For several years the travel industry has heavily debated "the billboard effect" – the media impact of a hotel's participation in an online travel agency (OTA) merchant program. While direct bookings are easy to track, it has become evident that OTAs, via their promotional capabilities, also account for bookings made on other channels such as the hotel's own website. In 2009 and 2011, Professor Chris Anderson of Cornell's School of Hotel Administration conducted the first published research designed to measure that impact. A 2012 report, Distribution Channel Analysis: A Guide for Hotels (DCA) , summarized many of the criticisms of Professor Anderson's research and then excluded OTA media effects in an evaluation of the costs of OTA participation. Phocuswright commented on the DCA report in a recent spotlight
(see Sage Advice for Hotels: Another Point of View on a Popular Industry Report).
- OTA Display Effects
- Measuring the Media or Billboard Effect
- PhoCusWright Response to Criticisms of Professor Anderson’s Studies
- In Anderson’s initial study, when hotels were “on” Expedia, they appeared on the first page. This may not be a realistic scenario for any hotel and all circumstances. Reported results overstate the effect if the hotel does not consistently appear at the top of the Expedia (default) display.
- Anderson’s studies did not mention whether the subject properties had lowered rates relative to competition during the study periods.
- In his second experiment, Anderson estimated an average 1:1.5 ratio for OTA to Brand.com bookings across thousands of independent properties. DCA research found that the industry average ratio of OTA to Brand.com bookings for all hotels was 1:1.45, which would mean that over 100% of Brand.com bookings came from OTAs (in this case, Expedia). This is unlikely
- It was not clear in the second study why consumers were on Expedia – to shop air, hotel or car – before booking a hotel on Brand.com. If known, this information could have deflated the attribution assigned to Expedia’s presence in the clickstream.