Decision traps in live game development

• InnoGames product manager Anna Sieprawska says developers should listen to their players, but try to understand what they are really asking for.
• Be careful when making decisions based on data and be sure to understand the context behind the numbers.
• “If you won’t make your game fun today, tomorrow doesn’t matter.”

Making a wrong decision for a game during live ops can lead to lost revenue, unhappy players, missed deadlines or even a game being cancelled.

Speaking during a session at Devcom entitled ‘Who Really Knows Best? Decision Traps in Live Game Development’, InnoGames product manager Anna Sieprawska offered a deep dive into the process behind making and iterating on game updates, events and new features.

Beware of loud player minority

Quoting another game designer, Sieprawska said “you should always listen to your players… but you should never make the game players are asking for”.

She referenced the most discussed forum thread on the Tribal Wars forum titled ‘Tribal Wars, A Dying Game?’. It has 227k views across 114 pages and was started in 2010.

“It’s a very, very slow death and I wish your game this kind of death,” she joked.

Sieprawska highlighted an issue when gathering player feedback - how many of your players are represented on social channels and in Discord, and how many of those actually engaging.

She said when players ask questions, it’s important to focus on player context and motivation to understand their actual problems.

Don’t trust data without context

Sieprawska showed a chart displaying ARPDAU going up in Elvenar overtime. While the data might seem positive, it could potentially reflect better revenue or the fact that there are fewer new players, with only dedicated users staying.

Sieprawska also highlighted a chart which showed the most popular buildings in the game. She said in some cases, if the team just looked at the data and thought certain buildings are good or are underused, it could have stumbled by stopping development of certain types or giving more things players don’t care about.

Is this game even fun?

Touching on the key topic of fun in a game, Sieprawska tied the motivation to spend money on virtual items with players having a fun experience, whether that’s to express themselves, continue playing or having a meaningful goal to achieve.

She quoted a colleague as stating: “If you won’t make your game fun today, tomorrow doesn’t matter.”

Sieprawska referenced ways to test fun, such as feedback from your target audience and other game designers and product managers, playtesting tools, and A/B testing.

Sieprawska closed the talk by stating that making a bad decision can be better than no decision - as long as you iterate. If a new feature needs work, it should still be worked on and not forgotten.