PythonThe Question
This analysis uses the Airline Passenger Satisfaction dataset on Kaggle, a public dataset of over 100,000 airline passenger records covering service ratings, trip context, and flight delays. It's used here to explore a question shaped by years of contact center CX work: loyal customers vs. one-and-done customers, what actually separates them? Using each passenger's Customer Type (Loyal / disloyal) as a proxy for loyalty, the analysis tests that question against service ratings, trip context, and delays to see which dimensions actually move the needle.
Approach
Rather than jumping straight to a model, the analysis builds up in layers, ruling out simpler explanations before trusting a more complex one.
Data Overview
Checked shape, types, missingness, and target balance across the full 103,904-row dataset before drawing any conclusions from it.
Loyalty vs. Trip Context
Tested whether loyalty simply tracks who's flying, class, type of travel, age, flight distance, before crediting the service experience with anything.
Loyalty vs. Service Ratings
Ranked all 14 in-flight and ground experience dimensions by the gap between loyal and disloyal average ratings, so the biggest drivers surface first instead of eyeballing 14 separate charts.
Loyalty vs. Operational Reliability
Compared delay frequency and severity between loyal and disloyal passengers using the industry-standard 15-minute threshold, the one dimension in the dataset that isn't a subjective rating.
What Predicts Loyalty
Trained a Random Forest classifier across every dimension at once to see what separates loyal from disloyal customers once everything else is accounted for. The model reached 94.6% accuracy.
A caveat worth carrying forward: loyalty isn't evenly split across trip purpose. Personal travel is 99.5% "Loyal Customer," while business travel is only 73.7% loyal, the reverse of typical CX intuition, where routine business flyers are usually the most loyal segment. That suggests Customer Type may be capturing loyalty-program membership more than satisfaction-driven repeat business, a distinction worth holding onto before treating every finding below as a service story.
Key Findings
- Who's flying beats how the flight went. Trip context, mainly business vs. personal travel, age, and trip length, outweighs the entire service experience.
- Booking and boarding stand out among the service ratings. Online boarding, on-time convenience, and easy booking separate loyal from disloyal customers the most. Comfort and in-flight extras barely register.
- Delays don't move loyalty. Whether a flight was on time made almost no difference, the "a late flight kills loyalty" assumption doesn't hold up in this data.
- Read this as a technique demonstration, not a verdict. The dataset doesn't say why someone is labeled "loyal," so these findings illustrate a multivariate analysis approach rather than a validated loyalty strategy.