How have you changed your marketing strategy and spend, during this time? Given what you’ve just said about demand, how are you thinking about changing marketing spend?

In two ways. First of all, we’ve stopped completely, the so-called non-performance marketing. We have a very clear separation between performance and non-performance. The non-performance has been stopped, because we wanted to understand, because we saw that things were deteriorating very fast. We wanted to see where the bottom was. We’re not yet seeing the bottom, but at least, things are starting to calm down a bit, from a medical perspective. We’re going to continue doing campaigns, but we’re changing the tone, the messaging, the copy, even the platforms on which we are communicating. That’s one thing that we’ve changed.

On the more performance marketing spend, the distribution marketing, if you will, we’ve changed, for example, the CPC models, to pure success-based models. This means that we are not doing any campaigns on Google, for which we pay for a CPC that we then need to track if it becomes a conversion which, eventually, becomes a stay, so we can deduct the ROI from that campaign or that click, specifically, and see how efficient or inefficient it is. We’ve shut that down completely and we have activated risk-free bookings, which means we only pay, not at the time of booking, but post-stay. The different partners with whom we work, metasearches, like Google Hotel Ads or Skyscanner, or even some affiliation platforms, we’re saying, continue selling; we’re happy for you to sell, but we won’t be paying you until the customer has checked out. This puts us in a very comfortable position, because it’s risk free. We don’t want to kill the business that’s coming in, those bookings, as I just mentioned, for Q4. We want to stimulate that demand, as much as we can, but in a logical and smart way.

So those metasearch engines, such as Google, they’ve been pretty flexible in adapting their model, in that sense?

They already had these models in place, such as the commission program for Google Hotel Ads; that was launched a few years back. Some of them have adapted quickly, to cater for or bring new solutions to this space. There’s another interesting aspect. I was looking, yesterday, at overall CPC costs and so forth. For our perimeter, meaning the markets in which we operate and advertise and the audience to which we advertise, the CPC costs have plummeted, by over 60%, in the last four weeks, which is logical, because there are less people advertising. This means there is less pressure on the auction and, as a result, that traffic is cheaper, which probably opens an opportunity to generate new bookings. We will probably be doing that in the short term. We have moved to a completely risk-free model, but we might be coming back to the CPC. Even though we know that some of the reservations that are going to be made might be cancelled down the line, because of the volatility is almost 100%, we prefer to stimulate bookings and receive those bookings, in order to keep the wheel turning and hope turning and try to find the new normal, as soon as possible.

CPMs and CPCs have both been declining but, also, the conversion is still pretty weak, as well, because people are not actually buying, because they are uncertain, but there is that potential to spend on cheaper traffic to, potentially, have that booking?

There is, overall, much less traffic, which is a problem. The only good side is that the traffic that there is, is slightly cheaper. We are also changing the way we measure success. So from a pure, complete booking, which is the traditional way of really measuring the conversion rate, we are looking, in some cases, at micro-conversions. For example, we are trying to direct as much traffic as we can to our contact centers. We’ve got contact centers in the Americas and in Europe. We are seeing extraordinary results from voice interaction, which is obviously more expensive. But we are also seeing higher ticket prices and better ability to convert. In these times of uncertainty, people want to know things. They ask a lot of questions which maybe could be answered on the web, but if there is a human at the other end of the line, people prefer to interact with a human. We are considering a conversion, a call that has been provoked by an original click and, not necessarily, a conversion with a credit card on the website. We see that as one factor of converting; either we are calling them back or they are calling us at a contact center. We are altering our measures, slightly, to make sure that we capture everything possible, out there.