Recent ECJ decision confirms: be careful with online scrapers and crawlers, they are often illegal

Database protection?

Database protection is a rather peculiar part of intellectual property rights. Copyright, trademark law, design law and other intellectual property rights protect a specific intellectual creation (a book, an image, a model, a brand name). Database protection, to the contrary, mainly protects the investment (in thinking, time or -especially- money) that someone (usually a company) has put into building a structured database around the content that is usually protected by other IP laws. It is in other words not the content of a database that is protected, but the database, i.e. the structure, in itself, at least insofar as it can be proven that considerable investments have been made to create the database.

The Belgian Database Act of 1998, as so often the transposition of a European directive, prohibits third parties from simply copying an entire database and at the same time prohibits systematic and repeated (smaller) retrieval requests from databases “if they conflict with the normal exploitation of those databases or they cause unjustified damage to the legitimate interests of the producer of the database.”

Databases and the Internet

As mentioned, the Database Act dates from 1998, a time when the Internet was still in its infancy and a database was often an Access or Excel file that ran somewhere on a company’s own server, internally.

However, the world has changed quite a bit in the past 25 years and today a lot of “databases” are actually publicly accessible websites. Every booking website of an airline, a tour operator or a concert hall is potentially a protected database, just like, for example, job websites or websites with real estate advertisements.

It is exactly in these cases that a kind of “secondary market” has developed on the internet, in which websites search the internet to collect data from various websites of, for example, airlines in order to collect and bundle all prices of all flights on one platform, where that platform itself tries to make a financial profit from the collective offering of all that collected content.

Not surprisingly, this evolution has led to a lot of legal squabbles, with companies that have invested heavily in setting up their database (and their website) being anything but happy with the fact that others just scrape their content and use it themselves.

Ryanair, for example, fought a war with PR Aviation in 2015 before the European Court of Justice to see the protection of its flight data as a database confirmed and to ban PR Aviation from systematically taking over Ryanair flights on its website. In that specific case, Ryanair bit the dust, because according to the Court its website did not show the necessary investments to obtain database protection.

Search engines that search the content of other websites permanently and in real time, in their turn, also potentially violate the database rights of those websites (that is, if those websites can prove that they have made substantial investments to ensure that their database is protected). This is also the case if only small parts of the target website are retrieved with each search. This was already decided by the European Court of Justice in 2013 in a dispute over the online search engine Gaspedaal.nl.

Recent judgment clarifies the rules

Online company Melons thought it had found a way out to still be able to scrape other websites. It didn’t search those websites in real time, like Gaspedaal.nl in the example above, but it periodically made a local copy and then analysed that local copy systematically. CV Online, a Latvian online job database, was all but happy with this way of working. It was their vacancies database that was used by Melons via a search request from users on the Melons website. Melons made a copy of (meta) data from the CV Online site and searched that local copy every time a search was made by users on its website.

It should come as no surprise that the European Court of Justice ruled on 3 June 2021 that even if a search engine does not search other websites in real time but does make a local copy to search them, it still (potentially) violates database law (again IF there is of investments that ensure a protected database, of course). The Court confirms that the terms “reclaim” and “reuse” from the Directive must be interpreted in the broadest sense and that the aim of the Directive (and therefore also of Belgian law, of course) is to prevent someone else taking the income on the back of those who made the investment to establish the database.

… But also sets new limits

The judgment does not only bring good news for owners of databases. The Court adds another bounce to its decision that potentially opens a lot of doors for third parties who still want to sift through someone else’s database. The Court says that a balance must be sought between database rights and free access to information and the development of innovative products and that for that reason it must always be examined whether there is any damage on the part of the database holder. Database holders will therefore have to prove their damage. That is likely to be very difficult in many cases. After all, aggregator websites very often refer exactly to the source website where they have copied data and in many cases the question will be whether more visitors (and therefore potential customers) are not led to the source website in this way. In that case, there can hardly be any damage…

Be careful!

So food for thought, but our permanent warning still stands: be extremely careful before systematically collecting data online. You are potentially committing violations of database law, but there is also a good chance that you will encounter problems under copyright, unfair market practices and unfair competition, GDPR and quite a few other regulations. So seek advice before you set off…