Seven points to consider before starting your search

Here are some basic things you need to know before you start searching.

1. The concept approach

Usually, a prior art search boils down to a combination of concepts defining the subject you are looking for.

A patent prior art search consists in finding relevant documents in a collection of patent documents. The fields searched are usually abstracts or classification symbols.

In a patent record, the content of the invention is described in the abstract and the classification symbols attributed to the patent have been chosen among the thousands available in the classification scheme and are usually indexes best characterising the technical content of the patent document.

In the abstract, you will find words, adjectives and verbs that linked together form sentences expressing your searched subject. If the expression of the subject remains the same in all relevant documents, words, verbs and adjectives vary from one relevant document to another.

The expression of the technical subject is a combination of concepts that will remain common to all relevant documents.

Knowing that the documents you are looking for have these common features ? these so called concepts ? your search will consist in finding documents where a given combination of concepts is present.

Since patent documents are classified, searching for a given concept should not be limited to a keyword search. A concept can perfectly be represented by a classification symbol.

A sound search strategy will therefore involve combination of keywords and classification symbols. An individual concept is usually defined by an "or" combination of keywords and classification symbols all relating to this concept.

The experience shows that it is not easy from the start to define all possible synonyms and classifications covering one concept. This usually remains an imperfect process. It can be tuned during the search itself: relevant documents reveal new ways to express your concepts. Using these new concepts more documents can be retrieved and so on. This becomes an iterative process that an experienced searcher brings to an end when he has the feeling that additional efforts will not bring more results.

2. The more, the easiest

The easiness of a prior art search is proportional to the number of documents relevant to your subject that are contained in the searched collection.

If there are hundreds of documents relating to your searched subject in the databases the chances were very high that even using a clumsy combination of keywords you would find a relevant document. It is quite simple to understand why. Hundreds of relevant documents mean that there are almost hundred different abstracts in the database that describe more or less the same invention. Chances are therefore very high that all possible keywords one would think of are used in some of these abstracts. In other words, if you ask 100 people to describe this invention in a few sentences, you are sure to be able to retrieve at least one of these descriptions using your own combination of keywords.

Consequently, one can say that a search is difficult when the number of documents in the database corresponding to your subject is low. This is usually the case in real life: none or very few relevant documents relating to your subject exist. In such a case the iterative concept approach exposed previously is extremely powerful and should yield these few documents.

3. Think lateral

To get out of a search dead end try to think lateral to find new concepts defining your subject.

To retrieve additional concepts possibly covering the invention, you sometimes have to think lateral: exploring multiple possibilities and approaches instead of pursuing a single approach.

4. The limits of the concept approach

When a subject cannot be expressed in a "normal" sentence, the concept approach is not applicable or can only be applicable in presence of good indexing schemes.

It would be wrong to present the concept approach as "the" universal search strategy.

5. I found nothing

Absence of evidence is no evidence of absence

Even experienced searchers cope with cases where no relevant results have been obtained. This does not mean that no relevant document exists or can be found. The above maxim is well known among archaeologists: if you do not find remains of an old temple, it does not mean that the temple did not exist or cannot be found elsewhere.

In patent searches such phenomena happen quite often and for various reasons:

  • A searcher may not have used the winning combination; a common case among novice searchers.
  • Some fields you are using like the classification are absent from certain patent records. This can be the case in databases covering collections that have not be classified for instance.

6. Use the Web to its full extent

Specific patent searches are sometimes best conducted using both the information available on the Web and in patent databases.

One often overlooks combined information that can be obtained from the World Wide Web and from patent sources. Patents or inventors are often mentioned in texts found on the Internet giving precious indications to be able to continue the search in patent databases.

7. Use the right tools

Using free patent search tools is not necessarily more cost effective than using fee based databases.

This last principle belongs to common sense. But many discussions with patent professionals often reveal that free tools are sometimes too blindly trusted. This is due to the incorrect impression that free services offer as much information if not more than commercial databases. Users need to be clearly informed on the limitations of free products. For instance, abstracts in espacenet, are not always available when they are present in some well-known commercial databases. In addition, even if you push free systems to their limits being aware of their strengths and of their weaknesses you will often spend much more time finding results than by using fee based tools. The costs of a patent professional spending 2 hours on a free system are often higher than the costs - working and database costs added together - of such a professional spending 15 minutes on a fee based system.