En savoir plus sur l'Académie européenne des brevets et sur la manière dont elle promeut et soutient la formation et l'enseignement sur des thèmes liés à l'OEB :
Désolé. Actuellement, cette page n'existe pas en français.
We will now take you step by step through two prior art search processes: a product search and a patent search.
You must do both to be confident that you have done a thorough prior art search. You must also do them before spending significant amounts of time and money on your idea.
To maximise your chances of finding relevant information, spend some time thinking of key words or search terms which best describe your idea.
When using search engines, the most obvious key words may be unhelpful. For example, let us say your idea is a mousetrap. A search for ‘mousetrap' produces over two million hits - many of them irrelevant, and an impossible number to search.
But a search for ‘rodent trap' (what else it is) and ‘trapping mice' (what it does) produces 20,000 and 700 hits respectively. These are still not small numbers but they are likely to be more relevant, so we can usefully start searching here.
Another problem: the most productive search terms may be specialist technical terms that you do not know. For example, a search for external devices that pump blood round the human body required the crucial medical term ‘extra corporeal'. A searcher with no medical knowledge would be unlikely to know this term, but might find it while examining the results of other key word searches. It may therefore take a few preliminary searches to find better keywords to use for more accurate searches.
Look out too for new terms for new technologies: for example, ‘virtual fit' for software systems to replace trying on clothes in shops, and ‘telemedicine' for remote monitoring of patients in their own homes.
You need to find out what is already on the market:
Obsolete technologies or products may be prior art, so check historical as well as current sources of information.
Products in development but not yet on the market may be prior art, so search news sites, industry journals, trade show and exhibition websites. Perhaps especially search academic research activity, as this is where many new products start out, often years before a commercial product appears.
You should also of course search offline - in shops, books, periodicals, printed catalogues etc.
And talk to people with relevant experience - for example, retailers and suppliers - who will have seen products come and go over the years and may have seen your idea among them. (People who have retired from relevant careers can be valuable sources of information, as their experience may go back much further than current practitioners!)
For many ideas, patent searching will be far more important than product searching. Although many products on the market do not have a patent, they are probably heavily outnumbered by the many ideas that are successfully patented but never reach the market.
Patent searching involves two skills:
We are going to show you how to use the European Patent Office's free Espacenet database, which is easy for beginners to use.
However, you are unlikely to be as good as a professional searcher, so in some cases it may be advisable to ask a professional to search for you. (See Professional patent searching .) An exception is if there is so much prior art that your search ends quickly!
It could range from a few minutes, if your first keywords are accurate and there is a great deal of prior art, to many hours.
The best advice is that you must be prepared to spend all the time it takes to be confident that you have done a proper job. Your mission is to find evidence that disproves the novelty of your invention. Your hope is that you will fail, but in the interests of a thorough search you must put that to the back of your mind.
For the same reason, assume that if you are not finding prior art, you are looking in the wrong places.
Keep searching until you are confident that there is nowhere else left to look. And keep records of everywhere you look and everything relevant that you find.A thorough and well-recorded search is essential - because how else do you prove an absence of prior art?