Using a Bow Tie to Make Optimal Marketing and Technology Investments
The Croatians are credited for inventing the cravat which then evolved into the bow tie, but as successful as they have been in pushing Dubrovnik and the Dalmatian Coast as a tourist destination in recent years, they haven’t yet used the bow tie to try and explain consumer purchase behaviour in the travel industry. In recent months I’ve been thinking about two upcoming conference presentations I am due to give in May, first in Miami and then in Bangkok, and how best to explain my thesis that airlines can maximize return on investment by putting more focus on the time between ticket purchase and flight departure. I’ve decided to refer to this as the Bow Tie Model.
Late last year I asked a colleague to trawl through all of the bookings created in one month using aggregate data from the approximately 80 airlines using the Amadeus e-Retail internet booking engine. My question was this: Tell me the average number of days between purchase and date of departure. I also asked this same question for products that would typically be referred to as ancillary revenue. The results were as follows:
- Air: 44.1 days
- Hotel: 41.7 days
- Rental Car: 19.4 days
When I presented these results at the Ancillary Revenue Airline Conference in Budapest in November 2008, I referred to it as the Late Buy – Early Buy Continuum (slides, audio). The purpose was to show that airlines focusing solely on website cross sell were not targeting passengers at the most likely point of purchase. In short, unless you are considered a destination website for non air content, you will not be optimizing ancillary revenue just by adding hotels, rental cars and destination content to airline.com. The disparity in timing of purchase between air and non air became even more apparent when looking at international travel and online redemption of frequent flyer travel. Using one major airline to illustrate this point, we found the average days prior to departure for purchase of air tickets was 76 and 82 days respectively. “Destination content” from a tiny sample size came in at 24 days prior, but due to low volumes, ineffective marketing, measurement being on days to consumption and not departure of first air segment etc, I’d recommend totally ignoring this number. When done properly I would expect to see an average of around 7 days for this category of product.
So how did the Late Buy – Early Buy Continuum evolve into the Bow Tie Model? And why is it relevant to direct sales channel managers within an airline? Part of the answer lies in this diagram:
Attend any conference on travel industry innovation, and the majority of start-up companies present will be competing in the phase from the moment a person decides they may want to travel somewhere up until the point in time when the decision is made on where to purchase the flight. Apart from an endless supply of start-ups, you also have established metasearch sites, destination guides, trip blogs, photo sharing sites, opinions from friends and family, and even government tourism boards all competing to influence the consumer’s purchase decision. Google research found that the average traveller spends 6.7 weeks searching the web and performs 8.1 travel related searches before booking. Over time the range of influences is narrowed down, a destination is chosen, and a flight is purchased. At the exact point of buying the flight there are no other influence factors, and immediately thereafter the destination is committed and a whole new phase begins, with an ever widening source of influence factors coming into play regarding the chosen location, right up to the point at which the person arrives at the airport to board the plane. This is this phase where influence translates directly into dollars.
The final part of the above diagram is shaded grey, as one challenge for the airline is to bring forward as much of this passenger expenditure as possible into the phase between air ticket purchase and day of departure. In-flight sales compete with airport duty free and cause excess inventory to be carried on board. Purchases made when the passenger arrives at their chosen destination yield no ancillary revenue to the airline. For these two reasons, and the fact that on board sales is not my area of expertise, I will not address these two phases any further.
This model is still very much a work in progress, but I intend to continue working with existing and new airline customers to fill out the different phases with hard numbers to back up these assertions. It is becoming clearer to me by the day that in an environment of financial belt tightening and increased scrutiny on ROI attached to marketing and technology innovation, that an airline putting a heavy investment focus on trying to be a leader at the point of “inspiration” is missing the point; especially when so many more agile competitors, including competitors without a direct revenue objective already dominate. The smart and successful airlines will be the ones investing in more effectively selling at the points where they have much better information than anyone else; this is during the booking flow and especially once the passenger name record has been created.