AVGEEK
Reading India’s fleet and profitability risks through data
Courtney Miller explains how visualisation and forecasting inform airline decisions, examining aircraft orders, utilisation, load factors and yields against India’s growing demand and profitability challenges.
The conversation
Episode 54
What happens when a pilot, an airline operator, and a strategist decide to reimagine aviation through the lens of data? In this episode, I sit down with Courtney Miller, the mind behind Visual Approach Analytics, to explore how storytelling, visualisation, and forecasting are reshaping the way airlines make decisions.
What we discussed:
• Courtney’s journey into aviation: from the cockpit to Bombardier and DHL, and eventually into analytics. • Why he believes the industry was “oddly slow” to recognize the value of aviation analytics. • Lessons from DHL’s cargo forecasting (yes, Valentine’s Day and Mother’s Day were the most stressful!) and how they differ from passenger airline planning. • The art of turning complex aviation data into accessible intelligence. • Fleet planning quirks: Allegiant’s unique approach, Akasa’s Boeing delivery challenges, SpiceJet’s survival instincts, and what Indian carriers like Air India and IndiGo should be doing differently. • The profitability puzzle: India’s booming demand versus chronic losses—what data signals separate hype from traction. • Are Indian airlines over-ordering? Courtney breaks down the metrics that reveal overcapacity, underutilization, and systemic risk. • India’s ambition to be a global hub: should carriers double down on regional connectivity, long-haul expansion, or partnerships? • Air India’s post-privatization struggles and what it could learn from South Korea’s model of global talent and best practices. • Disruptive shifts in aviation—from AI and machine learning to evolving fleet strategies—and what’s hype versus reality. • The next frontier for aviation analytics, misunderstood metrics like load factors and yields, and Courtney’s favorite airline innovation he’d copy worldwide.
Along the way, expect witty detours (including a Ryanair comparison), sharp insights on pricing (“AI doesn’t always mean personalized pricing”), and a candid look at how data has evolved in aviation decision-making over the past five years.
Key Takeaways
- Passenger and cargo economics differ fundamentally because cargo has its own handling and distribution network.
- Strong forecasting traces chains of operational consequences rather than extrapolating one curve.
- A fast-growing market can become less profitable as competitors compete away yields through excessive capacity.
- Network differentiation and operational execution matter as much as a glamorous fleet growth plan.
- Ryanair’s consistency is offered as a challenge to the claim that the low-cost model itself is obsolete.
- The industry’s analytical frontier is increasingly decision support, not simply a larger data warehouse.
- Maintenance forecasts using specialist machine learning may have more directly observable value than generic AI customer-service claims.
Additional subjects
Editorial analysis
The industry problem behind the numbers
Courtney Miller began as a pilot and moved into aviation analysis. He brings a practitioner’s scepticism to stories about demand and capacity: the existence of a numerical trend does not mean an airline should extrapolate it into another aircraft order. A particularly instructive contrast is passenger versus cargo networks. Passengers can enter and leave the aircraft under their own power; cargo must be physically moved, sorted and often trucked, changing both cost and network design. The lesson is methodological: the right unit of analysis depends on the operation.
Forecasting cause and consequence
The discussion takes off when Miller recalls forecasting a shortage of airline pilots and a later tightening of aircraft availability during the industry’s post-pandemic adjustment. He describes working through causal chains rather than treating the initial shock as an isolated data point. The value is not that one analyst foresaw every development; this is his account and the interview supplies no formal forecast scorecard. The value is to ask where a change in aircraft production, training pipeline, labour costs or maintenance downtime will surface next in the industry system.
India’s capacity discipline test
On India, Miller is wary of the assumption that rapid growth automatically rewards every competitor. He argues that persistent additions of capacity can undermine returns and that disciplined execution matters more than a perpetual race for market share. He contrasts different airline business models globally and invites India’s carriers to be strategically distinctive rather than reproduce a foreign model mechanically. The revealing problem is behavioural: when an industry grows quickly, managers can continue to pursue scale beyond the point where it creates value.
Where aviation AI might actually pay
His perspective on AI is similarly precise. He traces the move from collecting very large datasets to actually asking answerable operational questions. He is sceptical that generic language models will deliver the biggest gains. Rather, he highlights specific machine-learning approaches to aircraft and engine maintenance, including predicting life-limited-part requirements and organising shop visits. That is a plausible high-value application because a useful forecast can change an actual schedule or maintenance procurement choice. It is not a proven quantified benefit in the transcript.
Questions & answers
What distinguishes cargo route planning?
Its loading, handling and inland distribution requirements make the network economics different from passenger services.
How does Miller describe a valuable forecast?
A causal chain exposing what happens next to aircraft, capacity or labour—not a single headline number.
Why does India’s growth not guarantee airline profits?
Repeated capacity additions can depress yields before operational discipline has matured.
Does the low-cost airline model still work?
He points to operators such as Ryanair and distinguishes the model from carriers whose execution and costs have deteriorated.
Where might AI improve outcomes?
Specific machine-learning forecasts for maintenance planning and engine-part lifecycle management.
Sources & attribution
Original interview: The Indian AvGeek — Episode 54. Published 16 November 2025.
Editorial context: This is a researched interpretation of the original conversation, not a verbatim transcript. Guests’ statements, forecasts and business or policy assessments are presented in their original 2025 context; they should not be treated as independently verified current conditions.
Editorial Q&A: Questions and answers are explanatory paraphrases, not quotations from the speakers.