Identifying optimal price rise in telecommunications services

Identifying optimal price rise in telecommunications services

For Telenet Belgium, we tackled the challenge of determining the optimal price increase for telecommunications services. Faced with the delicate balance between maximizing revenue and minimizing customer churn, we employed a comprehensive approach. By modeling the impact on various factors, including ARPU, churn, and gross adds, we identified a bell-shaped revenue curve. This curve pinpointed the optimal price increase, providing valuable insights for the client on high-risk groups and more.


Data Science

Project Length

2 Months


Telenet Belgium

Our Planning Process

Due to high increases in inflation, companies also need to readjust their prices. The higher the price you set, the bigger revenue you get, almost immediately starting the next month. Price increases, however, negatively impact customers, and some customers may leave if the prices are too high. There is a sweet spot for a price increase where you maximize the revenue impact, i.e., you push the price as high as possible, until the point where customer churn outweighs the increased revenue. The data was prepared in a relational database using SQL, and analysis was performed using Python. 

What we did for Telenet Belgium

To identify the optimal price rise, we need to consider the impact on customer base revenue, the impact on churn, gross adds decrease, as well as package downgrades. The following steps were undertaken: 

  • Model price increase impact on ARPU of the existing customer base 

  • Model price increase impact on churn 

  • Model price increase impact on gross adds 

  • Consolidate the impact at product level and perform sensitivity analysis 

  • Detect the optimal price increase, suggest possible products to exclude from price rise, estimate the revenue impact and output the list of high-risk customers 

The existing customer base was modelled using elasticity curves based on customer product, tenure, usage, behavioral changes, income level, past behavior and more. Some customers will remain being customers almost no matter the price increase, some customers however, will decrease their spending or even downgrade to cheaper packages. These curves were forecasted for various hypothetical price increases.  

High prices affect both customer churn (i.e., customers leaving to the competitor due to high prices) and gross adds (i.e., new customers who may decide to stay with their current provider instead of migrating). Apart from modelling the customer base, we need to take these into account as well when accounting for the impact on revenue a potential price rise may have. 

Final Results

The end result when combining the price rise impact using all models and consolidating them together is a bell-shaped revenue curve. At the beginning the price increase also increases the revenue, but after a certain point the revenue impact from churn, reduction of gross adds and package downgrades outweighs the benefits and the slope turns negative. The highest point of the curve is at the optimal price increase, which at the end was suggested to the client alongside providing details on high-risk groups and more.