Why being passive can be unhealthy for revenue management
Too much intervention can be problematic but a balanced and active approach to airline RM works best. Tom Bacon reports
‘Passive’ revenue management (PRM) takes human intervention out of the equation. It means complete reliance on the sophisticated analytical model for setting fares – letting the sophisticated forecast and optimisation model allocate inventory across fare levels, and price dynamically based on a vast database of historic performance.
‘Passive’ means no analyst intervention, nobody trying to ‘outthink’ or override the sophisticated model. One old study showed that analyst intervention was often counter-productive, reducing revenue relative to PRM. Potentially, ‘passive’ management produces better results than constant intervention by even highly trained business analysts. It is also true to say that effective analyst intervention certainly requires understanding the nuances of the model, as well as correctly interpreting information outside the model.
A simple way to measure the success of any RM strategy is a calculation of ‘spill’ and ‘spoilage’:
Spill is the opportunity cost of accepting a lower fare when the seat could be sold to a higher fare passenger who books later.
Spoilage is an empty seat that occurs because the airline foregoes a low fare booking in anticipation of a higher fare passenger that doesn’t materialise.
So, what are the implications of an industry that embraces ‘passive’ RM?
Slow adjustment to market changes. A passive management philosophy will be subject to relatively slow adjustment to large market changes. When a significant increase or decrease in demand occurs, the model will be reluctant to immediately adjust the forecast going forward. It will tend to require a more consistent trend before fully adjusting the forecast.
The point to note here is that if the whole industry is slow to adjust to market changes, the entire industry will see spill or spoilage associated with suboptimal performance. If actual demand is weaker than the forecast, then the model will set aside seats for demand that doesn’t materialise, driving spoilage. Similarly, a sudden increase in demand, not immediately incorporated in the model, will drive spill.
Slow adjustment to airline-specific initiatives. The model will similarly adjust its forecasts gradually in response to airline-specific marketing, scheduling, or product innovations. Just as the model won’t pick up sudden changes in overall market demand, the model can’t rely on history to identify airline-specific initiatives that drive shifts in demand. If a schedule or product change drives a sudden increase in demand for a specific airline, at the expense of the competition, the model will be slow to adjust. In this case, the airline with the increased market appeal is likely to experience spill until the model catches up; its competitors will see spoilage as demand shifts to the innovative carrier.
Here the point to note is that this mitigates the value of innovation. Unless analysts intervene, the model will cause the airline to be slow to realise the whole revenue value of any change.
Lockstep pricing. If the whole industry relies on the same history, similar demand patterns will drive industry pricing. Collective reliance on historic patterns will show up in industry-wide pricing patterns, including seasonal, day of week, and time of day variations.
The asymmetric value to being out of step with the industry has this impact:
If all competitors rely completely on the model, then intervention by a single airline has asymmetric results. If the market is strengthening – and the model is slow to adjust – a competitor that correctly perceives the market change will not necessarily be able to unilaterally raise fares. The availability of low fares by the competition will limit any benefit it may see by its accurate market assessment.
On the other hand, if the market is weakening more than the model, the carrier that correctly perceives this can hold out lower fare inventory longer and gain a market share advantage.
This asymmetry could lead to an industry propensity to lower fares.
Such ‘passive’ management by the industry is not healthy.
If all airlines pursue such a passive strategy, the entire industry is held back in capitalising on market or airline-specific changes. Given that changes are constant in the industry, industry revenue is constantly underperforming relative to its potential.
If airlines assume its competitors will be ‘passive’, they are less incentivised to be ‘active’ even when they correctly predict market changes. Since such active management only pays off when the market change is negative (demand is reduced), the industry will tend be biased to lower fares.
Fortunately, no airline has fully adopted a ‘passive’ revenue management strategy, even if academic studies support such an approach. All airlines employ RM analysts whose job is to oversee – and regularly override – the model. Although analyst overrides are often too frequent or too large, the industry as a whole may perform closer to potential if revenue management is ‘active’.
Tom Bacon has been in the business 25 years, as an airline veteran and now industry consultant in revenue optimisation. He leads audit teams for airline commercial activities including revenue management, scheduling and fleet planning. Questions? Email Tom or visit his website