6 Feb 2012

Event Round Up - 28th Jan, Mobile Games - Part 2

In the second part of the round-up from our event of 28th Jan on Mobile Games, Steve Devo, @sdevo, offers his view of the proceedings.

Steve's been attending Mobile Monday London events since the beginning (November 2005), so we were particularly touched by him tweeting:

Valuable insights from the  panel last night, good to hear what does not work as well as what does. this is momolo at its best.

What follows are Steve's views, put more expansively than 140 characters allowed.

Once again, many thanks to our event partners, ICTKTN.

There is a thriving and vibrant mobile gaming industry in the UK,  and it is starting to have effects beyond gaming with the so called "gamification" and "socialisation" (you are allowed to smile or groan at this point) of applications starting to become more mainstream.  So let's take a look at the gaming market and some of the more prevalent aspects.

I have written this after attending a Mobile Monday London conference this week and combining that with my own experiences of talking with the mobile community here in the UK.

There are three parts: commercial, technical and buzzword (the third is,  you guessed it,  about those two concoctions mentioned above.)

Commercial Aspects

Freemium model has taken hold and is now the main driver of revenue.

Freemium is where a game or app is provided free to try,  and if you like it you pay (in app charging) to have access to the rest of the game,  or to buy upgrades in the game such as better weapons, extra levels, better cars etc.

It should be noted that when freemium is added after a game has been built it nearly always leads to a poor user experience,  therefore "in app upgrading" must be built into the design.

Lots of games cross sell/advertise the publisher's other games,  and analytics is starting to be used to maximise these opportunities.  We are starting to see that games will send back usage and activity data to the server to drive this analysis, indeed this is critical to success.

Traditional app developers should consider a more explicit freemium model in their big tools, the ability to buy access to advanced functions as and when I as a user need them would make the management of license fees far more acceptable, especially in the mid market.  This also support the gamification of apps - more of which below.

Discoverability is a big issue,  because there are 380k and 500k apps in the android and apple app stores,  and they are averaging 100 new games per day.  You can pay to be in the Apple preferred top 25 lists or selected lists - however this is very expensive and if the game is not a natural hit it will die quickly. Obviously cost of getting those people back is much higher again, so make sure game is ready before you attempt this type of marketing.

Conversely it is wise to get the game out early and see what the user reaction and usage patterns are.  This is aided with freemium as you can add new sections and better replacements later (if the app is able to load modules) or use the upgrade functions of the platforms to send new better versions to the users.

Yoshi Monsters spent 10 of their £11M funding getting the game of the ground,  that's not good enough and way beyond the budgets of most game publishers so the Games industry investors are now looking to see that a publisher has not just build and creative skills,  but also analytical marketing skills.

Acquisition of these analysis skills is a problem for a lot of these smaller companies, but it is one of the ways that some small publishers are rising above the crowd and solving the discoverability problem. 

A further use of the analytics is to alter the game design as well as the price of the premium parts,  thus parts of the game move in and out of the premium sections, and the costs change in response to usage, including the equivalent of abandoned carts within the game.  Clever stuff,  you might say this is taking scrum/agile methods into the post launch period,  and that is what is so smart. You could say this is games companies version of "live life in beta".

The inability to get your game noticed without being clever and/or having some decent wads of cash meant that out of the 248 mobile game startups in England last year (most in London) 135 died within the year.  This is about double* the usual attrition rate. (* having trouble confirming this statement though, so grab some salt)

Copycat games are a problem because a successful game will get copied ad infinitum.

There is still money to be made from Java games,  but this is only in the growth markets rather than the Western markets where smartphones are dominating.


But what is the income split for a game? Well typically  it looks like this:

  • 20% advertising- other people advertising within your game/site
  • 20% is from the game - purchase of the game, remember you give a lot of this revenue away to the store owner, hence this may be lower than you think.
  • 60% is digital goods - upgrades and additions -  e.g. in a zoo game you might buy extra ground, a cafe, more exotic animals etc.

You'll notice there is no product placement revenue in that list,  this has not worked on the console gaming world because the amount of effort required to manage the relationships / partnerships is too great for the small amount of income it generates.  Also gamers are quite a harsh audience and decry any placement that are too obvious or get in the way. It is expected that these issues will not change with a switch to a mobile platform.

Formats, technology and the like

Smartphones are mostly for casual games,  that is games that are just for fun,  there are very few successful mainstream games on these devices.

The tablets however are for more "serious" or hardcore games and the new ranges of tablets have the processing power (processor and graphics engines) to properly take on hardcore games. BUT there is a real challenge in moving a game from a console/PC to a touchscreen and that's because of the controls.  

There is a triangle of controllers: buttons/joystick for consoles/PC, movement for consoles (Wii, etc) and touch.

Moving a game between these controller types without this change being built in from the start has not been successful yet.  Also games scaled up from a phone to a tablet tend to have the controls in the wrong place and too far apart,  needs to be thought through.

TV's & Multi-device: As yet the game companies do not see a great deal of use for a game that spans several devices _and_ a TV.  I think this is missing a trick, playing a board game like monopoly, scrabble, etc,  or other turn based games could benefit from a large shared screen,  and then local hand held screens.  On the negative side a lot of casual gaming is played at the same time as watching TV.
For software vendors it would be good if these guys could consider dual screen/device working for some tools.  e.g. being able to use a subset of RSA or Doors etc on a tablet when working away from the desk, or even in presentations, meeting  makes a lot of sense to me.  Multi-person edit sessions that go beyond screen sharing makes sense too,  life is more about collaboration these days,  yet we are more likely to be in physically separate spaces.

HTML5 is not ready for gaming (yet).  Games need fast reactions to complex series of inputs. HTML5 does not cut the mustard.  And then there is the old saw of fragmentation.  HTML5 supposedly will fix this by having the browser act as middleware between the HTML5 and the device.  However this is dependent on the browsers acting consistently and they never do.

For an iOS game you're likely to produce 4 different versions, and for an Android about 6 perhaps as many as 10 in some markets.  Compared to the bad old days of Java when 800 to 2000 versions (skus) was not uncommon - this is vastly improved.

Tools: So how does one build apps for multiple platforms, remembering that Android is Java and iOS is objective C?

Well the answer is either two build efforts or the use of some clever strategy.  Some folks simply use C++ for the core elements of the game (physics engine etc) and then can port this to the devices quite easily.  The libraries that deal with device specific elements (camera, input, accelerometer and other gadgets) are written per platform.

There are some tools for writing once and building to multiple targets such as unity and marmalade and IBM's newly bought Worklight.

It seems that opinion is split on these tools.  Generally if you are stressing the device with a lot of work then you need to hone the code by hand,  if you're more  a turns-based casual game (eg Sudoku) then these tools are ok. But be careful. Here be demons.

Artificial Intelligence: A lot of platform games compete with the quality of the AI that controls the game.  This is becoming true with mobile games,  and the AIs are starting to fit into the constraints of the mobile.

Siri has sparked a lot of interest because it is not just a voice recognition AI,  it has a personality all of its own and that is what intrigues gamers.  Expect to see this kind of system in use in games quite soon.  
I suspect that we will start to see eye tracking soon,  then you can sneak things onto the screen whilst the user's gaze is elsewhere,  that kind of input into the AI will make a big difference.

Sadly you'll have not noticed the BlackBerry or Nokia/Windows phones in anything above.  these are just not working for games.  Yes a _lot_ of kiddies have BlackBerrys leading to these devices being the biggest seller last year,  however they are not set-up for gaming,  and a lot of the devices go to corps where games would be frowned upon or banned.

Nokia/Windows are just not a big enough market to justify the costs of entry, and the way things are going this is likely to remain the case for some time to come.

Some horrible buzz words that are emerging.

Gamification: (I did warn you) this is the use of game type techniques in apps to gain traction and maintain interest.  A Lot of apps,  and by that I mean the vast majority of apps have a short usage life on a device.  People download,  have a short flirtation with the app, and then it just kinda lurks on your screen unloved and unused.  there are figures for this and they are scary.

According to a Harris online survey on behalf of mobile ad provider Pontiflex, only 3 percent of apps downloaded through incentivised install campaigns are used frequently, and only 62% of downloaders use an app more than once.  Given this challenge and the popularity of games what can a game teach us to make an app more likely to be used.

Well initially people used "badges and leader boards" like Foursquare did,  and this can help but is a little cheesy to say the least.

What games do to increase usage is, I think, the following: -

  • They take you through a period of learning,  where you are introduced to more and more parts of the interface and the application.  This prevents the user being overwhelmed. 
  • Then they discover what you want to do,  and adjust accordingly,  and with the freemium model this means targeting the best upgrades / extensions at you.  
  • Then they let you explore and move around the game as you like.
Seems to me that this is the way to go with other types of apps,  understand the user's journey from wide eyed innocent through knowledgeable user to comfortable expert.  Forget the cheese.

Next up games use incentives that appeal to the user at a quite simple level,  often involving completing something and feeling good about it.  We can see that web sites are using short, chunked up, incentives to get folks to fill in their profile data,  for example Linked-in and Yahoo! properties do this, encouraging you to get 100% on your profile,  but not all at once,  they will ask you for more as you progress through your journey with them,  and will reward you with encouragement and kind words,  and, yes, sometimes a little competition, especially for men.

Socialification (it's worse then gamification,  and that's saying something) simply put, this is the use of social networks and interactions into an app.  

Games are starting to use presence in an asynchronous way,  that is knowing about other players without necessarily playing against them.  This provides a sense of belonging and with care can become an incentive to return to the game.

Multiplayer is not social networking,  get over it.  OK?  

Thanks again, Steve, for this really interesting piece!