EyeforTravel North America 2018

October 2018, Las Vegas

Understand how data, digital and partnerships can make your marketing work again

Why applying probability to personalisation could be the answer

How confident are you that your marketing messages really are engaging your audience rather than infuriating your customers? Tom Bacon considers how two companies are tackling the challenge

‘Personalisation’ is terrific in concept. The idea is to create a highly targeted campaign – including the ‘right’ price, product, messaging, channel – that is customised to the desires and needs of an individual traveller. Certainly, this is the way to drive higher conversions, and with a strong probability that the traveller will respond positively with a purchase.

Unfortunately, most suppliers don’t know their customers that well. Perhaps they have considerable information about their most frequent, loyal customers but the majority of their customers are not so well known. Even if suppliers have access to multiple customer databases and touchpoints – emails, credit card purchases, social media posts, for example – suppliers don’t typically know how to sync with the mass of information. 

How confident are you that the email address for a ‘Tom Bacon’ is really the ‘Tom Bacon’ who purchased your service on Facebook? Or that it’s the same ‘Tom Bacon’ whose residence is listed in public records as Denver? 

Such highly targeted campaigns – designed to yield the highest conversion rates – may be too targeted to drive the desired marketing results. In fact, broader campaigns are often the preferred tool for driving more significant new purchases.

Two examples of companies that claim to be able to extend personalisation to much larger audiences with their analytical capabilities are Amperity and Wiland, both sponsors of recent EyeforTravel events.

1. Tackling risk

Amperity helps clients design campaigns around the probabilities of customer identification. At an Eyefortravel conference earlier this year, Stuart Greif, Lead for Travel & Hospitality at Amperity, talked of the challenge consumer businesses have in sorting through the various identities individuals have across channels and touchpoints. Amperity works with the various references and touchpoints associated with ‘identity’ and assesses probabilities that multiple contact points apply to the same person.

According to Greif, based on the estimated probabilities of customer identification, Amperity creates personalised messaging. Their approach attempts to be as targeted and customised as possible, but they recognise the risks of inaccurate identification. A misquided personalised campaign, for example, can have the opposite impact of what is intended: it annoys and infuriates rather than engages the target customer. A less targeted, more generic message is designed for more ambiguous situations. For example, since one can’t be sure that the Facebook purchases were by the same ‘Tom Bacon’, those purchases won't be incorporated in any new highly targeted campaign.

2. Ultimate data

Ultimate Data, a division of Wiland, explicitly designs custom audiences around the likelihood of customers to respond to a specific campaign. They score customers based on analytical models of purchase behaviour, intent signals and social personas. Using a vast consumer database, they can offer multiple-sized audiences based on a ranking of targets and consistent with a client’s specific goals. Each audience they offer needs to have an appropriately customised marketing plan designed to maximise response for each campaign. 

Wiland began as a direct marketing firm providing direct mail audiences to its clients. In direct mail, marketers strive to limit high-cost mailings to high-value targets. In the internet world, Ultimate Data offers a similar analysis of ‘high value’ targets – and claims to bring direct mail-like precision to digital campaigns. It also offers larger audiences based on explicitly calculated purchase probabilities.

Many travel merchandisers lament the fact that they ‘know’ only their most frequent travellers, key members of their loyalty programmes, for example. They reject broader ‘personalisation’ since customisation doesn’t work for a completely anonymous purchaser. However, as Amperity and Wiland would attest, one can apply probabilities to personalisation for less frequent travellers and still tailor the merchandising approach (product, price, content, messaging, channel) based on the probability that the traveller has particular characteristics and desires.

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

Related Reads

comments powered by Disqus