Track Beyond These 3 Customer Feedback Metrics: NPS, CES, and CSAT

• CSAT, NPS, and CES are the three basic metrics under the umbrella of Customer Feedback Metrics (CFM).
• Organizations in different sectors use these metrics by customizing them according to their needs.
• It is possible to come across examples where the scale and question structure is changed and adapted to the context.

• CES is a metric with a weak theoretical basis just like NPS. These metrics attempt to measure a very large and complex concept with a single question and predict loyalty with this measurement.
• There are conflicting results in the literature on the relationship between these metrics and market performance.
• However, they are widely used due to their ease of application, their ability to identify problems in real-time, and their ability to fulfill the need for sectoral benchmarks.

• The CES has a 5-point scale in the original source.
• The question is “How much effort did it take to accomplish your request?” and is scored from 1 (very little effort) to 5 (a lot of effort).
• The important point in the question pattern is to avoid the use of difficult and easy expressions that cause mental manipulation and to ask about the effort expended. It is inconvenient to compare CES measurements obtained using different question patterns and scales.

• Posing the question as difficult or easy affects the answers. If you ask how difficult it is, the answers will tend toward the more difficult end.
• If you ask how easy it is, the responses will tend towards the easier end. The important point in the scale used is that the scale should be symmetrical.
• It is important that a customer who does not characterize their experience as difficult or easy can choose the middle item of the scale.

• The top-2-box approach to calculating customer feedback metrics provides a better prediction of loyalty.
• The top-2-box approach calculates extreme responses (e.g. easy or difficult) while ignoring middle options. NPS uses the categories Promoter and Detractor, while CES uses the categories Low Effort and High Effort. This calculation method yields more reliable results than indexed or unindexed scale averaging.

• The analysis of the analytical model developed to explain or predict a metric should be determined in accordance with the approach to calculating the metric.
• When a top-2-box method is used to calculate a metric, people need to move from lower-category to upper-category membership in order to improve this metric. It is not possible to identify the factors that enable this transition through a linear regression-based analysis.
• If you use an analysis method that is incompatible with the way a metric is calculated, your results will be misleading.

This article first appeared in Cloud4Feed’s CX Kaleidoscope Yearbook 2024 and is published here as written by the author. Part 1 of 8