Last week I introduced the first three of six critical game mechanics to consider when designing a leaderboard. Here are the final three visualization considerations that can significantly affect your leaderboard’s effectiveness: multiple leaderboards, points, and metadata.
Multiple Leaderboards
Many people assume there can be only one leaderboard per site, but this isn’t true. By narrowing the competitive space through filtering and aggregating, a leaderboard comes to represent a context — a subset of the larger experience. Not every context is relevant to every user, and multiple leaderboards are more likely also used to appeal to diverse users.
Context doesn’t always have to be based on user preferences either; boards can be used to encourage different behaviors. In social situations where both teamwork and individual effort are desirable, two leaderboards might be appropriate: one team board to encourage teamwork, and another, individual leaderboard to apply a little individual pressure.
Points (an abstracted measure)
So far I’ve spoken about “scores” and avoided using the word “points,” because they aren’t the same thing. A score is any measurable quantity and could include either points or direct behavior counts. In fact, behavior-based and point-based leaderboards are fundamentally different.
A behavior-based leaderboard is the simplest leaderboard you can have. Participants are ranked by the number of times they perform a specified behavior. What if you have multiple relevant behaviors and can’t be limited to just one? The solution is a points system: each behavior gets a point value whose amount corresponds to its perceived significance. For example, on the National Hockey League leaderboard, a win is worth 2 points and an overtime loss 1 point.
![]()
This leaderboard ranks by points (PTS), not wins (W)
Points are much more flexible than behavior counts but there are a few disadvantages to using a points system:
Audience Rejection
The biggest risk with points is that the audience doesn’t accept the valuation assigned to behaviors. By assigning (or not assigning) points values to behaviors, you are making an official statement of their relative value. If your valuation doesn’t match the general sentiment of the community, your leaderboards won’t be effective motivators.
Loss of specificity
The more behaviors you add to a points system, the less real meaning an actual score has (because there are multiple ways that score could be achieved). In an American football game, if a team has 15 points, it’s hard to tell how they got that point total. Maybe they scored 5 field goals (3 points each), or three field goals and a touchdown (6 points) but missed the extra point. Or maybe they scored two touchdowns and successfully went for two extra points after one of them. The fact is, once you abstract behavior with a points system, there is no way of knowing exactly what a participant did to earn their score. This becomes even more relevant when you consider the next problem…
“Gaming” the leaderboard
Once you introduce a points system, players will be encouraged to seek out the behavior that provides the easiest route to earning points and focus on it. This is a form of “gaming” the system, and the only solution is to take the best care possible when first balancing your points assignments.
Metadata
Think about a leaderboard that measures sales team performance. If the leaderboard counts the behavior “closed deal,” what’s missing? The number of deals a salesperson closes is relevant, but it’s not enough to provide the whole picture. You also need to know the dollar amount of the deals closed. This is an example of metadata. A single instance of the activity “close deal” might include a metadata value of $10,000, while another instance includes only $3,000. A metadata leaderboard is a behavior-based leaderboard that ranks players by the metadata sum of all instances of the behavior being tracked.