Travel is known to be among the most competitive spaces for SEO. Facing entrenched competitors with technical savvy, shifting user behavior, and Google’s moves to take over more travel search results with their own features takes a lot of hustle and smarts – and in the age of COVID, the stakes are even higher with less demand to go around.
We sat down with Antoine Eripret of Liligo.com, a French travel meta search engine, to ask for his insights on the unique challenges faced by travel SEOs and how to approach them.
Antoine: In 2014 or so I started as a freelancer in France, while I was studying, and then my first professional experience was in 2016 – my girlfriend is Mexican and doesn’t speak French, so we decided to move to Spain, where I joined a small agency. There were twenty people at the agency when I joined and forty something when I left, and we worked with e-commerce, travel, a bit of everything, to be honest.
Since the beginning of this year, I’ve been working at Liligo.com, a mainly French travel meta search engine, like Orbitz or Expedia in the US. Liligo doesn’t actually charge users, we just refer them off-site to where they can complete the purchase.
I focus primarily on technical SEO, but I get involved in content as well, depending on the projects.
Antoine: Starting with content, the biggest challenges are probably how broad the keyword options are for informational queries, as with so many options it can be hard to know where to focus, and the need for unique content for commercial query landing pages (like flight routes) to avoid duplicate content issues.
On the technical side, internal linking is probably the biggest challenge. If you think of, for example, a typical e-commerce site, where you have, say, women’s dresses, and as you go deeper in that category the search volume falls off while the conversion rate tends to go up because those users know exactly what they want, so internal linking easily aligns with search volume.
For travel, it’s not that simple, sometimes search volume is higher on more specific queries (for example, “flight to London” can have less search volume than “Paris to London”), and for internal linking it becomes difficult to know where to link, with so many origins and routes to cover, you can’t include links to them all on the homepage, for example.
Antoine: I think it’s underrated. It can be easy to focus on optimizing content, publishing to earn backlinks, and so on, but internal links we can directly control. Obviously you have to work with your development team, you have to prioritize projects and wait for resources to be available, but basically you can do whatever you want, and with quick results.
In the travel industry, demand is super fragmented, and the search volume is all around the routes. So your homepage, say, gets just 1% of your non-branded traffic, with the rest landing on other pages. You can’t just say, “OK, let’s optimize these top X pages” assuming Pareto’s principle is true and 20% of pages get 80% of traffic – in travel, that’s not the case, demand is spread out, so the pages you link to from your homepage is just the tip of the iceberg. You need to have a well designed architecture that ensures all your valuable pages are linked to.
Antoine: For commercial queries, the travel user experience is pretty straightforward. List the flight/hotel/etc options matching the user’s query. But, while you can have the best landing page for a given route, a given hotel, etc, you can’t assume the travel dates, number of people traveling, so you also want to get the user’s input as quickly as possible.
What we want the user to do when he or she lands on a page is perform a search with those details.
For informational queries, it gets more complicated, because demand for topics varies by destination, route and other factors. So you need to build a demand matrix, if you will, to help you identify the query types and groupings that have the most demand, so you can decide what content to build.
We focus on French users, so that helps us to narrow down what the most popular destinations and topics are for French travelers, and there are fewer origin airports than, say, in the US. It is more complicated for rental cars and hotels, because you can rent a car or book a hotel in virtually any city, so you may need to add more complexity to the architecture, maybe by adding regional level pages, to include internal links to more cities – and then you need to find ways to optimize internal linking to those pages.
Antoine: One example that might demonstrate a challenge is where you have search demand for an island vs. a city, and your website may only link airports to cities – so building a landing page for island destinations requires an exception in your architecture, new logic, in order to create pages for those queries. Building a case for that requires research, to identify where search activity demands these exceptions, and then working with developers to build it.
Antoine: It’s been pretty massive. We knew trends would be impacted by late February, not because COVID was already bad in France yet, but we knew it was coming. The drop-off for us started around the 16th of March, I think, and it was a huge decrease day after day after, until we hit a minimum.
Things got a bit better when the French government decided, OK, you can go out again, you can go to 100 kilometers maximum, so we saw the traffic recovering a bit.
You can’t do much when demand collapses. You can work on your content, because you know that at some point the demand will return, but you can’t do anything to get demand back. It’s important to maintain rankings for when demand does return, as you might for Black Friday, so we have of course kept working to improve, but the impressions and clicks obviously decreased a lot.
On the other hand, it’s a good time to do work you might otherwise put off because of the risks. For example, if you were waiting to do a site migration, you could just do it now, because it doesn’t matter so much if a key landing page gets screwed up for a few days.
Antoine: I get the reasoning for cutting SEO Budget – when things get bad, most companies can run for some time without the marketing department, SEO included.
I have friends in SEO on the agency side, here in Spain, who work for travel websites, and it has been tough for many of them. Tourism is a huge part of GDP here. But some have also seen an increase in new inquiries, as businesses are cutting back spending in paid search and are looking into what can be done on the SEO side.
SEO takes time, so if you’re starting now it’s probably too late to rely on SEO to make it through COVID, but if you’ve invested in the past and have good organic visibility, it’s paying off now.
Antoine: Yeah, it is. At the same time I think maybe there is some danger. As some companies are doing better and have more to invest in SEO, they’re not anticipating the drop in demand that will come when things are back to normal.
So if you’re getting more traffic from people who are stuck at home and shopping online for things, you should expect when that is no longer the case, traffic will come back down. You should be ready for that, and for marketing budgets to drop for business that have done particularly well during COVID.
Antoine: Content hubs. In travel, for informational queries, you have a lot of ways of building your architecture. Blog style content can be approached from the flights, routes and hotels side, discussing the options and tips for travellers, but you can completely flip that around and focus on destinations as well, linking in to flights, routes, etc. from those pages.
It’s not strictly internal linking, but as you map out your architecture, it’s important to consider how you will fit this content into the user experience on your site. It is often overlooked because it’s easier to convince senior management to invest in pages that target the commercial queries, like flight routes.
It’s harder to justify the value of informational content, to prove that at some point this will build value for the business, that it’s part of the customer journey, that it helps with branding. There is also a cost to translating and localizing that content for users in different locations who speak different languages. It can be tough.
Some websites seem to do fine without this content, for instance Kayak.com doesn’t invest much in content marketing and it works for them, but I wouldn’t recommend a strategy that completely leaves out informational queries.
Antoine: I’m not aware of Kayak’s entire strategy, but I have analyzed their site as a competitor. It’s clear that they’ve done a very good job on commercial pages. The architecture for those is very well optimized.
I think maybe Kayak shouldn’t overlook the informational queries, because while they may be leading right now in some countries, what happens in a few years if they are not as well positioned? Without the informational side, when the commercial query visibility falls, the business is more exposed.
Antoine: Branding is really everything. Look at Amazon – they recently massively cut their affiliate commissions, which they can do because they’ve built up the brand loyalty so well over the years.
As you said, with Google trying to take your traffic (or better answer user queries, depending on your point of view), it’s important to have a safety net – if Google takes over your commercial queries, you still have a brand, you still have customers, you still can rank for informational queries. The less reliant you are on Google and the commercial queries that Google may want to answer with their own features, the less risk to your business.
Antoine: Thank you for having me.
We enjoyed speaking with Antoine about the unique challenges facing a travel meta search engine. To recap some of our takeaways: