its effectiveness on user retention.
- Understand the interest in the National Geographic articles.
These are the results we got back:
- Users had a few questions about voting on tiles but generally understood what they were doing
- If the user had an interest in the game, they understood how it worked
- We had a good response from very enthusiastic users ready to take on thousands of tiles
Because of a few schedule pushes, we only had one week between the user test results and the launch of the public site. This was enough time to solve some of the user feedback concerns by changing some copy and adding some FAQ questions. Although we didn’t have time to solve any of the major concerns, it gave us confidence that it would work and objectives for phase 2.
Success bragged about on the TED stage.