Within this document you’ll find a narrative postmortem for the game We Went Back. This is not a postmortem of the game at large but it does touch on elements of production that ultimately impacted the final narrative development of the game at its release. It represents my recollection and analysis at the time of writing and in no way represents others on my team or their personal experiences in development.
We Went Back is a time-looping horror game in which a player wakes up in an abandoned space station utterly alone. The exit door is locked and the player must find a way to escape what lurks within. The player must find clues, experience dynamically changing environments and confront terror in order to find a way out.
The game has an average playtime of 28 minutes and was released as free-to-play on April 3, 2020. As of the posting of this postmortem, the game has achieved 100,000+ downloads, 2-3 million+ views on YouTube, and has a Very Positive designation on Steam with an average of an 86-87% approval rating. It also achieved Top 10 streaming status on YouTube Gaming, Twitch and Steam.
We Went Back was developed by Dead Thread Games, an independent game studio that was formed in Salt Lake City, Utah. Development occurred over the course of eight and a half months with the help of sixteen developers.
The original pitch was formed in August 2019 by Emilee Choate who proposed a game with P.T. style gameplay combined with Firewatch style dialogue branching. A prototype of the environment was greenlit for production in November and the initial development team was joined by ten other members to form its current composition. One of those members was myself, Lis Moberly, a narrative designer. I proposed that the Firewatch dialogue branching would be out of scope given the time and resources of the project and from there the project pivoted to reflect its current design.
With the tone and reference material from Choate, I used my background in Early Modern philosophy and literature to create a unique backstory and narrative foundation for the game that centered around the Apollo 11 mission and the 16th century philosopher, Giordano Bruno.
What Went Well
Scoping Down Early
Scoping the narrative early was a key decision that aided We Went Back from the beginning. Emilee Choate had been brainstorming the initial pitch of the game for several years, but as a designer in specialty, it wasn’t until we were able to meet that it was clear that the ideas needed to be consolidated in order to make the narrative communicable within the game.
Two primary references were chosen in terms of design and gameplay: Hideo Kojima’s P.T. and Campo Santo’s Firewatch. I emphasized early on that creating the dialogue tools and branching to achieve Firewatch style quality would probably not be in scope for what the team could manage. However, a P.T. style narrative was possible.
Later, this decision would end up being pivotal to the game’s success as the manpower on the team was centered in art and engineering. As the only narrative designer and writer on the team, I knew that my bandwidth to write several branches of dialogue and maintain the narrative integrity of the game with my team’s time and resources was limited. Had dialogue branching been introduced to the game, We Went Back would have certainly failed in terms of quality and quantity due to the lack of other narrative specialists on the team.
The game succeeded in part because the reference material was so good. In the case of We Went Back, P.T. was the primary gameplay inspiration for the team, as was Renaissance philosophy, literature and art.
I conducted several playthroughs of P.T. (fortunately a team member had a PS4 with a downloaded copy on it) and I learned that the game, despite its reputation for being obtuse, has a clear story. A pregnant woman has been killed by her husband and until you learn all the circumstances of her death, you cannot escape. By the time the game is finished, the player knows how she died, where she died, where her body was placed, and what happened to her unborn child. Furthermore, I found that the scares were built up to connect to the narrative elements of P.T. – the scares were deliberate to the story and were not there for their own sake. The build-up and pay off of those scares were connected to the player’s understanding P.T.’s premise and this unlocked what the experience of the game narratively needed to become during pre-production.
In addition, an incredible gift was given to me as I was in the early stages of constructing the We Went Back’s backstory. After researching several conspiracies surrounding the Apollo 11 Moon landing, I discovered that conspiracy theorists often point to the Giordano Bruno crater as the location where an ancient alien species dwells, proving that the Apollo 11 mission was fabricated. Giordano Bruno had been the focus of my research in graduate school and I knew that I had hit the jackpot. Digging out books, pictures and academic articles – some of my own publication, I was able to construct a narrative for We Went Back that incorporated both the elements of moon conspiracies that were the foundation of the game’s pitch with the Early Modern philosophy that would undergird our narrative in terms of character, location and plot.
Without P.T. or having the amassed information that it took me years to collect on Giordano Bruno, We Went Back in its current form wouldn’t have been possible.
Nixing Traditional Walking Sim Features
We Went Back has been described by many as a P.T.-like walking sim and these players would be correct in that assessment. With the insertion of escape-room style mechanics such as a puzzle with clues you must find, it is an easy conclusion to draw.
Understanding that, I took a hard look at the best and worst parts of traditional walking sims that have both hindered and helped them as a genre.
The natural impulse of those creating a walking sim would be to do what most walking sims have: include post-its, letters, notes, journals and several recordings. While this text-based information interests those who like narrative-specific games, I understood mechanically we’d have tension between the desire to keep our player continually looping through the space station and stopping to read these found documents. Each time a player picks up a letter or note, they are forced to pause movement in the environment, which risks interrupting the gameplay flow and immersion for the player.
While recordings were discussed several times and one is planned for insertion within the game in a later patch, I made the conscious decision that the arbitrary nature of these elements needed to be largely absent from the game unless we could find a tangible solution to them not interrupting gameplay. All text in the current game is not interactable, meaning that a note cannot be picked up and closely inspected by a player. There is a zoom mechanic which allows players to inspect text closely on posters, whiteboards, and different assets within the game but it is deliberate and limited.
This decision was controversial on the team and it went through discussion several times. Walking sims as a genre by now are deeply entrenched, even with their flaws. However, I think the success of We Went Back is partially due to abandoning these features and choosing to focus on the core gameplay loop and pacing.
The team in turn relied on environmental storytelling through our visual design and event loops, particularly in terms of the creature’s presence in the space station and the dynamically changing environments. Since the game’s release, many players and reviewers have expressed appreciation for the conservative use of these elements that often are inserted without a close analysis of their necessity.
We Went Back has two core pillars: Move Towards the Horror, and Encourage Cultification. From the beginning, an ARG was discussed as an exciting way to engage players and invite them to participate in the game’s narrative. However, due to time and resources, scoping dictated that we needed to focus all narrative energy on the core game and its development. Therefore, the decision was made to abandon the ARG. However, in addition to my responsibilities on the narrative, I was also in charge of our studio’s social media and found that despite tabling the ARG, months of material gathering for the ARG allowed the remnant of it to exist on Instagram.
Later, this decision to include light ARG references into our Instagram became important to our player base and community. It encouraged them to play the game multiple times, going back each time a new social media post occurred in order to see if they could identify the connection it had to the core narrative.
The light ARG still allows for these affordances: our players can to continue to enjoy the game, participate in the narrative, and evolve in their personal interpretation of its meaning. It has also drawn our community closer together as they rally to postulate new meanings and theories between the game and the ARG.
Playtesting was, without a doubt, the smartest decision we made in terms of active development on the game. While playtesting is generally thought of in terms of gameplay mechanics and bugs, I specifically analyzed all our playtesting results for narrative feedback and for what our players were experiencing on a story level. I surveyed their responses and looked for patterns in what players found narratively satisfying or what they didn’t find convincing. Oftentimes these narrative problems coincided with design issues and on some level, we weren’t able to address all the issues that were identified by our playtesters. However, having the information empowered us to make decisions and execute them.
I firmly believe narrative playtesting needs to be utilized far more industry-wide. It was crucial for us to see what did or didn’t resonate with our players in terms of scares, our puzzle, game premise, and environment – all of which impacted the game’s story in some way.
What Went Wrong
One the primary complaints of players is that the ending is anti-climatic and upon reflection, I agree with them. As of the date of writing, an amendment to the ending is planned for patch on May 1, 2020 to address this issue.
Due to scope, time and the need for remote work with the onset of COVID-19, my team was unable to conduct the last rounds of playtests that would have allowed us to measure the success of the game’s ending, in addition to the pacing of the game’s scares.
There are two mandatory scares in the game, and the others are optionally triggered depending on player behavior in the game. However, the last mandatory scare happens midway through the game, leaving the rest of the game’s experience absent of fear or tension if the player doesn’t trigger other optional scares. This possibly leaves the game with more than fifteen minutes of zero tension – far too long in a horror game that’s designed to be a half hour experience.
We should have included at least one other mandatory scare that was tied to narrative in the game’s climax. We hope this will be amended in the forthcoming patch.
In addition, at the conclusion of the game the player comes to realize that they have been the creature they’ve been fearing the entire time. This was shown narratively in the game by setting up a scare event in the first few minutes of the game. The player approaches the suits room and after interacting with a grate on the left-hand side, sees the creature emerge, reaching out their arm to grab them. At the end of the game this moment is directly connected when the player solves the puzzle and wishes to exit out of the same vent into the space station. The player is paused in eyeshot of the vent and they see boots walk past in the hallway. After the boots pass the vent pops out. The insinuation is that in the final loop you are the creature, and the boots that walked past you was yourself at the beginning of the game.
Thankfully, many players immediately drew this conclusion. However, many haven’t. In the original conception of the game, animated arms were going to be included to inspect objects and assist with this plot point. However, the arms were removed due to scope. Had they been included the player would have been able to look down and see that their arms were in a space suit. Additionally, in the moment when the vent is popped out by the player, we would have had the creature’s arm push it to emphasize the connection to the first appearance of the creature.
It’s hard to know if including the animated arms overall would have helped or hindered the game. However, in terms of this essential narrative moment, I still believe they would have aided the player in understanding the conclusion of their story arc.
Puzzle’s Lack of Tie to Narrative
One of the biggest struggles during the development of the game was connecting the game’s puzzle to our narrative arc. The puzzle was designed by another team member and included narrative elements that were made independent of feedback from me in how it tied to the agreed upon backstory. Looking back, this issue should have been addressed by me immediately, but out of a desire to not step on others toes and pick my battles, I didn’t speak up. However, that single decision ended up creating a huge headache for the game later down the line as the results of playtesting showed that the game felt disparate between its gameplay loop and its narrative.
As a result, the team agreed late in development that the clue objects and the password needed to change to accommodate narrative moments we needed to achieve in the game.
Ultimately, the lesson I took away from this is that a close relationship needs to exist between narrative and design if both areas in a game are going to co-exist successfully. A lack of communication between these disciplines is common within games, which I’m afraid happened to our project as well. I don’t believe any one person on the team is to blame for this occurring. I think we managed to respond to player’s concerns as revealed in playtesting as well as possible.
Finding Documentation Formats that Communicated with the Team
A challenge during development was getting easy access to narrative documentation in one specified location during the project. At first, we put all documentation in Confluence. However, we soon found several limitations with the software and that some team members felt overwhelmed by the amount of information placed on these pages. In addition, it was limited in terms of integrating documents and sharing them easily across platforms and users.
Another of our biggest struggles during development was to find documentation formats that were easily readable for the studio to disseminate and understand between all the various disciplines. For example, the original documentation of the game’s events was first formulated in a bullet point list. Because the game features time-looping in its narrative, this list became confusing across departments in communicating what needed to happen where. Later that list was adapted by another team member’s smart decision to place it into a grid, although it was incomplete and needed to change as iteration progressed. Next the team tried using story boards with the help of our concept artist and while this worked at first, later it was difficult to communicate and account for every necessary gameplay moment in story boards, not to mention our artist had other pressing tasks. Next, we tried a screenplay format, which I was comfortable with for cinematics and gameplay sequences. Yet I discovered that not all members of the studio were as comfortable with this documentation as I was. Therefore, I adapted the screenplay format to include images, bullet points and gameplay breakdowns. However, I found that while some people read them, others still found it difficult to parse because of their unfamiliarity with the format. Ultimately, we returned to grid format and made sure it reflected where the game stood in active development. It was made as a shared document versus a pdf so that it could be collaborated on and changed throughout development. Having a visual and dynamic version of our events was ultimately the best choice for our team and was used for the duration of development.
However, too much time was expended trying to figure out the best way to communicate to everyone the needs of narrative across the team. I think had this been done sooner, less time would have been wasted and there would have been less confusion.
Didn’t Have a Pipeline to Approve OST
A struggle for our team was establishing a consistent pipeline in approving OST. Although I wrote the more prominent parts of the OST as seen on the posters, there were small blocks of text that I couldn’t script due to the demands of the art pipeline and production. Therefore, the artists would often fill in their own block text. At first, this was happening without double-checking if such text would conflict with the narrative of the game. I knew that if we placed text within the game, players would try and read it. I was sensitive about it and tried to create an OST sheet where artists could place all their text and then I could amend it as need be.
However, given the pressure and demands of production with other narrative aspects of the game, I simply did not have the bandwidth to correct or amend these OST elements in time for release. It still remains one of my biggest regrets because much of this text has been read and interpreted by players in ways that I wish had been different. I feel that the narrative lack or dissonance of this text has reflected poorly on me as a writer, although it wasn’t my work. However, because I am named as the associated narrative designer on the team, it’s logical to assume it was my contribution.
The OST pipeline and its failure is no one’s fault but my own lack of time and ability to scope my responsibilities properly and I have learned to be more precise about scoping this element of future projects in the future.
Working on We Went Back was a one the greatest creative privileges and challenges I have undertaken in my games career. Narratively it was a unique project that I am unlikely to do of its sort again due to the nature of its game design, length and format. This project reflected more of my personal passions than any other game I’ve worked on, which makes it particularly special. It is also unique in its composition of talented team members, without whom I could have never succeeded. Many of them have become like family to me.
I also hope that this postmortem helps current or future developers gain insight into their own projects, which I know they are laboring on with blood, sweat and tears. Thank you for working so hard to bring meaningful stories to your game. The world will always need new voices.
And finally, I cannot express how much gratitude I feel for our players. To see the world love the game and embrace We Went Back, dive into the story as I’d desperately hoped they would and have an experience they’re excited about has been the reward of a lifetime. To our players, this game was first and foremost for you. Thank you.