April 2018, San Francisco
Why airlines need to segment smarter
People travel for many different reasons but often airlines don’t go far enough in meeting their specific needs, writes Tom Bacon
Marketing textbooks generally speak of customer segmentation based on demographics. We are talking millennials vs baby boomers, singles vs young families, affluent vs the budget or, perhaps, value conscious. Some marketing firms track 20 or more ‘psychographic’ segments that exhibit different buying habits and propensities.
Airlines, on the other hand, typically use customer segmentation based on behaviour. The underlying segmentation is generally ‘purpose of trip’ (business vs. leisure) rather than customer age or affluence. Since business travellers tend to book their trips closer to the departure date and often demand additional ticket flexibility, airlines can take advantage of this with fares that vary significantly by days-before-departure (DbD) and refundability.
Days-before-departure is the most important driver of fare levels, and the most popular indicator of price sensitivity
These factors are used as indicators of a customer’s price sensitivity. Restricted fares (non-refundable fares booked far in advance) are the lowest and designed to meet the needs of the most price sensitive travellers. Full fares (the highest fully refundable fares) can be six to eight times the value of the lowest fares and are designed for business travellers who need that flexibility and are willing to pay for it. The range of fares in-between appear, in increasing value, between six weeks before departure and the day of departure. As such, days-before-departure is the most important driver of fare levels, and the most popular indicator of price sensitivity.
Missing a trick?
DbD is obviously an imperfect proxy for business vs leisure, and an imperfect indicator of price sensitivity. It is also means that airlines are missing a trick in maximising airline revenue. Without improved segmentation, airlines also miss out on new merchandising opportunities related to ancillary purchases. Questions airlines should be asking include:
· What customer segments want which additional features/amenities above the base travel product?
· With the new array of ancillary choices, how can airlines point customers to the ancillary features they most value?
Here, applying days-before-departure is a particularly weak approach. After all, certain leisure passengers booking far in advance may well wish to check their bags or purchase adjacent seats just like a business traveller.
For both reasons – improved estimates of price sensitivity and for merchandising ancillary products – airlines need a more refined way to segment customers.
Airlines need a more refined way to segment customers
Ian Williams, a principal consultant for Amadeus calls for a more refined approach to customer segmentation. The methodology is still based on behaviour: certain behaviours, he argues, reflect different ‘purpose of trips’. Williams suggests that airlines:
· Assemble a group from across multiple airline functions to identify possible segments around purpose of trip.
· Hypothesise the behaviors of the people within these segments and the factors within the booking and ticketing data that help describe these behaviours (for example, DbD, length of stay, day of week, time of day for both bookings and travel).
· Conduct a statistical cluster analysis around the hypothesised factors that are believed to differentiate the proposed customer segments. What does the data say? What attributes work together to form clusters of behavior?
· Compare #1 and #2. Then ask the questions: Do the hypothesised clusters emerge from the data? Are there clusters that don’t match up with the hypothesis? What could those clusters mean?
At one airline, Williams’ methodology led to airline was further breaking down business and leisure into sub-segments. He identified seven such segments; four business and three leisure. Business segments included both commuters (short haul; out on Monday, back on Friday) and seasonal workers (long stays); leisure included a differentiation between weekend vacationers and weekday vacationers.
The value of this new segmentation lies in both potential dynamic pricing and merchandising. Commuters are likely to value certain features far more than seasonal workers. And, although leisure is certainly more price sensitive than business in general, weekday vacationers (older, wealthier, less tech savvy) may value certain amenities more.
Airlines need to approach customer segmentation in new ways. Finding different ways to segment can offer increased opportunities for pricing differences and form the basis for more effective e-merchandising of ancillary services.
Tom Bacon has been in the business for 25 years, as an airline veteran and industry consultant in revenue optimisation. He leads audit teams for airline commercial activities including revenue management, scheduling and fleet planning. Email Tom or visit his website.