Compare trends across player clusters and spin bandings
Overview
KPI Analysis is a part of Player Behaviour section on Optimise. It can be used to view the contribution made to various KPIs be either player clusters or spin banding. For a table view of this information, it is possible to use either Player Cluster Comparison or Spin Banding Comparison, both in the Player Behaviour module.
Visualisations
This page contains two line graphs:
- KPI by Player Cluster
- KPI by Spin Banding
By Player Cluster
In this line graph it is possible to view GGR, NGR, number of Players and Stakes by each player cluster group. This will show you the contribution that each cluster has made to each of the KPIs. It is possible to focus on one or more of the clusters by selecting one or multiple options from the Player Cluster drop down above the chart.
Know more about Player Clusters
By Spin Banding
In this line graph it is possible to view GGR, NGR, number of Players and Stakes by each Spin banding group. This will show you the contribution that each spin banding has made to the KPIs. It is possible to focus on one or more of the player clusters by selecting one or multiple options from the Spin banding drop down above the chart.
About Player Clusters
How to utilise player clusters
Players are grouped together using a popular machine learning technique known as clustering. This technique considers your entire player base and organises them into clusters based on their behaviour.
How often player clusters are updated
Amplifier AI's models run on a daily basis and allocate players to a single or multiple clusters, based on their behaviour for every game they play.
As such, the total number of players in all clusters will exceed the number of unique players, and therefore it is intended to be used to understand the proportionate distribution between clusters.
Attributes of player clusters
The model considers the following attributes when assigning a player into a cluster:
- Number of spins
- Bets
- Average bet
- Average session length
- Bonus stakes (if available in data)
Using this information, players are assigned per game into one of the following categories:
- High Value Multi Session — these players tend to have highest number of rounds, sessions, bet value, and often have higher RTPs. They are termed 'Multi Session' as they play multiple sessions of the same game per day.
- High Value — these players share similar attributes to High Value Multi Session players, however they only play one session per day.
- Mid Value Multi Session — these players tend to play fewer rounds, place smaller daily bets, have smaller session lengths and less RTP than High Value Multi Session players.
- Mid Value — these players share similar attributes to Mid Value Multi Session players, however they only play one session per day.
- Tourists — these players contribute low amounts of GGR, play one session on average and tend to place few bets with low RTPs.
- Low value — these players have the lowest contribution to all the metrics compared to other clusters. It is likely that bonus abusers would appear in this cluster.
About Spin Banding
Spin banding is a useful tool to help you have a granular view of player behaviour, by understanding player engagement. On this page, every session is allocated to a spin band, depending on the number of spins played in that session. The bands are set as:
- 1-50
- 51-100
- 101-200
- 200-400
- 400+
How do I get support or make a feature suggestion?
Visit our 'here to help' form and we will respond accordingly.