pátek 21. srpna 2015

Postmortem of our Steam Greenlight campaign

Before we launched our Steam Greenlight campaign, we went through some studies about the approval process and a couple of blog posts about both - the successful and unsuccessful campaigns of other studios. What we found out is that there's no specific number of yes votes you need to be greenlit. Some games got approved with 1500 and some need even 10k or more. And as expected, everything was different with Grenade Madness also.

The preparation

This part took us most of the time - let's say 10 days of full-time work (there are only two of us in the team right now). We know that our game is not the most beautiful one in terms of graphics, so we decided to highlight some other aspects - e.g. the crazy action part of the game and the moments you don't even know what's happening, big explosions and other stuff.

We also tried to cut the trailer into very small cuts. As small as possible to make it as exciting and shocking as possible. 

Another part were the texts. Texts for our website and for the greenlight page of course. We need to make clear what the game is about because it was not very clear from the small cuts in the video and the screenshots.

What was very important to us was to let players know that we are ready to listen to the community and adjust the game according to the palyers feedback. But at the same time we couldn't just go out there and say "tell us what game we should make". This would look like we don't have a single concept in our heads. (And we have!).

Did we choose a wrong time?

Some people in the industry say, that even when your game sucks, you are given a wonderful comments on your Greenlight page, so you can only rely on the negative comments to have some real feedback. 

When you are a game developer and work on your own, it happens that you forget the summer holiday is starting...

We started the campaign on Sunday June 28th, 2015. From our experience with running several websites, we knew that on Sundays there's a significant traffic peak on the internet. But... This was one very bad choice because the summer holiday started two days before. Nor me or my colleague remembered (we both finished studies 2 years ago).

So from todays perspective we thing that we should have started it on Friday afternoon and about two months earlier or later. 

In the first few days we collected almost 460 votes and as our game fell into the second page the vote count stopped counting.

As you can see in the screenshot, we had very few unique visitors on the page. Does this have some effect in the algorithm? Who knows...

We left the page as it was almost for a month except a few announcements with gifs. And on July 28th the game was greenlit with only 486 "yes" votes.


A few days before the game was approved by the community, we released a demo version of the game. Maybe this had some effect?

We knew that we can't lose anything when we release an online version of the game. We quickly got some negative feedback from the Czech game dev community. This is of course normal. It motivated us to improve many aspects of the game.

The first day we submitted the game to kongregate, the demo was stolen by Russian and Turkish online game websites. This caused a very significant traffic growth to our Greenlight page and we finally had some numbers about the game itself. E.g. the average gameplay time.

The stats:

This demo had 3 different playable arenas. Each game was 5 minutes long, which was great for debugging during the development but extremely fast for real gameplay. Another mistake :/

The last statistics is a bounce rate in seconds.

0 - 60s -  As we found out later, a lot of players had problems with connecting to our servers. Also, the UI wasn't very user friendly and many players reported bugs in game menu. So those are probably the players who left immediately. 

61-180s - Players, who successfully connected to the game, but had no enemies to play with.Also, players with high latency.

181-600s - Players, who played one round and left. Probably because the game was boring. 

601-1081+s - Players who played 3 or more games. Thank you!

What is very interesting is that we got owned by others many times, when we tried to play the game ourselves.

We can't wait to release the next demo and compare the statistics. We hope it will be much better!

Things that we changed

  • So we got on Steam. Wow. That's something, so we decided to try our best to improve the gameplay.
  • We reduced the number of weapons slots from 5 to 3. It was too confusing.
  • We added the Equip menu, where players can chose weapons they want to play with in the next round. ( In the next demo you will still be able to change your equipment anytime you want ). Also the more experience points you have, the more weapons are unlocked.
  • We added bonus weapons with limited ammo. E.g. the rocket launcher.
  • We added a few more melee weapons for close fights like the baseball bat.
  • You can equip yourself with 3 bonuses before the start of the game.
  • Nové animace, interakce ragdollů s fyzikou.
  • New animations and ragdoll physics implementation
  • Complete UI redesign (thank you, players for telling us the game UI looked like a crappy website)
  • We simplified the game controls
  • We are working on more arenas so the game will be finally stylized.
  • The next demo will be released in September. 

This is how the game looks like now....

Equip menu & new HUD

Grenade Launcher in action

Ragdoll compilation

3 komentáře:

  1. zaujimavy clanok, len tak dalej hosi! :)

  2. Vdaka, ze ste sa podelili o zaujimave data a skusenosti. Hru som sice nehral, ale z videii to nevypada vobec zle, ked sa to poladi svojich hracov si to isto najde.

  3. Dík pánové, konkrétní data se vždy hodí a sám vím, jak těžký je se občas od něčeho odpíchnout :)