Entry
The story begins after riders already intend to buy.
Utah Transit Authority • BI + operations analysis
Mapped ticket-machine failure paths and turned a reporting gap into a clearer recovery and accountability plan.
UTA needed a clearer view of why riders were abandoning ticket purchases and why support teams were not seeing the same failures riders were experiencing in the field. The work focused on system-failure causes, rider flow, and which failure patterns deserved action first.
Short version: plain-language BI work made a machine problem easier to diagnose, prioritize, and escalate.
Customer Lifecycle
Where the leverage sat.
This was mostly a service-delivery and recovery story. The value came from understanding where riders were getting stuck, what support teams were missing, and which fixes had the clearest operational payoff.
Start
Riders enter the purchase flow at the machine.
Purchase Path
The main issue was where purchase attempts broke down.
Failure Analysis
Machine behavior and rider flow were reconstructed step by step.
Recovery
The output supported vendor fixes, alert tuning, and service follow-up.
Trust
Improving reliability supports rider confidence over time.
STAR Method
The case, kept short.
Situation
Riders were hitting repeat ticket-machine failures, but support visibility was incomplete.
Control Center saw some alerts, but not enough of the actual rider experience to understand the full failure pattern.
Task
Show where the failures were happening, how often they repeated, and what support teams were missing.
The goal was to turn a vague machine problem into something leadership could prioritize and escalate with evidence.
Action
Rebuilt the purchase path, pinpointed where riders got stuck, and compared that with what the vendor reported.
I used Excel, SQL, Power BI, Power Query, and Python to organize machine events into readable purchase attempts, rank the common failure paths, and compare those failures with the alert stream support teams were receiving.
Result
The team got a clearer recovery plan and a better basis for vendor accountability.
The work highlighted the dominant failure paths, surfaced roughly $15K in annual recovery opportunity inside a $2.0M fare context, and made the vendor reporting gap easier to challenge.
Growth / BI Levers Used
The practical moves behind the result.
Path reconstruction
- Turned machine events into readable purchase attempts.
- Made failure sequences easier to compare and rank.
Alert-gap review
- Compared machine-side failures with the vendor alert stream.
- Showed where support teams were getting an incomplete view.
Recovery prioritization
- Ranked fixes by frequency, rider impact, and recoverable value.
- Made remediation more practical for leadership and ops teams.
All Raw Data Remains Proprietary