Going Episodic

The decision to turn Orwell into a TV-style episodic serial was a bold, risky move. It had the potential to get people talking and elevate the game experience to greater heights, or create the indie game equivalent of a black hole.

In this, the second in a series of posts about the development of Orwell, we talk about going episodic: what went right, what went wrong, and why we’ve decided to do it all again and release Orwell: Ignorance is Strength episodically.

- - -

On October 20th, 2016, with our hearts in our mouths, we released the first episode of Orwell via Steam. Over the following four weeks, we continued to release a new episode every Thursday, creating a five week season (“Season One”). As far as we know, Orwell was the first game to release this way on Steam, and the second game ever to attempt this release structure.

When it was first conceived, Orwell was intended to be a single date release. But as development progressed and our publisher--Surprise Attack Games--came on board, meetings about what Orwell was trying to achieve led to numerous discussions about the game’s structure, how it was broken into “days”. Orwell felt like a Netflix series, or an old-school radio play. And we talked a lot about the influence that This American Life’s audio show, Serial, had on podcasts.

When viewed in those terms, creatively, Orwell really suited an episodic release.

Orwell is a thriller with suspense and plot twists being a critical part of the game. By forcing players to wait between episodes, we hoped to accentuate that feeling of suspense and encourage players to really savour the content in each episode rather than rushing through. We wanted players to think about the game’s theme and how we balance privacy, connection, security and freedom in the age of the internet and social media. 

Commercially, the decision was riskier.

On the positive side, indie games doing something different help them to stand out in a crowded market. It makes the game more remarkable, more likely to get talked about. Given that we were launching in October when the AAA season was in full swing, we needed every bit of remarkability we could get.

The episodic release would give us a launch window of five weeks instead of one, and five launch announcements instead of two. Most games live and die in the week of launch. We had five weeks of “being relevant” with a steady flow of information to deliver to players. By creating a unique experience that only the “early adopters” would get, we also hoped to create a stronger reason to purchase and generate a sense of being in a special club, similar to those that see a band before they get famous.

But, as good as the positives were, the risks were also there. 

When you do anything new, you run the risk that people just won’t understand it. Releasing episodically--with the demo for week one, commercial launch for week two and five “days” of content across four weeks--had the potential to over-complicate an already unusual approach and it was hard to explain clearly.

Reviews, too, were going to be a problem. There was the chance we wouldn’t get reviews of the full game until the end of the season because the media wouldn’t be able to play the whole game. Or worse, reviews were going to be based on only the first couple of episodes, which would miss a huge amount of what was in the game and might be overly critical.

But... as history records, we went for it!

With no real previous example to refer to (we didn’t uncover the fact that Resident Evil Revelations 2 had taken this approach until a week or two into the season), we were pretty much “making shit up” as we went.

The development schedule was tight. We finished each episode while already being in the middle of the release period, which was very stressful, but also exciting and strangely motivating, too. Mel (our Art and Business Manager) remembers listening to YouTube videos of let’s players playing one episode, while she was working on the next one. While this reduced feedback loop was great because it gave us the chance to see people react to important moments in the game, and helped us get a better idea of how to improve things, it was also distracting. For example, if someone found a minor bug, we knew we wouldn’t have any time to fix it before the final release.

Towards the end, time got more and more scarce. We ended up finishing Episode Four just on time and being several hours late with Episode Five, discovering a game breaking bug right before release. As the release time passed, people began to ask what was wrong and when the final episode would be available. This was a rather scary and extremely stressful situation for us.

But what could have been quite the game dev nightmare, turned out to become an engaging chat between our Orwell fans and our wonderful producer and community manager Steve. He informed the fans about our status and why we were late, and they were surprisingly kind about it. We even got late-night emails from people telling us to take our time to fix the bug, as they wanted us to have enough time to ensure the quality they’d come to expect.

At one point there was a person on a ship who really hoped that we would be able to launch the final episode while he still had internet available. We were absolutely amazed at the number of people waiting for the final episode to drop, but also by the fact that they were so friendly about us being late. We’d built an audience of fans who cared about the game and appreciated that we were updating them in real time about the delay.

In the following weeks and months after release we received numerous emails, Facebook and Twitter messages from fans telling us how engaging and meaningful they found the game and how much they loved the overall experience. This was incredibly rewarding, and, despite of all the stress, it was a very exciting and wonderful experience.

For our publisher, the ride was a wild one.

Surprise Attack Games had been hoping that the episodic approach would give them an extra way to cut through the noise and get media to pay attention to the game. And there was reason to be confident. They’d had really solid coverage of the game reveal in August with Polygon, Rock Paper Shotgun, PC Gamer and Kill Screen, in particular, and the angle of the weekly serial seemed like it would get a good reception based on the media they’d been talking to about it. 

Then, two days before we launched episode one, Rockstar made the official reveal of Red Dead Redemption 2 and announced they were going to drop the first trailer that week. The internet went crazy. We were all less thrilled—the trailer was going to drop an hour before we were set to announce episode one of Orwell. 

As if that wasn’t enough oxygen sucked out of the media landscape, Nintendo then announced that they would give the first look at the Switch console on the very same day. It was a nightmare week for any indie game to get noticed or talked about, let alone a small game from an unknown German studio and an Aussie-based games label.

Thankfully, the marketing plan worked. Polygon, Destructoid and PC Gamer all ran pieces on the day of the announcement and Rock Paper Shotgun, Gamespot and Killscreen ran pieces a few days later. The weekly approach was the headline lead for most of the stories, validating their hope that this would give it the extra push to be newsworthy enough to cut through.

One negative issue the publisher had anticipated turned out to be true. Many reviews ran very late with most appearing in December. While reviews were overall extremely positive, one Australian magazine chose to review the first two episodes in their print edition giving it a 4/10. Later, a different journalist gave the whole game a 9/10 in a review on the magazine’s website but Metacritic only lists the print score and that pulled down our average score significantly.

Throughout the season we continued to get coverage from smaller sites, YouTubers and Streamers, and some even ran weekly reviews of each episode. The larger sites didn’t end up covering it each week, but most came back for the final episode with a review or the news that the season had wrapped up. Unfortunately, we didn’t see journalists talking about it en masse each week on twitter, which we hoped might have happened for those that had got really into the game concept, but the weekly drop caused more general buzz with players.

Engagement in the forums started slow but grew each week. Fans were discussing other aspects in their own threads and responding positively to our weekly updates about each episode. Most rewarding for us was to hang out in the forums in the hours before a new episode launched and watch as fans would start to gather ready to play and discuss it.

One unexpected result with players was that we found they would replay each episode multiple times while they waited for the next episode to release. They’d try to get all the various outcomes in the episode and find all the achievements, helping each other to find the trickier combinations of choices.

During that time, we got plenty of steam reviews and players that didn’t wait until the whole season was out before reviewing it. The rating remained around or above the magic 95% positive rating for a number of weeks and still remains at 91% positive.

Sales were strong across the five weeks, spiking around episodes 1, 3 and 5. And now, more than a year later, Orwell: Keeping an Eye on You continues to be one of Surprise Attack Games’ strongest performing titles.

All of it proved to be a wonderful experience and we learnt a great deal about the episodic structure.

With Orwell: Ignorance is Strength (‘Season Two’) deep in development, it should come as no surprise that we’ve decided to do it all again.

Orwell Season Two will release across three complex episodes: Each episode will offer more choice on how to advance than before. The story will also feel more personal than the original, focusing in particular on three individuals and the complicated relationships they have with each other--and the truth. Return players should expect a more dynamic environment: based on the newly added “time of the day” characters may act on their own behalf and events unfold independently from the players‘ actions.

The player will be granted access to a more advanced build of Orwell. The old tools are all still there, useful in their own right. But the player will be invited to join “The Office”, a secret department of Orwell that has been equipped with tools that not only allow investigators to research individuals, but to also take that information and influence the way events are reported and perceived by the public. This gives the player more agency, not just in the way the story branches and consequences play out, but in the narrative itself--the story of Orwell: Ignorance is Strength.

When written out like this, it’s an exciting and daunting proposition. We had hoped to have the game out by now, but because of all this extra content--the story, the ideas, the narrative branches, the new tools--and, ironically, the way that episodic content comes together, we’ve focused on getting it all just the way we want it and delivering to our wonderful fans the very best game we can.

With an even stronger emphasis on narrative, decisions that alter the course of the investigation, tools that allows players to pry deeper into people’s lives and manipulate that information, and a thematic search for truth in a post-truth world, the opportunity for thrilling cliff-hangers and deep philosophical debate is plentiful. 

So, as this year comes to an end, we look forward to the next. We can’t wait to give our fans--and hopefully many new Orwell players--the opportunity to experience another unique surveillance thriller set in the world of Orwell. Week by week.

- Team Osmotic

The familiar seen from a new perspective. An early screenshot from Orwell: Ignorance is Strength.