Automated root-cause insights for airline customer complaints
Key Metrics
Complaints Analyzed
Categories Identified
Overall Negative %
Model Accuracy %
Analysis
Complaint Volume Over Time
Daily stacked sentiment breakdown
Category Severity Profile
Multi-axis comparison
Insights
Priority Ranking
Complaints sorted by severity and volume
Key Topics
Sample Complaints
“Look at that. The flight I have been trying to book for 3 days is gone. I guess someone got through. Thanks! Keep up the #Failing”Feb 21, 2015
“your customer service today was deplorable. There were plenty of ways u could have gotten us on that flight. #badcustomerservice”Feb 24, 2015
“your customer service stinks. Trying to book a flight for hours now and keep getting hung up on. #usairwayssucks”Feb 17, 2015
Key Topics
Sample Complaints
“thanks for delaying our flight for an hour, missing our connecting flight and terrible customer service. #neveragain #vacation”Feb 20, 2015
“can a real person help me here? This customer service is poor. I'm at the mercy of someone since the flights are disappearing.”Feb 22, 2015
“you have the worst service, you Cancelled Flightled all your flights FLL to PHL all flights flew. Stuck in FL 3 days. #done”Feb 22, 2015
Key Topics
Sample Complaints
“you guys rock!!”Feb 20, 2015
“you guys rock!”Feb 22, 2015
“thank you! ❤️❤️❤️ you guys!”Feb 19, 2015
Key Topics
Sample Complaints
“never had an airline refuse to help sit parents with tiny kids. But this guy is willing to do it. #JetBlue”Feb 22, 2015
“why don't you allow families with small children to board first anymore? Our next flight will be on another airline...”Feb 22, 2015
“being told by this guy that he will not help us sit next to our 5 and 8 year old on the flight. #jetblue”Feb 22, 2015
Key Topics
Sample Complaints
““: Our fleet's on fleek.”Feb 24, 2015
““: Our fleet's on fleek.”Feb 23, 2015
““: Our fleet's on fleek.”Feb 23, 2015
Key Topics
Sample Complaints
“where's my apology?”Feb 18, 2015
“no apology? Cool.”Feb 22, 2015
“i appreciate your apology. Sincerely. Thank you. That's really all I ever wanted to begin with.”Feb 23, 2015
Key Topics
Sample Complaints
“I love your company and your staff is amazing. They just made an uncomfortable situation comfortable”Feb 22, 2015
“was in a line a mile long at sky harbor this morning. Your staff was courteous and expeditious. Thank you. #onechildfourbags”Feb 22, 2015
“one staff on desk. Now been queuing for over an hr.”Feb 23, 2015
Key Topics
Sample Complaints
“719. Looks like we are about to get going, finally!”Feb 23, 2015
“2284, four hours Late Flightrs and we are finally flying out...too bad I missed my event”Feb 24, 2015
“3127. Just landed in LIT.”Feb 23, 2015
Key Topics
Sample Complaints
“Nope, I have not been rebooked.”Feb 23, 2015
“Already rebooked for tomorrow. Fingers crossed!”Feb 17, 2015
“yep, after long waiting on the phone we managed to rebook it. Hope it will be a pleasent one. Thanks for ur help.”Feb 18, 2015
Key Topics
Sample Complaints
“how come I'm not getting my points after I buy from the rapid reward shopping site?”Feb 24, 2015
“customer service was so bad updating my points, once I use my next points, I won't be flying you. Point system is almost a sham.”Feb 19, 2015
“thank you for the response. How can I got about getting the points onto my rewards account?”Feb 17, 2015
Key Topics
Sample Complaints
“can you follow me so I can send the DM?”Feb 21, 2015
“follow/dm please”Feb 23, 2015
“you have to follow me in order for me to DM...come on now”Feb 24, 2015
Key Topics
Sample Complaints
“Thank you!”Feb 18, 2015
“thank you!”Feb 23, 2015
“thank you!”Feb 23, 2015
Key Topics
Sample Complaints
“Celebrates 15-Year Anniversary With New Livery - Digital Journal”Feb 18, 2015
“unveils new 'Bluemanity' livery - USA TODAY”Feb 19, 2015
“Thank you that it is not just a livery; it is a culture that 16,000+ crewmembers embody daily #thanksDave”Feb 17, 2015
Key Topics
Sample Complaints
“'s CEO battles to appease passengers and Wall Street -”Feb 20, 2015
“'s CEO Battles to Appease Passengers and Wall Street -”Feb 17, 2015
“'s CEO battles to appease passengers and Wall Street -”Feb 19, 2015
Key Topics
Sample Complaints
“I don't think you should help him at all based on his behavior. The voucher and cot seem like enough lol 😃”Feb 24, 2015
“I don't think you should help him at all based on his behavior. The voucher and cot seem like enough lol 😃”Feb 24, 2015
“how can he try again it will be 5-6 hours before he gets help. Contingency plans non existent.”Feb 22, 2015
Sentiment Health
Overall satisfaction metric
Sentiment is severely negative
0–40
Critical
40–70
Caution
70–100
Healthy
Deep Analysis
Category Comparison
Sentiment composition across all categories
Customer Service
6821 complaints
55.93% negative
Other / Uncategorized
6231 complaints
55.08% negative
Positive Feedback
497 complaints
20.93% negative
Kids Old
99 complaints
52.53% negative
Fleek Fleet fleek
169 complaints
20.12% negative
Apology Experience
43 complaints
76.74% negative
Rude Staff
59 complaints
47.46% negative
140 Characters
53 complaints
47.17% negative
Rebooked Rebook
39 complaints
58.97% negative
Points Pts
39 complaints
51.28% negative
Dm Follow
157 complaints
8.92% negative
Positive Feedback (Thank)
321 complaints
4.05% negative
Selfie Marks
30 complaints
13.33% negative
Ceo Wall
70 complaints
2.86% negative
Customer Service (Help)
12 complaints
58.33% negative
Complaint Density by Day
Volume patterns across the week by category
Legend
Complaint Volume
High Negative %
Volume vs. Severity
Complaint volume against sentiment severity
Severity
Notes
Bubble size = severity score. X-axis = complaint volume. Y-axis = negative sentiment %.
Activity
Recent Complaints
Latest feedback from customers
“I don't think you should help him at all based on his behavior. The voucher and cot seem like enough lol 😃”
“I don't think you should help him at all based on his behavior. The voucher and cot seem like enough lol 😃”
“how come I'm not getting my points after I buy from the rapid reward shopping site?”
“you have to follow me in order for me to DM...come on now”
“I'm not a child. I'm someone who has an issue with flying and prepares ahead of time to reduce the distress caused by planes. You”
“your customer service today was deplorable. There were plenty of ways u could have gotten us on that flight. #badcustomerservice”
“2284, four hours Late Flightrs and we are finally flying out...too bad I missed my event”
““: Our fleet's on fleek.”
Methodology
Insightwell analyzes customer complaints through an automated pipeline:
- Topic Modeling— Groups complaints into thematic categories (e.g., “Flight Delays”, “Lost Baggage”) using unsupervised clustering on complaint text embeddings.
- Sentiment Scoring — A fine-tuned transformer model assigns a sentiment score (positive, neutral, negative) to each complaint, calibrated against human-labeled samples.
- Severity Ranking — Severity is computed as a weighted combination of volume (raw complaint count) and negative sentiment percentage, prioritizing high-frequency negative issues for stakeholder attention.
Dataset Note: This dashboard demonstrates analysis using a public airline customer feedback dataset as a stand-in for real company data. In production, Insightwell integrates directly with live complaint ingestion systems (email, chat, survey platforms) to provide real-time visibility into customer sentiment trends.