Gamification designers should consider whether it is helpful to expose all the scoring rules to all players to achieve the business objectives they want.
Much of gamification thinking assumes 100% score algorithm transparency – “make it obvious how people score and they will do those behaviours that lead to a higher score” – is a typical design principle.
However this can lead to a non-optimal, reduced set of behaviours: instead, a hybrid approach can provide the best of both worlds.
The typical, fully transparent score algorithm can be characterised as:
- contextual – you get points in known contexts
- open – the ways to achieve points are clear to all
- overt – nothing is hidden
- direct – there is a clear link between action and points earned
Score algorithm transparency has significant positive appeal to players:
- each player knows exactly what they need to do to score
- a score can be reverse engineered to understand how it was achieved (the ‘right to explanation’)
- a score can be verified against underlying raw data
- all players can be confident that other players are competing using the same rules
- the service is seen as simple which can lead to higher overall adoption
However when providing a gamified service, this is not always the optimum design strategy – a 100% transparent score can also lead to negative side effects:
- players may stop doing “unscored” behaviours (even if they are desirable behaviours) to focus on the behaviours that do receive recognition
- players may over focus on scored behaviours (gaming the system / spamming) to the detriment of overall business objectives
- players may over focus on the game itself rather than on the activity being gamified
- non-gamers may be put off by the rigidity of the system and the mechanistic behaviour it requires to succeed
- managers may not get the relative weighting of points per activity correct. This will make some activities too cheaply rewarded and others too expensive to achieve, leading to a misallocation of effort
- managers may find it difficult to justify an update to the score algorithm, as the score rules are seen as “set in stone”. This leads to brittleness which can lead to a failing of the whole service, if it fails to flex in response to actual behaviour and business needs.
Gamification designers should consider two alternative strategies for scoring opacity:
- Fully opaque
- Partially opaque
In an fully opaque score algorithm, players do not see their score at all. This means that a player does not know their score at all:
- websites being ranked on Google,
- influence scores on Quora (what is Quora’s gamification blueprint? )
- EdgeRank on Facebook,
- Lithium’s community score….
This means that they derive many of the benefits of algorithmic processing -i.e. ensuring that relevant content rises to the top while those that seek to game the algorithm for personal advantage are hindered by its lack of transparency.
In a partially opaque score algorithm the manager decides which scores to explain and which to hide:
In some systems only the final score is shared (Credit Scores, Klout) while in others a sub section of metric are made transparent while other remain as “hidden metrics”.
For me the key benefits of opaque and partially opaque scoring algorithms for gamification designers are as follows:
- You can flex your scoring algorithm to reflect the behaviour you want at any time
- You can add mystery to how players are scored, so ensuring that untracked behaviours are continued by most players
- You reduce unwanted gaming of the system
However this must be offset against the reduction in simplicity for players which is a major factor in driving engagement.
As usual gamifiers, the final choice is yours.