Improving revenue management offers great potential value because it goes right to the bottom line without additional capital spending. Moreover, it is needed in a world where pricing is getting faster.
What exactly is revenue management? The application of disciplined analytics that predict customer behavior at the micro-market level and optimize product availability and price to maximize revenue growth. It is an example of “analytics in motion,” which uses real-time streaming data (that can be big, fast and noisy) and dynamic optimization to improve the quality of an organization’s continuous decisions such as: pricing, order acceptance, resource allocation, scheduling, routing, and forecasting supply/demand.
American Airlines pioneered revenue management in the 1980’s when it innovated real-time price/capacity variability to fend off the post-deregulation threat of low-cost airlines. The new practice was considered a “spectacular success” that “saved the business.” A 1992 article estimated quantifiable benefit at $1.4 billion over three years and expected annual revenue contribution of over $500 million going forward.
Marriott, which also had perishable inventory, customers booking in advance, lower cost competition, and wide swings in supply and demand, applied the same methodology of scientifically varying prices to hotel rooms. By the mid-1990’s, revenue management added $150-$200 million annually in documented revenues.
Setting the Table
1. Start with a limited, usually perishable product, such as an airline seat, hotel room or truck. Note that network value can be an important additional requirement e.g. “We need 3 trucks in Chicago in 2 days.”
2. Identify different customer market segments for that product:
- Different price sensitivities
- Different things they value
- Different buying behaviors
- when they buy
- where they buy
- how they buy
You might begin first with premium segments and work down from those willing to pay the highest price or contribute the highest profit.
3. As the “spoilage date” approaches, dynamically:
- Re-estimate the demand from each segment (usually with a distributional forecast)
- Change the quantity and/or price offered to each segment
The objective is to optimize expected total revenue or profits.
In B2B, repeat customers will adapt their behavior to your dynamic pricing policy. If you drop prices at the end, they will wait to bargain shop. If you simply decrement bundled capacity, they will give you the lowest value orders early and deprive you of capacity you could have sold at greater yield later. Once you tell them you’re out of capacity, they probably won’t call with additional orders—depriving you of valuable information on demand.
Thought-provoking Successes
As mentioned earlier, Marriott was a successful early adopter of revenue management, but it was initially used only for individual reservations. Group bookings, a key driver of profit, were individually negotiated by the field sales force with a weekly minimum price guidance. Marriott developed a group price optimizer that provided real-time revenue management quotes for groups based on; desirability of the date, type and size of the group, and up-to-the-minute forecasts and bookings. The result was a revenue increase of $46M in the first year and $76M in the second year—during an industry recession with decreased individual room bookings. I examined this case in my book, The Optimization Edge.
In another example, the intermodal division of a Class I railroad sought a new business model that leveraged optimization to provide the immediate, real-time presentation of route options, schedules, capacities, and price directly to shippers and transportation brokers. Historically, the business focused on wholesale pricing that relied on third-party agreements and rack rates for intermodal marketing companies. Princeton Consultants worked with the business on a real-time system that accommodated spot truckload conversion sales. These sales yielded higher margins than those of traditional annual-commitment freight. Many came from truckload customers who had not recently used rail/intermodal and were favorably impressed. Many of the new shippers became repeat users and some even converted to high-volume annual-commitment freight. More details about this case are here.
For a high-frequency trading (HFT) hedge fund, Princeton Consultants designed and built a real-time system that leverages revenue management on both the buy and sell side. Using advanced forecasting, low latency cloud data processing and optimization, the firm trades US equities based on big data signals processed in micro/miliseconds. More details about this case are here.
If you would like to explore high-value revenue management opportunities at your organization, my colleagues and I would welcome a conversation. Contact us to set up a call.