Endless – Wrap Up
So Endless my first game developed in Unity ended up with about 100 downloads, heavily weighted towards iOS (80/20 split). That was off the back of about £35 worth of advertising on Facebook.
I learnt a huge amount from it and decided to use that experience to focus on my 2nd game Mozzie Squash.
Mozzie Squash – Beginning
The game concept is simple enough, mozzies appear on the screen and you have to squash them.
Initial testing focused on challenge modes, squash 100 mozzies as quickly as possible. Testing with friends soon established that mobile phones and the internet have indeed ruined concentration and 100 was far too many. So 50 became the default mode.
I settled on 3 game modes, unwilling to let 100 or the 2 min challenge go I kept them in. After more testing with friends I gathered feedback that actually levelling up and progressing was something that would keep them coming back for more so the “challenge” modes became secondary to a level based game whereby you squash 30 mozzies before 5 escape the screen. As you progress the mozzies get quicker, more agile and more likely to escape the screen. After the final round of testing I opted to add in a “boss” mozzie every 5 levels that took more than 1 squish to squash in order to break the game up and keep it interesting.
All this goes to show the value of getting as much feedback as possible from wherever you can to help shape the design of the game and avoid building a game that is fun only for you.
The Techie Bits
Game design was relatively simple, some tricky bits like keeping the mozzies on screen until they’re ready to escape but keeping them squashable. Good use of the Unity physics engine and layers actually made that relatively simple in the end.
Trying to make sure the mozzies actually appeared on the screen (they start off outside the screen boundaries) and not whizz off never to be seen was another fun challenge.
This time around I made sure I added in the facebook unity sdk so I could easily track how well advertising was going, turned out to be really easy and added a huge amount of value to tracking the statistics of the game.
I did my first in app purchase to remove adverts, again with unity’s cross platform engine it actually turned out to be relatively simple to configure and set up.
Lastly to make the game more social I used leaderboards on both google and apple. This was a bit more tricky, it’s not quite cross platform so I’ve ended up creating duplicate projects to manage the code better. It has to be said that apple/unity win the day here and make things easier to set up, the only tricky part was submitting a time based score. Google requires an additional SDK but once set up it’s relatively easy to work with.
The Business Bits
For endless I did an epic 30 second advert that I think was fairly witty and showed a little of the game, I was hoping to entice people with humour to download and play the game.
It didn’t really work, 80 downloads off the back of £35 worth of advertising or 44 pence per install (worth remembering that figure). Out of the 25000 people that had the advert served up for viewing just 71 watched the entire thing or 0.2%.
For Mozzie Squash I did some more research on what worked for facebook advertising;
- Shorter videos work better, 6-15 seconds, ideally 6 seconds. I settled on 8 seconds to get as much exposure as possible across Facebook and Instagram.
- Ratio matters, you want to optimise for Facebook and Instagram, last time I uploaded a traditional HD (1920×1080) format. People viewing on their phones use portrait, the aforementioned HD format sucks. This time I went with 864×1080 which gives you a 4:5 ratio that can be used across most Facebook and Instagram locations.
- Gameplay took up most of the advert (unlike last time) and finished with a picture of the squashed mozzie and told people to download the game.
- I paid £30 for Adobe Premiere (monthly cost that can be cancelled). To do the advert. It was worth it!
The new advert, compared to the old one was a success in every way, I opted for a daily budget of £10 to run for 10 days. I actually paused at 2 1/2 days down to retention rates (more on that later)
- Getting the aspect ratio right meant that for £25 my advert was displayed 48,109 times as opposed to the endless advert at 24,605 times. A huge improvement.
- A full play of the advert stood at 6.3% as opposed to 0.2% for Endless.
- Cost per click was down from 23 pence for Endless to 5 pence for Mozzie Squash.
- Post reactions 100% improvement, post comments 200% improvement.
- Cost per install 9 pence, down from 44 pence!
Like I said it was a much bigger success in every single way.
So why did I stop the advertising at £25? Well it turns out that there are some other statistics that are really important for a game to be a success from a business point of view and that is essentially game time (or retention rate).
If it’s costing me 9 pence per install I need the people that are playing it to keep playing and either see a lot of adverts (a lot!) or pay to remove them. For every person that downloads the game I need a good % of them to be playing the next day. Mozzie Squash is currently somewhere north of 10%. Really it needs to be somewhere closer to 40%. So that’s it advertising cut for the moment!
Lessons Learned (or learnt, apparently both correct) and Future Plans
Overall I am absolutely delighted at how much improvement there has been from Endless to Mozzie Squash. The game itself, the technical improvement and the marketing improvement are all big wins! 4x-5x the number of people playing the game (at less cost) is amazing. Although low for a commercial success the retention rates are far better as well.
Mozzie Squash has turned out to be a good enough game that I’m going to spend some time improving it and see if I can get those pesky retention rates close to where I need. So it’s back into the testing cycle with friends and family.
So what are my takeaways from my 2nd game.
- Commercial success is still a way off.
- Advertising, get the format right and shorter is better!
- If you’re going to use Facebook advertising get the SDK in the game early it’s invaluable.
- Listen and watch people play for feedback. I should have ditched the 100 challenge and 2 min challenge modes, no one is playing them! I will be removing them from the next update.
- I spent too much time thinking I need to build up a Facebook, Instagram and Twitter presence. My bro gave me the advice of relax on it and build the games first. On day 1 of my android release I got 82 downloads. Only 1 came from a social media post! I’ll keep my accounts active but post less regularly.
- Lastly plan the game better, I don’t spend a lot of time planning I just jump straight in and see what works. I’m never going to be the best planner in the world but I can improve. I’ve been reading Scott Rogers book Level Up which is well worth a read if you want to get into game design.