How our healthscore works
How our healthscore works
# Comparative analysis of customers
Skalin constantly compares all your customers on dozens of indicators to determine the strengths and weaknesses of each of your accounts.
Our AI constantly analyzes more than 70 KPIs across your entire portfolio to highlight customers at risk and detect those with potential.
These indicators are divided into 4 main categories:
- Criteria linked to the use of your Product: proportion of active users, time spent on the platform, depth of use, level of adoption of your key features, time between two connections, evolutions...
- The interactions your customers have with your team: proportion of contacts addressed, decision-makers regularly involved, level of customer satisfaction with each interaction (thanks to sentiment analysis, see below), email response rate...
- CSM Pulse, i.e. the CSM's assessment of the account
- Contract-related data: evolution of MRR, consumption of additional services, contract history, remaining duration, etc.
These indicators are calculated for each customer segment, automatically identifying accounts that are weak on a metric, compared with their reference group. A customer who is weak on too many KPIs will automatically see his health score drop. So you don't have to worry about defining the criteria to be analyzed - AI does it for you!
# Behaviour forecasting
Taking into account the evolution of each of the KPIs allows us to nuance the previous analysis. For example, a customer may have a low connection frequency compared with the average for his segment, but if he makes progress in relation to his own history, the AI will take this into account and consider that the customer is part of a positive dynamic. By analyzing the customer's history, we can refine the calculation of the health score and predict future behavior.
# Self-learning model
While the first point enables AI to be effective very quickly, the integration of Machine Learning models optimizes the score over time. When a customer churns or drops into long-term inactivity, the AI will reintegrate previous behaviors to progressively give greater weight to the most discriminating variables. The more historical data the AI has at its disposal, the more relevant it will be.