Customer case

Net Promoter Score improvement plan

Tue 04 Jan 2022

For several years now, the Net Promoter Score (NPS), which is the ultimate indicator of customer recommendation, has been scrutinized by all companies to track the level of satisfaction of their customers.

As part of its customer experience improvement plan, a telecom operator has identified 8 major indicators on which it is focusing its efforts. These indicators correspond to tangible and quantifiable customer pain points, and are taken into account in the variable compensation policy of employees, thus allowing global involvement of all collaborators.

In order to improve the monitoring and analysis of these indicators during the monthly customer experience committees, he called on our business analysts to:

  • Simplify and automate the indicator tracking tool,
  • Add a focus on Premium customers
  • Analyze the root causes of customer pain

A “one click” automation  

The analyzed indicators are divided into 4 categories:

  • Fixed network: number of outages, speeds
  • Mobile network: connection quality (time spent in 2G/3G...)
  • Offer: is the customer benefitting from the most recent offer? Exceeding bundle duration
  • Customer relationship: date of last call, frequency of calls to customer service...

In order to simplify the preparation of the monitoring table for these indicators, our business analysts first restructured and reprocessed certain data (deletion of duplicate customers ....) to allow for "one click" automation. Thus, all the data is set up in a single action.

Then, they created a second dashboard to specifically track customers with a Premium offer for a better follow-up of these customers.

To assist in the decision-making process, facilitate the analysis of the indicators produced, and thus improve the relevance of the resulting action plans, our experts have worked to identify the drivers, i.e., the root causes of the downward or upward movement of these indicators.

In this exercise, it is important not to confuse causality and correlation, as this could lead to false conclusions and inappropriate action plans.

Correlation determines a relationship between two variables. However, the fact that these two variables move together does not necessarily mean that one variable is the cause of the other. A strong correlation may indicate a causal link, but there may also be many other explanations: It may be the result of chance, or there may be a third variable to consider that makes the relationship look stronger (or weaker) ....

In the case of the analyzed indicators, it is the business sense that will be able to determine whether the correlation is good or bad.

For this purpose, Sofrecom's analysts relied on our client's business teams to identify the list of the most relevant drivers.

For example, in the case of an ADSL offer, low speeds can be explained by the distance between the connection node and the customer's home. By minimizing this distance, we minimize the problem. The distance between the home and the ADSL node is therefore the cause of the low speed. For line cuts, finding the driver is more complex because customers can voluntarily turn off their box every night, the weather (storm, tempest) can also be the cause of cuts ....

At the end of this work, the Sofrecom team has proposed actions to minimize customer dissatisfaction.

Result

The improvement in the monitoring of indicators has allowed:

  • A significant time saving in the preparation of reports: from several days before the project, this preparation is now done in a few hours.
  • A better clarity of the tool which allows to reduce the risks of errors.
  • Specific monitoring of indicators for Premium customers and the development of specific levers.
  • A better understanding of the original causes of customer dissatisfaction and the levers needed to reduce this dissatisfaction.

Benchmark: 

NPS - Net Promoter Score: Recommendation score ranging from -100 to +100

Customers answer the question: On a scale of 0 to 10, how likely would you be to recommend this service to someone close to you?

0 means you would definitely not recommend it, 10 means you would definitely recommend it.