Don't miss
  • 2,232
  • 6,844
  • 6097
  • 134

Guest post: How we publish an online game

By on September 28, 2010

As part of the rebrand of GAMESbrief (coming soon), I’m actively looking for guest posts on interesting topics. This post by Phil Stuart, Creative Director of Preloaded, originally appeared on the Preloaded blog.

Over the past few years we’ve learned a lot about how to design casual games that large numbers of people play and enjoy. Having just read the excellent ‘How to Publish a Game‘, I thought I’d share some of the tactical thinking that goes into releasing our games online and our approach to reaching the largest audience possible.

Traffic peak following Newgrounds upload

Seed your game smartly

A broad understanding of what role specific sites play within the online gaming community is important in providing the right information, to the right people at the right time. Knowing who wants a press release before launch, who wants to play an early Beta or who wants the game live before they even touch it, comes from experience and a large dose of trial and error.

Create a matrix

We have a matrix of every major games portal (about 100) which lists, amongst other things, audience demographic (often unqualified), upload/submission policies, submission links, revenue models, log-in details, max file sizes, and game and thumbnail dimensions. This matrix forms the basis for a bespoke seeding plan that is linked to the game’s genre and target audience.

Preloaded's seeding matrix

Example seeding plan

Having all this information on hand has huge benefits for the end product. Knowing, for example, the largest file size Newgrounds will accept or having a list of all the thumbnails sizes required by the major portals means you can make informed decisions during production that can increase the potential audience and save you time and money.

Time poor seeding

If you are limited in time, choosing key sites to seed and upload your game to is the best approach. In our experience, Jayisgames and RPS carry the best editorial coverage of online games, whilst Newgrounds, Kongregate and MiniClip have the best active community for game plays and feedback. Once your game appears on these sites, it will begin to seed itself.
Paid for placement exsists on some key sites but from our experience the fee can be negotiated if you have a good game.

Integrate with Portals

MiniClip and Kongregate each have their own APIs which they use for their scoring and achievement systems. The documentation is decent in both cases (Miniclipedia, Kongregate API) and are relatively simple to implement. In our experience, doing this goes a long way in building relationships with the portals and can often help when negotiating placement and editorial coverage.

Flomo with Kongregate API integration

SuperMe’s FloMo with Kongregate API integration

All our Channel 4 games integrate with the Kongregate API. You can view Trafalgar Origin’s implementation here.

Adapt to the environment

By the time you’ve implemented the API and adjusted the game to conform to the specfic portal’s terms of service (logo placement, click through policy etc.) you can end up with quite a few permutations of the game. Maintaining one file is crucial for our development, so all functionality is developed as modules which can be easily turned on or off – the game simply checks the host domain and configures itself accordingly. Its takes a bit of planning but makes a big difference when releasing updates.

Work with the ‘Pirates’

A large number of plays will come from ripped versions of the game, unofficially embedded on third party sites. This can be mitigated against but our strategy has always been to embrace this ‘pirating’, using it to increase the game’s reach.

Traffic with notes

Game traffic showing ripple effect of ‘pirating’

To be pirate-friendly, our games often end up as one self-contained SWF file. It can make large files but makes the ripping and embedding process easier. 1066 and Trafalgar Origins are just over 10mb. All necessary logos and branding are also included within the game, along with any instructions. In this context, any information on the official site obviously becomes useless.

Listen to your players

It may sound obvious but seeding your new game on launch day is a little unwise. The only thing worse than discovering that odd bug is having to deploy a new release to 100 game portals. Setting aside time for reactive tweaks in the first few weeks is a very important strategy in building a large user-base. Listening to a gaming community breeds interest and ultimately loyalty. Typically we spend about 5% of the budget post-launch.

Trafalgar Origins feedback on Kongregate

Feedback for Trafalgar Origins on Kongregate

Nothing beats the directness of an online gaming community’s feedback and Kongregate has by far the most passionate. In the case of Trafalgar Origins, we changed amongst other things, the way the camera followed the boat after a group of players complained about the system – A simple toggle system now exists for those players in Flash’s right click menu.

Community-led features in Trafalgar Origins‘ right-click menu

As an aside, when you are making iterative releases, tracking versions is incredibly important. All our games carry version numbers in this same menu. This becomes particularly useful as ripped games get out of sync very quickly with the most recent release.

Analytics rule

Releasing a game gives you a rare opportunity to find out first-hand about your audience. I find the process of identifying what you want to learn and then tracking usage accordingly one of the most exciting parts of publishing a game.

In-game tracking is the only way to track a portable game. We use Google Analytics but also wire-in our client’s preferred tracking system. A key requirement of portable games is to track the ‘host’. Big stats are great, but knowing where in the world the game is being played is incredibly important to our clients. It also feeds directly back into our constantly evolving seeding matrix.

Tracking hosts in Google Analytics

Tracking hosts in Google Analytics

Learning from our mistakes is a core part of this process. Analysing the multiplayer system in 1066 and making changes to it’s implementation has increased multiplayer game plays from 20% in 1066 to 40% in Trafalgar Origins.

Be SEO friendly

In our experience, only a fraction of game traffic (3% – 1066, 1% – Trafalgar Orgins) arrives through search engines. However, maintaining a primary SEO’d destination for all our games is very important to us and our clients, particularly if there are many versions of the game in existence online.

Google's results for Channel 4's 1066
Google’s results for Channel 4’s 1066

Alongside good semantic mark-up, copywriting is very important. I recently read a brilliant post about copywriting game descriptions – Highly recommended and easily transferable to SEO. We also include thumbnails for Facebook shares and a Fav Icon. This level of detail makes a big difference to a very small amount of people (namely us).

‘Big name’ sites linking directly to your SEO’d page will increase your game’s visibility in search engines. Traditional PR plays an very important role in this respect, making sure the high volume sites (ie Guardian Games Blog) get to see the game and hopefully link back accordingly. 


A publishing strategy is an integral part of the overall game proposition. Considering who will distribute the game and how the audience will find it is vital in realising it’s potential.

Syndication remains the simplest way of reaching a mass audience, but brings with it a community that needs listening to. Managing this ‘conversation’ takes time and is very public – something which some clients (read brands) rightly struggle with (*). Syndicated content is also often monetised by the portals, creating an imperfect relationship between publisher and platform. The trade off for mass audience is often a lack of purity in the branded experience and a notional loss of control.

It goes without saying the above won’t work for all games (or clients). The challenge is assembling the correct configuration to successfully target the audience whilst balancing the needs of the brand or client. We certainly don’t have all the answers, just some great clients who like to try new things.


(*) Kongregate offer a closed testing service in which they give their highest ranked players the opportunity to play and feedback on unreleased games but we’ve yet to try it.

Update 16 August 2010:
At the moment we are getting a feel for how successful click-throughs from syndicated games can be. We will post results in the next few months.

About Phil Stuart