Knowledge Research

This simple and easy to use query screen which is part of the Business Intelligence module is a powerful feature which pulls together much of the ‘under the hood’ analyses that GAT can perform.

It helps you answer two important questions

Who knows whom?

Who knows what?

Knowledge research lets you find who in your domain is the ‘expert’ on a particular topic, the person with the strongest connections to an individual or the person with the most contacts with people in another domain.

For example, if you want to know who in your domain has the strongest contacts to Google, you can search for the domain name ‘’, this will identify all the people who have email, document sharing, calendar appointment or even just Google addresses in their Contacts. The results will rank these people in order of strength of the connection.

Having identified the strongest connection you can decide if they are the right contact to approach for introductions.

GAT offers options at the search stage which can improve search speed. You can search by group or Organizational Unit. This will limit the search to just those members of the domain covered by the selection.

In addition, if you are not searching for a contact you can leave the ‘Search in Contacts’ choice unselected. The default is not to search in contacts.

By default, we also have ‘Narrow search’ selected. A narrow search only returns results for ‘primary’ sources. In the case of documents, it will search only for writers or editors who are explicitly named in the share. In the case of emails, it will only search for senders or receivers. If you are cc’ed  or bcc’ed on an email you will not be returned when this choice is selected.

This setting is on by default.

In the case of ‘subject’ search, all sources that mention the subject are searched and the results are presented in table form, with the each domain member ‘ranked’ based on their use of the subject word.

Sometimes knowing the type of document you are dealing with can help increase the relevance score, so clicking on the ‘?’ for the documents found will give you a breakdown of the document types.

The search strings can be joined by Boolean operators to build more complex searches.