Greenlight in 16 days Post mortem: What worked well, what didn’t
Last February 10th we excitedly launched a Greenlight campaign in order for Wincars Racer, our game in which a team of 8 people have been working for two years, to be released on Steam.
Wincars Racer is a fast-paced online arcade racer game which pays homage to classic titles like OutRun, Burn Out, Mario Kart, Sega Rally, where strategy also plays a role. The aleatory item system is dropped in favour of the playera��s ability behind the wheel and the use and configuration of the special skill that he/she chooses.
Before reading this article I advise you to have a look at the campaign to have a better understanding.
Everything went ok
Not that I thought this would go wrong for us or that our game didna��t deserve to be greenlit by the community, but truth is the result was remarkably good achieving it on February 25th, ending 24th place after being in the Top 100 nearly through all the two weeks.
After reading so many articles in Gamasutra while I was preparing my campaign, I though it was fair to analyze it publicly and draw some useful conclusions.
I will focus on what worked
We went for an approach as similar to a pitch elevator as possible. Then, a hook phrase which defines your game, and at the same time explain what makes it special. After this, a list of main features, a link to play the demo (I will talk about this later on) and a flashy little section for press mentions and awards.
Filling the description with images, GIFs and headings for each section has been a good idea for games which have done well on Greenlight.
We went for something simpler and we saved a lot of time to be used on other tasks.
Same approach as before. We used the best six images we had and it was enough.
Valve makes you use a minimum of four images and we had many more available in the press kit but we want to go straight to the point. Most of the players who check Greenlight devote a very small amount of time to each game so it is best to highlight your best features: a�?Therea��s no second chance to make a good first impressiona�?
Choosing your branding image
We devoted more time to this task and we think that the number of visits increased for this reason. We made an animated GIF with some of the most spectacular moments of the game which were representative of the gameplay and we also interspersed the logo.
Although Greenlight lets you upload a bigger GIF we decided to optimize at 150 x 150 pixels which is the size shown in the page where you compete with other games to get peoplea��s attention. In your pagea��s profile the resolution will look like crap but thata��s where people will mainly pay attention to the video and the images.
Views and votes over time
The graph says it all.A� From day 21 the organic traffic practically dies. The upturn from day 15 shows our maximum effort giving it visibility in the threads we had opened on forums as well as having a bigger activity on Twitter, Facebook groups, etca��
It is clear that the best thing to do is to try to concentrate your resources on a good launch and letting Valvea��s system keep you visible for as long as possible. Never try a long term a�?leaka�? campaign as you will lose most of your options.
Today, an independent studio launching a Greenlight campaign is no big news. It is very difficult to get the media publishing your press release unless the game is extremely polemic or original.
Our plan was to launch an open Beta version of the game the same day we kicked off the Greenlight campaign.
The mail we sent had the following subject:
Wincars Racer launches a campaign on Greenlight and announces open Beta
A much more appealing headline in order to get a journalist, a youtuber or any streamer interested in your game and publish a post or record a video about it, which on top of that means added value giving their readers and followers the chance to play and try the game.
Some examples of pieces and articles written about Wincars Racer:
Some publications like IGN Spain directly titled a�?Wincars Racer launches its open Betaa�? and talked about and recommended the greenlight inside the article.
Ita��s always advisable to write each email individually, but if you cana��t afford to do this for lack of time there are great tools which will help you.
We used Mailchimp to send these automated press statements. A great tool which guarantees your mails will reach their destination.A� It doesna��t make spam filters go off and provides data about who actually opens your mails and even the links which they click on. If your emails are not opened you can try sharpening your aim and drawing conclusions about what you are doing wrong, it has A/B tests integrated in the mails.
However, do not forget there are people at the other end opening mails. Personalize them the most you can, with the recipienta��s name and surname, personally answer each journalist who writes to you and make things as easy and straightforward for them as possible. Dona��t forget having a press kit ready and include it in your mail.
Dona��t worry if there are people who open your email but finally dona��t post anything about your game. This can be due to a thousand factors and not because you did something wrong or your game is not attractive enough (you will be aware of this in many other different ways).
Journalists usually follow many games until they decide that ita��s time to talk about them or that they fit into their publishing agendas. Having them periodically updated with the most important events relating the development of your game will help to make yourself a name and it can be the first step to build a stronger personal relationship for the future.
Mailings for players who signed up at our site
Since we started our development blog we added an option so those players who wished so could join a Wincars Racer mailing list. In exchange we offered them exclusive content like early access to Beta Tests and invitations to participate in tournaments where we had prize draws and we gave away gaming helmets, t-shirts, etca��
We have more than 3000 registered players and the statistics say that more than 600 opened the email the same day we sent it, and that 200 clicked on the Greenlight link. We can assume that means 200 positive votes, cana��t we?
Besides, many players told us they had received the email but hadna��t had the chance to vote on greenlight in that precise moment, and that they would do so later.A� For this reason we assume that the number of positive votes coming from the mailing list is substantially bigger.
Google Analytics: Where the visits came from
Once you launch your campaign on Greenlight you get access to a basic analytics site on Steam. If you want more detailed information it also allows you to synchronize a Google Analytics account. I recommend to do this because it gives a clearer view about where do people coming to your page come from. It will also help you to make future decisions about how to approach your marketing strategy for the game, the launching or even to decide into what languages should the game be translated.
These are the final statistics according to Steam:
And these are the Google data:
As you can see theya��re not exactly the same but dona��t give it more importance than it has, this is due to the different criteria used to determine which user is a unique user and which one is not.
Do not worry either about the rebound rate, conversion, or the ratio of first time users vs users who come back. These data can be very useful for your sitea��s Analytics but not for a greenlight.
TheA�image above is indeed relevantA�if you are thinking about what countries or languages should your launch campaign should be focused on, even more so if your game as strong online approach as ours, as it can also help you to optimize your servers.
This graph also shows a very important fact:
Up to this moment I had read in other post mortems that the usual is for 70-80% of your traffic to come from Steam itself. Although the graph says 30% comes from Steam, most of 53% of direct traffic also corresponds to clicks on the game icon on Steam. Crosschecking these data with those from Mailchimp regarding people who clicked on the link in the emails we sent (which is also included in direct traffic) we could deduct that 60-70% of our traffic came from Steam.
The referrals graph is a bit disappointing. Despite being featured and recommended on many sites, the number of visits coming from thereA�is not very big. Although the sum of all of them has been of great help spreading the word (adding the ones we dona��t know about, disguised among direct traffic).
Social networks, mainly Twitter and Facebook also helped with 500 visits, a 5% of the total.
Building a community around our game from the beginning of its development was the key. Keeping and taking good care of the relation with journalists was also helpful in order to get more visibility, but not so much as we expected, I think that this would be different if we were discussing a game release and not a greenlight campaign.
The positive/negative votes ratio
First, I would like you to have a look at the ratio between positive and negative votes.
This is the ratio we had 4 hours after launching our Greenlight:
It began decreasing 8 hours later
This is what it looked like 24 hours after the launching
Data after a week
Finally the ratio we ended up with when the game got greenlit
We had a yes/no ratio close to 50% among the players who got to know our game for the first time through Steam Greenlight.
However, we managed to increase this percentage to our favour thanks to all the votes we received during the first days from our followers, who got to the site through our emails and social networks as well as from articles published in the press.
The site Green DB, as well as indicating your rank position on Greenlight, also awards a maximum of 5 stars to each campaign. Ours got 4 stars from the beginning to the end.
No one knows exactly how does Steam work in order to decide how many people it will show your Greenlight to, but I suspect having a ratio above the average was beneficial in order to get thousands of extra organic traffic.
Trying to achieve a good initial a�?momentuma�? can be the key to manage to prolong those first days of maximum exposure.
Events and Trade Shows
I assume everyone knows already how important it is to attend fairs and to show your game to the public first hand.
We couldna��t, in economic terms, afford to attend fairs which are held abroad, but we did take the chance to make our greenlight coincide with an event which was going to be celebrated in our city, The Japan Weekend. We also organized our own event at TelefA?nicaa��s Flagship Store as a release act for the open Beta and the Greenlight where we invited the press.
These events allowed us to win more followers in our web, positive votes on Greenlight, plenty of photographic material to share on social networks and above all to have the chance of having the press play our game and solve their questions.
This is not an action to be synchronized with the first day of your launching on Greenlight (you will have too much work to pay attention to and preparing for a fair is a hell of a lot of work), but without a doubt it helped prolonging the exposure in the press and social networks.
Finally, as highlighted by Thomas Brush, the creator of Pinstripe, in his article a�?Why your Kickstarter Will Faila�?, attending fairs will help you establishing new relationships with the media and other videogame creators besides reinforcing the ones that already exist.
As a general rule, only show the best about your game.
I know how great it is to have lots of content, but if you have to choose, do not hesitate to go for quality over quantity. This is not Kickstarter and therea��s no need to show everything you have to justify the economic support. You better focus on making a good impression shooting your best shot in a clear, simple and straight way.
What didna��t work:
We didna��t manage to appear in almost any of the big videogame publications.
It may be useful in order to get the media to know and follow you, but we didna��t get a lot of visits through the media that covered our campaign.
We hardly attracted any traffic from Reddit. My advise is that if your not regular users of some successful subreddit, dona��t even try it. Youa��ll be marked by the punishing cross of Spam.
Optimizing the Greenlight page to the max (video, GIF and texts) to keep a good yes/no ratio.
Building and keeping a good community of followers from the beginning of the development was the key.
Concentrating all the most important marketing actions in one day to create momentum to achieve exposure rocketing organic traffic.
As with the community, working on building quality personal relations with journalists, streamers and fellow developers.