Teamwork! - The operations and marketing teams were seated quite close together, so this allowed for simplified collaboration and easy communication within our company, and the developer was easy to contact. We were very comfortable with each other, which can sometimes be difficult in larger companies, and after the first few brainstorming sessions everyone settled into their grooves and opinions came more freely. Compromise was a bit difficult at the beginning, as we have some very hard-headed people on the team, but balance was eventually found here as well.

Comprehensive Testing – We went balls to the walls with our testing, doing both the open and closed beta over five months. We did the regular Google Play tests, trying out a couple of icons and screenshots. The icon we thought would succeed did, which was reassuring, and we ended up with some good screenshots. We also did a lot of balance testing. The developer’s original idea was to make cards as powerful as they would be if they existed in the real world. This meant that god cards were extremely powerful, to the extent that five common cards would still probably not be able to defeat one of them. This, unfortunately, left us with more problems than benefits. Heimdall, for example, could block the first three points of damage. Unless a player had just the right cards, such as an instant kill or something with extremely high attack, there was virtually no way to defeat him. The testing led to a lot of changes, and we ended up with a better game because of it. Finally, we did a lot of placement testing for our UI. Chinese players have a much different idea of how everything should appear on the screen, and through playing around with different variations we eventually ended up with something that players found simple and intuitive.

Community Management – We had a really good community throughout the beta. There were a couple of little bumps, but overall they were extremely fond of the game and helped us out a lot. Our main community contacts were Howard and myself, and players seemed to get along well with both of us. The main theme for the beta was ‘Become a Designer’. We kept track of which players were most involved, gave the most and best suggestions, and helped us find the most bugs, and then we acknowledged their efforts both in-game and on the website. We ended up with seven designers, and made sure that they got extra rewards, as well as a place to gather together and share their thoughts. Our social media was also a priority. We made sure we posted at least once a day throughout the betas, and had several weekly contests on Facebook where players could get extra rewards. Freebie Friday, where we would have players complete random tasks (write a poem about Valhalla Lost, send a meme, link a Viking song, etc.)  was the most popular. Strategy Sunday, where players could discuss some strategies we suggested, was also pretty successful. Our final effort had a bit of a sad tone, but one player in particular really appreciated it. After the November terrorist attacks in Paris we decided to make a card to show moral support for the people of France. With the help of one of our most dedicated players, himself a French national, we designed a card called Fraternity. We were a little afraid that the effort might come off as a bit disingenuous but in the end the players seemed to appreciate it.

Experimentation – We tried out a couple of new programs before the closed beta launch, one for PR and one for marketing. Prelaunch.me is an Android app that lets players register to be notified when a game is launched. Ultimately the CPA was about average, but the player involvement was awesome. We also tested out Pitchbox, a system that helped us send pitch emails to different media sources. This ended up being a bit of a disaster, which I’ll go into later, but the willingness to experiment with this program let us to learn a lot about media, and we got quite a few contacts out of it, so it paid off more or less.

Art – This wasn’t really us, it was the developer and their freelance artists, but it’s worth mentioning. There were hundreds of painstakingly drawn images of gods, spells, monsters and more. They put a hell of a lot of effort into those cards. 

Scope: Too big, too soon – We had huge ambitions for Valhalla Lost, but we got a bit of ahead of ourselves which ultimately did more harm than good. We wanted to test with hardcore CCG players, but instead of choosing a select group for the soft-launch we decided to launch in several countries at the same time. This led to a lot of players trying the game once, then abandoning it, never to return again. The reasons for this were obvious towards the end, we ended up getting a lot of players that weren’t as hardcore as we had been expecting, and weren’t that familiar with CCGs. We had expected our first round of players to understand the basics of games like Magic, or Hearthstone, and so they would be able to pick up and play Valhalla Lost with minimal difficulty. This assumption led us to not creating a good enough tutorial. We also prioritized new content, introducing new PvE levels and world bosses, over fixing bugs. The intent had been to keep the most dedicated players interested, but the bugs eventually pushed them away.

Designing in a way that was more appropriate for PC – As I said before, our initial target audience was hardcore CCG players. This is a great strategy in China, where the average amount of time for a single mobile session can be 20-30 minutes. Unfortunately, western audience tend to have much shorter sessions on mobile. A single game in Valhalla Lost could take up to 15 minutes, whereas most western mobile sessions cap at about seven minutes. We probably would have been better off doing one of two things 1) Making the first version for Steam. 2) Starting with less mechanics and making games quicker. We were afraid, however, that players had already become accustomed to CCGs and would get bored without enough options. In the end the exact opposite was true. Mobile also has a lot more limited real estate, which led to some problems with the menus and UI. These were eventually figured out, but it hurt us at the beginning.

PR – We made a few mistakes with PR that cost us pretty dearly later on. First off, we definitely could have afforded to get help from a PR company, and have since done so, (check out Novy Unlimited if you’re looking for one) but we wanted to test the waters for ourselves. We were hoping to get as much exposure as possible, so after making our press kit and getting our materials together, we started contacting a ridiculous amount of media. It would have been safer, and likely more effective, to just contact a few sources that we knew would be interested, but our eyes were bigger than our stomachs. We also didn’t take advantage of all the platforms we could. We should have created our own subreddit, for example, to go along with our social media and forum. Finally, our community building didn’t start until right before the beta. This wasn’t anyone’s fault, the game was already ready for beta when we finally got a contract signed, but it would have been good for us to plan the timing more appropriately.

Launching overseas first – Valhalla Lost is a Chinese-made game, which was localized to the west. The localization ended up not being as strong as it could have been, however. There were quite a few spelling mistakes even in later versions of the beta, as well as some parts of the story that should have been changed. Same thing with the UI. It probably would have been better to perfect the game on our home turf before launching it overseas. For now it is being optimized here, and we will try again with the English version later.

PVE over PVP – In the earliest parts of the beta we tried to push PvP, but there weren’t enough players yet. This led to quite a few players leaving, and so we decided to concentrate more on the PvE to keep these remaining players engaged. We should have created some sort of alternative for when there weren’t enough players. Instead we ending up adding content which was fun, but not really what people wanted.