Think about the main ideas in your research question. For example, in the question,"What are some best practices related to web analytics?" the main ideas are: best practices and web analytics.
These keywords can be used in your search statement:
"best practices" AND "web analytics"
If you do not find enough useful articles, expand your search by using OR to include additional, related keywords:
"best practices" AND ("web analytics" OR "google analytics")
Look through the subject terms, title, and abstract of useful articles you find for keywords related to those that you have used - then try connecting them using OR as shown above. This will often locate useful articles missed in your first search.
Many databases allow you to use "*" (an asterisk) to truncate a term. Truncating allows you to search for all variations of a word using a single search. For example, a search for ...
... will find articles containing both "metric" and "metrics" with one search. Broadening a search like this can help you find more useful articles.
NOT lets you exclude words from your search to remove unhelpful articles from your results:
"web analytics" AND conversions NOT (retail OR sales)
Subject terms are words assigned by a database's managers to all articles in it that have the same subject. In SingleSearch, for example, all articles on web analytics are assigned the subject term "WEB analytics." Because these terms are assigned to all articles on a specific topic, using a subject heading as a search term will retrieve only articles on that particular topic. Databases that use subject terms allow you to use their advanced search page to search only the subject terms field of the articles they contain.