Some data that is used to predict churn for products
This dataset appears to contain customer information and billing details for a telecommunications company, likely a cable or internet service provider. The column names suggest various aspects of customer demographics, billing practices, and services used, such as gender, senior status, dependents, tenure, phone and internet services, security measures, and payment methods.
Overall: The dataset has an overall quality score of 96.41/100, indicating a high level of data quality. The completeness and validity scores suggest that the dataset is nearly complete and accurate. However, the inconsistency score indicates some variation in data entry or formatting.
Completeness: The dataset achieves a completeness score of 99.9/100, meaning that most rows are fully populated with relevant information.
Validity: The dataset has a validity score of 100.0/100, indicating that all data is accurate and reliable.
Uniqueness: The dataset has an uniqueness score of 100.0/100, suggesting that each customer record is unique and not duplicated.
Consistency: Despite the high overall quality, the consistency score indicates some variation in data entry or formatting, with a score of 82.18/100.
Analyze customer demographics and billing patterns to identify trends and opportunities for targeted marketing
Evaluate the effectiveness of different pricing models and service plans
Develop predictive models to forecast churn rates and optimize customer retention
The dataset may not be representative of all customers, particularly those with missing or inconsistent data.
The quality scores should be considered in the context of potential biases or errors in data entry or formatting.
telecom, cable, internet, customer analytics, billing management
Visualize dataset coverage on a map. Geospatial columns (latitude/longitude, lat/lon, x/y) are detected automatically.
Loading analysis steps...
Please log in to view your saved notebooks
No comments yet — be the first to share your thoughts.
Please log in to comment.