Emmanuel Scuto, CEO of WeYield, expert in car rental optimization :

Have you ever found a yield manager not stressed? How many of you got a bit of anxiety finalizing his/her capacity optimization of a train segment, for a peak departure day at the airport for an airline or in August at La Croisette hotel in Cannes? Let him who is without sin demand spoilage or price under-optimization cast the first stone!

Back in 1999, I remember my first inventory control experience at Disneyland Paris, managing the 7-hotel properties with 6000 rooms to fill. Orders were clear enough: do 100% occupancy from 1st July till 31 August. To do so, it was (still is I imagine) a mix of high season prices calibrated by market sources with reservation flow controls to avoid dilution combined with a « smooth landing » on the d-day to deliver a full resort with a minimum of over-booking. During every evening in Summer, my team and I were on duty to make sure that the booking arrival instructions were correctly applied to fill every single property and manage overbooking in a smooth but efficient way to close without any capacity empty. The next morning briefing was pretty intense if the goal was not achieved!

Nowadays, with instant performance measurement tools in place and competitor benchmarketing, doing good for the Yield manager is not sufficient: doing better than the objective and better than the competition is crucial. Tools like STR Global are enabling to share average daily rate and occupancy of last day in a click. MPI (Market Penetration Index), RGI (Revenue Generator Index), and ARI (Average Rate Index) are the « new instrument of psychological torture putting more pressure on the Revenue Manager. To optimize even better and reach the Grale, many algorithms and mathematical models have been invented for RM teams with more and more complex tools. However, despite all these software and applications, implementation efficiency remains based on humans.

In 2011, US Professor Elliot Bendoly from Emory University (Atlanta, Georgia), expert of human behavior in operational management, realized a study called « Linking Task Conditions to Physiology and Judgment Errors in RM Systems». He then published an article in Performance (volume 5, edition 4).

ARE YIELD MANAGERS GOOD ENOUGH TO MAKE EFFICIENT DECISIONS IN AN UNDER PRESSURE RM ENVIRONMENT?

The author conducted an experimental process to test pricing decisions vs available capacities over a particular period of time (lead time). The goal of the Yield Manager is to sell the maximum unit of C-Capacity over a period of T-Time to maximize expected revenue. During every stage, Yield analysts got various scenarios to play with (combining price and volume) to accept or to deny: for each of them, a revenue expected was computed. Every denial meant that the Yield analyst was expecting a more profitable situation to happen at the next scenario either in price or in a volume of sales to maximize his revenue. He must also take into consideration that there was no other scenario available and therefore lose all gain opportunities. The study went even further, trying to evaluate the Yield analyst motivation impact on the performance by measuring eye blinking frequency, a good indicator to evaluate stress. Three categories of behaviors were highlighted :

1. the indifference of the analyst or how lack of motivation and sufficient challenge for his task generates a superficial and distant analytical commitment
2. awakening of the analyst or how with a little more challenge on every single task can generate a much higher level of attentiveness resulting in enhanced performance.
3. desertion or how a given situation seems to be frozen or blocked to the analyst, without any room for improvement, generates a high level of stress and demotivation.

These three behaviors may change in intensity depending on the environment: multiple sites or a single site or stock to optimize.

As a starting point, managers should be aware of the conditions under which RM decision-makers will make suboptimal decisions, reject or accept errors specifically. By doing so, they can begin to monitor for the root causes of such behaviors and reduce their frequency” wrote Bendoly.

How to do that? By settling more training and dedicating more resources to improve and sustain awareness of these dynamics over the performance. Also, reducing the size of capacity to be optimized by cutting the inventory in slices and allocating more time to analyze and take actions in advance to reduce pressure naturally generated closer to the d-day.

‍