Tuesday, 5 June 2018

How Travel Brands Should Tackle Attribution


Knowing your customer is critical but complex journeys across multiple devices present travel brands with huge challenges. A new report investigates how travel brands can conquer attribution.

Travel brands should attack the issue of attribution from multiple angles says EyeforTravel’s new Understanding the Customer report, a completely free report, which can be downloaded now. Approaches range from tracking through cookies, data-driven attribution and encouraging customers to log-in to own brand sites and apps, all of which needs to be underpinned by a strong data program that measures changes over time.

Travel brands need to undertake an extensive and multi-faceted approach because attribution requires capturing interactions at so many touchpoints over extended periods and with different devices.

Mobile bookings in particular have additional complexity in terms of attribution, as mobile remains a last-minute channel, with short leads times leaving limited room to build up tracking data. In EyeforTravel’s Mobile Industry Survey 2018, 80% of accommodation business surveyed reported that mobile lead times were shorter, with 58.7% saying that they were much shorter. Just 1.3% found that mobile lead times were longer.

Click to download


This can leave brands reliant on just last touch attribution, particularly when it comes to mobile bookings, and often looking at an incomplete view of their customer’s journey.

Maria G√≥mez Bada, analytics expert at HomeAway.com, argues that travel firms should fight the tendency to analyse just the last click, and see the whole picture. She believes that Google Analytics default attribution model, which will show you your customer’s last known direct click, gives an incomplete picture. Instead, she proposes data-driven attribution models for most online, complex companies.

“Data-driven attribution understands the value of the whole path, not only considering clicks but also impressions,” she explained. “[For example], a customer has seen your ad in Facebook, doesn’t click on it but recalls your brand, and might come in later through SEO [search engine optimisation] or SEM [search engine marketing]. Data-driven attribution gives a value to each medium and channel, considering clicks and impressions. Not only that, it works with predictive models to try to understand how you can invest in better marketing channels and ultimately get more conversions.”

To do this however, you need a strong foundation of data around your customers. “We’ve implemented a data management platform that basically enables us to very much look at our audience,” says Alessandra di Lorenzo, chief commercial officer, media and partnerships at lastminute.com group. Core to this is tracking, particularly cookies: “We call our cookies essentially unique users who come to our site [and when] altogether, make up our audience. What we do is look at this audience and combine segments… that we then use to personalize all of the messages that pop up on the website. This information enables us to make the customer journey more intelligent and more profiled, and therefore more relevant, we hope, to the unique users coming to browse our pages.”

Combining data from your sites cookies with third-party tracking enhances the potential to build out profiling and increases the accuracy of data-driven attribution. “Access to a person’s behaviour from retargeting sites and affiliates helps us, with a breadth of exposure,” says Steven Consiglio, product performance manager at Booking.com. “A map site, a luxury vacation site, it goes on-and-on. [Through these] you learn increasingly more, because your learning platform is not your own page: it’s wider.”

However, Lorenzo argues that even when armed with a host of information, “Unique identification is pretty much impossible from one device to another. Our solution to this is giving customers a value that brings them to log in… [for] a more tailored experience.… We have implemented all of the social log ins to make it very seamless. It’s about exchanging value and creating something special for our users.”

“Who can argue if a probability-based algorithm is right or wrong unless you benchmark it against another probability-based algorithm or a deterministic approach – which is what you have with a log-in base? This is the power of the Googles and Amazons, Facebooks and eBays who have a massive amount of logged-in customers and can deterministically tell if you are the same person.”

Del Ross, senior advisor at McKinsey & Co and a travel distribution and digital marketing expert, adds that the most important thing is to ensure that your data methods stay the same over time, so you can understand trends. “Behavioral insights have much more commercial and strategic value than transactional data,” he believes. “Transactional attribution is useful, but it is more important that the attribution method be consistent so that changes over time can be understood.  The absolute data is less important than pattern changes, which can reveal shifts in customer preferences and needs.”

Click here to download the Understanding the Customer report, part of EyeforTravel’s Behavioural Analytics Report Series. This report features insights from:

  • Booking.com
  • HomeAway.com
  • Hyatt
  • Hertz
  • Expedia
  • Lastminute.com Group
  • McKinsey & Co
  • Stena Line