Description

Churn (loss of customers to competition) is a problem for telecom companies because it is more expensive to acquire a new customer than to keep your existing one from leaving. This contest is about enabling churn reduction using analytics.


The Business Pain:

Most telecom companies suffer from voluntary churn. Churn rate has strong impact on the life time value of the customer because it affects the length of service and the future revenue of the company. For example if a company has 25% churn rate then the average customer lifetime is 4 years; similarly a company with a churn rate of 50%, has an average customer lifetime of 2 years. It is estimated that 75 percent of the 17 to 20 million subscribers signing up with a new wireless carrier every year are coming from another wireless provider, which means they are churners. Telecom companies spend hundreds of dollars to acquire a new customer and when that customer leaves, the company not only loses the future revenue from that customer but also the resources spend to acquire that customer. Churn erodes profitability.


Steps that have been adopted by telecom companies so far:

Telecom companies have used two approaches to address churn - (a) Untargeted approach and (b) Targeted approach. The untargeted approach relies on superior product and mass advertising to increase brand loyalty and thus retain customers. The targeted approach relies on identifying customers who are likely to churn, and  provide suitable intervention to encourage them to stay.


Role of predictive modeling:

In the targeted approach the company tries to identify in advance customers who are likely to churn. The company then targets those customers with special programs or incentives. This approach can bring in huge loss for a company, if churn predictions are inaccurate, because then firms are wasting incentive money on customers who would have stayed anyway. There are numerous predictive modeling techniques for predicting customer churn. These vary in terms of statistical technique (e.g., neural nets versus logistic regression versus survival analysis), and variable selection method (e.g., theory versus stepwise selection).


Objective of this Contest:

The objective of this contest is to predict customer churn. We are providing you a public dataset that has customer usage pattern and if the customer has churned or not. We expect you to develop an algorithm to predict the churn score based on usage pattern. The predictors provided are as follows:

  • account length
  • international plan
  • voice mail plan
  • number of voice mail messages
  • total day minutes used
  • day calls made
  • total day charge
  • total evening minutes
  • total evening calls
  • total evening charge
  • total night minutes
  • total night calls
  • total night charge
  • total international minutes used
  • total international calls made
  • total international charge
  • number customer service calls made

Target Variable:
Churn: if the customer has churned (1=yes; 0 = no)


Technology:
All tools that could develop machine learning techniques and predictive modeling algorithms such as Java, Python, R, Rapidminer, Orange, WEKA, Octave, and SVM-light are welcome. Although, please note that conditional winners will be required to submit their algorithm code and tools that don't allow this should not be used. This includes proprietary frameworks, which cannot be submitted to CrowdANALYTIX for validation.


Solvers Expectation:
Participants may submit one (1) entry every 24 hours of the competition period.  CrowdANALYTIX reserves the right to request that a participant submit the prediction algorithm associated with an entry to CrowdANALYTIX through the “Responses” tab on the contest page. Once an entry is selected as eligible for a prize, the conditional winner must deliver the prediction algorithms code and

Evaluation Method CHURN is the targ Top six solvers based on this evaluation method will be shortlisted for further evaluation by a review board. The review board will comprise independent experts and the six shortlisted solvers. Further details can be found after joining the contest and going to the Criteria tab.


Timelines:
Submission Deadline : 30th July, 2012
Results announced by: 10th August, 2012


Prize money:

  • One first prize of US$500
  • One second prize of US$300
  • One third prize of US$200

Latest Participants

Latest Entries

Forum Updates

Competition Administrator


Prize Type Amount in $ No.of Prizes

Solvers Entry Last Submission Leaderboard

Solvers Last Activity

Total Entries


Submission deadline has been passed! And you can no longer submit responses or edit your existing responses
Add an entry
File Uploaded Date
Add Topic
Topic Name Last Post Replies




Winners


Winners