When I began my career in the fascinating world of air cargo, I worked with a brilliant team that processed all available data, both internal and market, and compiled it into a dynamic Excel report. With a variety of filters and regional segmentation, this report was essential for daily decision-making, supporting the commercial team in executing sales and shaping long-term strategy.
That strategic planning process took weeks to complete, relying heavily on manual inputs for demand forecasts, pricing, and capacity, along with multiple iterations to meet the revenue targets set by the company.
Each year, our analytics team introduced new features, and we were constantly amazed by the improvements they made. More historical data, additional input parameters, and business rules enabled the creation of a model that automated not only data processing but also the efficiency of revenue and contribution calculations. This powerful model took an entire night to run and had a dedicated desktop computer to speed up execution. If the commercial team had new information, operational plans changed, or RM adjusted seasonal pricing, we had to rerun the model, posing a major challenge for maintaining targets amid shifting inputs.
Daily operations in such a dynamic environment also involved multiple spreadsheets, market data, ad hoc calculations, and commercial inputs feeding macros that guided decisions on shipment acceptance or rejection, often backed by emails for post-flight traceability.
At the time, those spreadsheets, macros, and our in-house model delivered peak efficiency. However, customer management still faced delays, not only because clients couldn’t self-manage, but also because of internal validations, especially around dynamic pricing and demand fluctuations.
Welcome Automation
Today, airlines not only develop internal solutions but can also rely on cargo revenue management optimization from experts like RTS, who lead the way in designing algorithms that drive automation and continuously improve result accuracy.
A demand engine powered by machine learning, trained daily with current bookings, can now predict demand with remarkable precision, even down to seasonal patterns and class density level. This allows teams to understand market behavior, anticipate trends, and focus sales efforts in the right direction.
Defining a dynamic bid price no longer requires analysts to evaluate hundreds of markets over several days. An integrated system can now consider all relevant variables and deliver recommendations in minutes. Sales experts can then review and adjust these recommendations, using them in quotes and negotiations to maximize profitability. These dynamic bid prices can be visualized directly within reservation systems, with pre-established business rules automating booking evaluations and optimizing revenue.
In my opinion, we are leveraging AI in the right direction and in the right way.
Automation models are streamlining data processing with multiple factors and making it available to RM analysts and commercial teams. Can we imagine a scenario where AI acts as a commercial team assistant, identifying customer preferences for products, OD connection times, and making them available to the sales team to offer better quotes and rates with a higher probability of acceptance?
Smarter Booking and Pricing
Freight forwarders (FFWs) now make bookings through H2H, EDI, or marketplaces, seeking the best applicable rate. This is made possible by pricing automation, which not only calculates rates but also selects the most suitable offer based on service options, the routing options carefully selected by the model (most commonly known as business rules), or manually generated based on the user’s expertise.
Could AI soon recommend the best route for sensitive cargo, anticipating delay risks to ensure successful delivery at a competitive price? The future of air cargo sales is not just automated, it’s intelligent, adaptive, and customer-centric.