A recent study raised an interesting debate about the livelihood of the infamous billboard effect. The study conducted by the American Hotel & Lodging Association’s Consumer Innovation Forum reported that the 2009 Cornell billboard effect research is no longer valid.
The billboard effect is the direct correlation between visibility on an OTA site and direct bookings. Chris Anderson conducted the original study at Cornell in 2009 and it has been a widely accepted phenomenon ever since.
However, the new study offers a lot of data that they say proves the billboard effect is dead. The study looked at various key indicators from 2012 to 2014 and made some pretty bold statements. First, the study broke up consumers into three categories:
They then show how these same consumers behavior has changed over time.
On the surface, these numbers indicate that consumers who visit intermediary sites are rising. And, those who visit only hotel sites or both sites is decreasing thus reducing the billboard effect. But, first let’s look at the other numbers they report:
The new study acknowledges that these numbers may seem good for hotels on the surface, but it is not because of the disproportionate amount who people who no longer visit both sites.
If the study is trying to say that less people shop at both types of sites in general, I believe that to be true. But, to say that the billboard effect is no longer real, is a bit of a stretch. Consumers are still shopping on OTA websites and then returning to the hotel website, they proved that. The billboard effect does not come with a certain percentage threshold, it simply says that presence on OTAs increases direct bookings. To me, these numbers prove just that, OTA presence drives direct bookings.
The study also provided these metrics:
Again, the study reports that this is only good news for hotels on the surface because a consumer who visits an intermediary site and returns in their next session is 90%. Meaning, only 10% of the people who start on an intermediary site will go on to a hotel site in their next session.
While it is obvious that intermediary sites are good at retaining their customers, that doesn’t disprove the billboard effect.
It is also important to note the differences between the new study and the original Cornell study. The new study used clickstream data from a market research firm and was then analyzed with attribution modeling. The Cornell study used a preset number of listings from Expedia’s first page results that were methodically cycled on and off the site. The listing would stay up for 7 to 11 days and the resulting direct bookings were recorded.
I think causation is easier to determine when you are looking at specific listings instead of making assumptions through a collection of clicks. The properties likely to benefit the most from the billboard effect would be those that appear on the first page of search results. The new study pulls data from all types of properties, which may be better to tell an overall story, but what do the numbers look like for an OTA’s most popular listings?
In order to accurately prove or disprove the billboard effect’s livelihood, Cornell or another research firm should replicate the original study. Using the same methods, a duplicate study could compare the state of the billboard effect to its 2009 results.
Both studies also neglect to include mobile searches. Tnozz agrees that in 2009 this probably didn’t affect the study significantly, but the new study has done itself a disservice by not including mobile data. Mobile booking accounts for more than 50% of travel searches and 25% of travel revenues. That sounds like a lot of missing data to me.
And even more, Frederic Gonzalo offers that in reality, OTAs are only one piece of the puzzle. And I agree. A hotelier should drive bookings through many different channels. OTAs are only one distribution channel that should be utilized when marketing a property. Gonzalo notes the best ways for hotels to market their hotel including:
At Cloudbeds we strongly believe in creating a dynamic marketing distribution strategy. While we do not believe the billboard effect is dead, it is never good to rely on one channel to drive revenue. It is obvious that OTAs have become stronger since the 2009 Cornell study, but many of the same benefits still exist.
If the death of the billboard effect scares you, or you find yourself heavily relying on OTA channels, check out our posts on how to strengthen your distribution strategy.