Validating Your Game Design

My notes on the talk Validating Your Game Design: Ways to confirm the potential of your game idea before launch by Oscar Clark

Presentation mostly Targeted to Live games
Some questionable design patterns, in Validating Your Game Design#Mechanics for example

Direction

What success looks like?

Pick 3, and put number on them! => KPI/Metrics
It can be:

What are you making

Understand you audience

3 Personas

=> Targetable audience
Everything needs to be build in the design

Design

Loops

Mechanics

Think about the emotional journey of the player

  1. Start condition
    Does the player enter the game in a relaxed state? Does he enter the game with anticipation
  2. Challenge
    Are we introducing a sense of tension?
  3. Resolution
    Move away from anticipation toward FOMO
  4. Reward
    => Relief, but without satisfying the player fully so he continue to engage
    Keep player in the FOMO, even when we give the reward

Context

Metagame

Not the game
What happen outside of the game

Extensible interaction

Interaction that can

Art/IP Fit

Is the art suitable for the IP

Game economy

This encompasses everything, like
Gold / Player Time / XP / Consumables / ...
Think about how they all interact together and where does your business model fits in

Data

Where do we get data?

User testing / Qualitative testing

Online playtesting platform : https://antidote.gg/
Great to recruit 5/10 people.

Have some telemetry ready to understand what path do they take / where they are stuck.

Ask 3 Questions

Attribution partner

https://steamdatasuite.com/
When I run a marketing campaign, where does the player comes from?

In game data

It's increasingly difficult to extract raw data from game platform, so we need to back data gathering into the game -> ship to an Object Storage and observe with another software (Dataiku :))

Example of data

  1. Event trigger
    With context, anonymized of course
    2. Level start
    3. Level end
    4. Shooting action
    5. End of a match
    With context, so we can mesure a Win / Loss ratio

Market data

vginsight, steamspy, steamdb

What do we do with the data

=> Average of 10 games in the subgenre
=> Create revenue possibility
⚠️ It's not forecast, it's what success can look like

Top down analysis with Market data:
What the market is looking like?
What the growth is looking like?
=> Create KPI to use to mesure our likelihood of success

Bottom up analysis

Build Data targets

Example Indicators

Game report

Rate every next points with N/A, ISSUES, PARTLY, MOSTLY, GOOD
Detail every category by answering the questions:

Game design

Compelling core mechanic
Gameplay is extensible
Sense of purpose and progression
Play fits lifestyle & Mode of use
Compelling social experience

Monetization Design

Focus on player utility
Monetization fits platform/genre
Good use of hybrid monetization
Reasons for repeat purchase
Purchase experience is satisfying