Automotive Service Churn

Proving that churn is predictable from service data


1 October 2006

Prompted by declining profits from their automotive service centres caused by lower than expected utilization, this client called upon CYBAEA to reduce service churn and at the same time look for other commercial opportunities from data and a better understanding of their customers, markets, and competitors.

First page of case study
Download PDF case study.

Download PDF

The client, which is part of a large retail giant, has about €750m revenues from importing, selling, and servicing four automotive brands. For this phase of the project we looked at two of the car brands, high- and mid-tier, in a single Nordic country, and the client’s own service centres only (excluding their dealers).

Initially sceptical that they would not have enough quality data, the client was astounded by the level of insights and commercial actions that even this first phase delivered in only four weeks.

Project approach

First phase was an accelerated effort to deliver quick commercial value on the problem at hand and to identify additional opportunities.

Project approach

The project was divided into two parts: investigation, to understand the data, the processes, and the commercial drivers, and analysis to deliver the models, identify the commercial opportunities, and propose the next steps.

Regular workshops ensured constant alignment with the business and ultimately rapid implementation of the changes to deliver on the commercial opportunities.

Business results

Churn is very predictable from service data, and this gives deep insights into why customers are not coming back. Based on these insights a clear list of immediate action was developed in three broad areas: Commercial changes, data opportunities, and improvements in execution capabilities. Selected examples of these quick wins are included below.

Churn is very predictable from service data, and this gives deep insights into why customers are not coming back.

Commercial action examples:

  • Trial a BTL communication to at-risk customers around the time of the next service, using the prediction model to select the targets and the descriptive model to select the messages.

  • Trial service programmes focusing on paid-for convenience options such as express service, collect and return service, courtesy car, Sunday openings, and more. The options depend on the capabilities and spare capacity of each service centre and they target specific pain points in the customer experience as identified by the analysis.

Data action examples:

  • Instigate qualitative and quantitative research to understand churn reasons and customer experience for (only) one very specific group, namely the customers who churn after the first visit and for whom the data available from the systems is therefore limited.

The project clearly made the economic case for the Net Promoter System™ (NPS) and even with a world-class score there was commercial value to be had from expanding the programme.

  • Deep dive into the Net Promoter programme and ways of increasing responses. It turned out that the Net Promoter Score was an important predictor of churn and while the client overall had a world-class score there were gaps in the programme and the data was under-utilized.

Execution action examples:

  • Put in place robust campaign management with consistent use of test and control groups and rigorous capture of campaign outcomes to ensure future learnings. This was primarily a process and training change and thus easily implemented.

  • Ensure customer-facing staff capture accurate data. This project clearly demonstrated the commercial value of data but the challenge is to ensure that the front line staff sees it as their job to capture accurate information. This would be done by incentives, education, or cultural change, or, more likely, a combination of these. While it may not be a quick change, it is one with no particular constraints on execution and so can be readily started.

All of these actions could be implemented immediately to deliver quick commercial value to the organisation and an improved experience for the customers.

Initially sceptical that they would not have enough quality data, the client was astounded by the insights and commercial actions from this rapid first phase. All the actions could be implemented quickly to deliver immediate commercial value to the company and improved experience to its customers.

Download PDF