Repeat Caller Model after Self-Serve
Do you want to know why customers are calling the call center or initiating a chat session instead of completing their transaction online or on the app?
Use key driver analysis from this model to help you understand opportunities for improving customer pain points!
This model is designed to be used in any call center and for a range of industries such as Telecom, Streaming Media Services, Entertainment, Hospitality, Airlines, Retail, Banking, Health Insurance and more.
Full model price: $60,000
- $6,000 initial deposit
- $24,000 first payment (invoiced at data access)
- $30,000 final payment (invoiced at model delivery)
PRODUCT INFO
Use Case
Understanding how to improve self-service channels, predicting and preventing calls while increasing customer satisfaction and retention. Lowering the cost of handling high volume calls in call centers.
AI/ML Modeling
A range of algorithms will be tested for the best AUC/outcome like XGBoost/GBM, Neural Network, SVM, ANOVA, KNN, K-Means, etc.
LLMs and NLP techniques may also be used to enhance model performance.
Model Delivery
One-time purchase:
- Propensity scores
- Customer ranking
- Leading model predictors
- Model performance
- Insight session
On-going monthly service additionally includes:
- Tracking leading predictors
- Model monitoring
- Seemless model refresh with market trend changes
- Strategy session every 3 months of service
- 10% discount on full model price
DELIVERY AND PAYMENT
We will contact you within 2 business days of your deposit to set the engagement date and provide instructions for the required data set.
The first model payment is invoiced at the data access date. Model delivery usually occurs 3-4 weeks from the data access date. The final model payment is invoiced at the model delivery date.
REFUND POLICY
Deposit fully refundable before engagement date.
If we cannot detect a pattern for a stable model, the final model invoice will be waived and we will provide you with data assessments, findings, insights and further recommendations.