Drawing on lessons learned from the experience of the traditional financial sector, the Neosurf model emphasizes that building a mature database on user behavior allows the game’s business partners to manage their operations much more closely than by the past. Historically, data has been pulled from banks and processors for analysis by service providers, but with little input from the learnings that APMs can add, either from wallet usage or vouchers.
Data that can identify and report bonus abuse, and that can detect unusual trends that may indicate fraudulent use, are just two of the extensions established APMs can provide; beyond the basic “know your customer” analysis. Identifying digital wallets where abusive behavior is evident and can be addressed allows operators to confidently optimize the use of desired services. To achieve a good level of trust, digital wallet data must be shared openly and integrated into account management reports.
Immature data has limited value, but deep mature data can be used in four distinct ways:
- To inform improvements to the client proposition
- To indicate fraudulent activity and adverse KYC analysis
- To identify and block bonus abuse
- Add an overview of promotional opportunities and communication strategies
The irony of providing the best bonus fraud and abuse data is that, anecdotally, this often results in poor trust scores on public trust measurement sites because there is no wholesale filtering capabilities on trusted sites to exclude scores and comments posted by scammers and abusers from the bonus system; those who suffered at the hands of data indicators. Those APMs with lower trust scores may be the ones providing the industry with the best service standards, and indeed this is the picture that emerges when customer satisfaction survey data collected internally is included.
No payment system provider, whether established traditional or alternative, will share their backroom algorithms in detail, as this would damage the integrity of the process, but some are more transparent than others to make it a virtue of openness to the data they collect and why.
There is a mix of factual and behavioral data collection, with factual data used initially to meet KYC regulatory requirements, with behavioral data more positioned for intelligent analysis; this is where the volume and depth of mature data points will dictate the value of the outputs.
The factual data elements verify predictable aspects such as addresses, credit history and others; but it is the interpretation of behavioral data that is particularly relevant for operators.
Behavioral data is captured from:
- The time of the registration process
- Time models at each stage of account registration
- Image metadata sent as proof of identification against other uploaded data
- The technological formats of POA-POI (develop)
- The pattern and frequency of the loading value, with volume and speed parameters
- The game model
- Spending behavior over time, based on the value and spending of traders at
By building on these elements and combining the results, patterns emerge to indicate:
- Indicators Volume Value Velocity to identify fraud and create actions to be taken
- Predictions of behaviors that trigger positive and preventive actions
- Player behavior indicates a clear pattern of regular use across different sites at different times, and depending on their winnings and offers, switch operators
- Behavior recognizable as fraud / criminal activity and can be stopped in real time
- Integration behavior and use of documents / types that indicate a potential risk of crime / fraud and measures created to stop, increase risk requirements
- Behavior models adapted to external influences: weather, seasonal occasions and disturbances. The Covid-19 is an obvious influence!
In this case, the blocks were created, stopping a number of activities. But more importantly, the use of data has been added to the analysis to establish insights. This information was then shared with specific operators so that together a proactive action plan could be put in place.
This excerpt from a very complex system shares insight into a high level process of finding consistency. Data output from this and other in-depth actions is a sustainable approach to matching behaviors to current records by geography, time, values, etc. Using analytics is essential internally, but sharing analytics is even more valuable for partnering with operators.
Industry commentators stress the importance of providing operators with mature, quality data. Customer satisfaction levels with the overall service proposition indicate quarterly trend data between 95% and 98%, highlighting not only the responsiveness of the customer service team but, just as importantly, the underlying capabilities of the customer service team. the platform driven by data analysis.