L34: Managing Demand Fluctuations

Good morning, everyone. Welcome back to MGA-301. Last lecture we established the fundamental demand-capacity challenge in services and introduced the two strategic approaches for managing it. Today, Lecture 34, we go deeper into Managing Demand Fluctuations — specifically yield management, dynamic pricing, and queuing theory. [0–10 minutes: Introduction] You have just searched for a flight from Goa to Mumbai on IndiGo. The base fare shows at one thousand two hundred rupees. You bookmark it and come back two days later — now it is two thousand three hundred. You book it. Your friend books the same flight a week later — she pays seven hundred and fifty rupees. Same flight. Same seat class. Three very different prices. This is yield management in action — one of the most financially powerful and sometimes most maddening tools in the service management arsenal. Yield management, also called revenue management, is the practice of using variable pricing strategies to maximise revenue from a fixed, perishable capacity. Today we unpack how it works, why it is effective, the conditions under which it makes sense, and how to manage customer reactions to price variation. [10–40 minutes: Core Content] Yield management originated in the airline industry. American Airlines developed its SABRE reservation and pricing system in the 1980s and used it to vary fares based on booking lead time, load factor, and customer segment. The results were so dramatic that every major airline, hotel chain, and car rental company in the world eventually adopted some version of revenue management. The basic logic is elegant. A service firm has fixed capacity — say, two hundred aircraft seats on a given route. Once the plane departs, every unsold seat generates zero revenue and its value is permanently lost. The goal is to maximise total revenue from those two hundred seats. If it sets one fixed price, it will either sell out early (missing the opportunity to charge higher prices to late-booking business travellers) or leave seats unsold (because leisure travellers who would have bought at a lower price were priced out). The solution is to offer multiple price points to different customer segments at different times, calibrated to the relative price sensitivity of each segment. Leisure travellers are price-sensitive but book well in advance — they receive lower fares in exchange for advance commitment. Business travellers are less price-sensitive but book close to departure — they pay higher fares for flexibility and availability. The yield management system dynamically adjusts the allocation of seats at each price level based on real-time demand signals. Conditions under which yield management works most effectively: fixed, perishable capacity; variable demand with different willingness to pay across segments; the ability to segment customers by booking behaviour and price sensitivity; and an inventory control system capable of managing multiple price levels simultaneously. In the Indian context, yield management is now standard in: airlines — IndiGo, Air India, SpiceJet all use sophisticated revenue management systems; hotels — the Taj Group, Marriott, Leela, and OYO all vary room rates by day of week, season, and booking lead time; railway berths — IRCTC's dynamic pricing on certain premium trains; car rentals — Zoomcar and self-drive services. The technology behind yield management is a demand forecasting system that analyses historical demand patterns, current booking velocity, competitor pricing, and event calendars to predict demand at each price point. A hotel in Goa will have dramatically higher rates during December 25 to January 1, during the Goa Carnival in February, and on long weekends. It will have its lowest rates during September and October when demand is lowest. These are not arbitrary price changes — they are the output of a revenue management system optimising revenue across the entire demand calendar. Queuing theory. Queuing — managing the lines of customers waiting for service — is both an operational science and a significant service experience issue. The mathematical study of queuing was developed by Danish engineer A.K. Erlang, originally for telephone traffic management. The core insight: queuing systems have predictable properties based on three parameters: arrival rate (how many customers arrive per hour), service rate (how many can be served per hour), and the number of service channels (how many servers are available). When arrival rate approaches service rate, average waiting times increase non-linearly — a small increase in arrival rate near capacity causes a dramatic increase in waiting time. From a services marketing perspective, the operational efficiency of queuing is less important than the psychological experience of waiting. Maister's Laws of Service provide the key insights. Maister's First Law: Satisfaction equals perception minus expectation. If you wait fifteen minutes and expected to wait ten, you are dissatisfied. If you wait fifteen minutes and expected to wait thirty, you are delighted. Managing waiting expectations is as important as managing actual wait times. Maister's Second Law: Occupied time feels shorter than unoccupied time. A ten-minute wait during which you are doing something — reading, watching a video, filling in a form — feels shorter than a ten-minute wait standing doing nothing. Service firms should fill waiting time with productive, engaging, or entertaining activities. Other key observations: unexplained waits feel longer than explained waits — tell customers why they are waiting and how long they have left; uncertain waits feel longer than certain ones; unfair waits cause more dissatisfaction than fair ones of the same length — if someone who arrived after you is served first, the emotional response is disproportionate. Practical waiting management tools: appointment systems that spread demand in time; virtual queues (like Ola's "reserve your spot" function) that allow customers to wait wherever they choose rather than physically standing in line; real-time wait time displays; and occupying customer attention through app-based interactions, in-environment entertainment, or guided tasks. [40–55 minutes: Activity and Discussion] Yield management simulation. You are the revenue manager of a 100-room Goa beach resort. Design a pricing strategy for the next four months: November (pre-peak), December (peak), January (peak), February (post-peak). Your costs require at least fifty percent occupancy to break even. Leisure tourists who book in advance pay an average of four thousand per night. Business and last-minute travellers pay up to ten thousand. In pairs, design: a pricing strategy for each month, a promotions strategy for the slower months, and an overbooking policy to handle cancellations. Eight minutes, then share. [Allow eight minutes. Debrief with focus on trade-offs between occupancy rate maximisation and revenue per room.] Discussion question: Yield management creates different prices for the same service. A business traveller on the same IndiGo flight as a student may pay five times more for the same seat. From a fairness perspective, how should we think about this? Is dynamic pricing inherently unfair, or is it a rational market mechanism? Does the answer depend on the service category? [Healthcare and essential services are different from luxury travel — a good framework for discussion.] [55–60 minutes: Summary and Assignment] Today we covered yield management — its logic, conditions for effectiveness, and application in Indian aviation, hospitality, and rail. We examined queuing theory and Maister's Laws — the insight that waiting psychology is as important as waiting time, and that occupied, explained, and fair waits feel shorter and generate less dissatisfaction. Assignment: Next time you wait in a queue at any service — a bank, a hospital, a food court — observe and record: how long did you actually wait, how was the wait managed, how did the firm communicate expected wait time, and what could be done differently? Write a one-page reflection using Maister's frameworks. Next lecture — Lecture 35 — we examine Productive Capacity in Services in more depth — how service firms measure and make strategic decisions about expanding or contracting their capacity constraints. See you then. Thank you.