“Years ago I presented results to a company called Business Objects, with their entire team, and agency gathered to watch. It was the early days of text analysis, and I was so excited because for the first time we’d searched a massive online database for mentions of the client’s name. They were pleasantly surprised by the volume of their coverage—until they asked to see a sample story. Uh-oh. It turned out that a great many journalists had written “small business objects to this proposed legislation” or “big business objects to this ruling,” and they were all included in the coverage.”
Typical search technology can’t solve that problem. Most search engines used by media monitoring services have the same problem. Even a Google News search would produce the same extraneous clips.
However, some media monitoring services such as Glean.info (formerly CyberAlert) include advanced search technology that solves the problem and avoids delivering the irrelevant stories.
Advanced Search Capabilities
What’s Glean.info’s “secret?” Glean.info’s search engine technology can specify capitalization. In the case cited by Katie Paine, Glean.info’s search query would specify a capital “B” on the word business and a capital “O” on the word “objects.” It would ignore any use of the phrase “Business Objects” that did not include initial caps on the two words.
The ability to specify capitalization helps improve clip accuracy for many companies – especially companies with generic words in their name.
Orange, the mobile telecommunications company in Europe, is one example of a company with a difficult name to eliminate irrelevant clips. A Boolean query can focus the search to the company and its categories of services. It can also use the “and not” command to eliminate other irrelevant articles:
Orange AND (mobile OR cell OR telecommunications) AND NOT (“Orange Bowl” OR “Orange Julius” OR crayon) /\ Orange (specifying the capital “O” on “Orange”)
Requiring capitalization of the O on Orange greatly improves accuracy. Using the “and not” terms eliminates obviously irrelevant articles. By combining a Boolean query with forced capitalization, CyberAlert has consistently delivered over 97% accuracy on a test query for “Orange.”
Specifying Capitalization in Searching for Acronyms
The capitalization requirement can also be applied effectively to searches for acronyms.
The media often uses acronyms to refer to organizations. Use of acronyms is even more prevalent in social media. As shown by Abbreviations.com, virtually no acronym refers to only one organization or industry. ISIS refers to the Islamic State in Iraq and Syria. It also refers to Israeli Secret Intelligence Service, Independent Schools Information System, Iowa Student Information Services, International Spine Intervention Society, the name of a pharmaceutical company, and many other organizations and processes.
Another example: ACA stands for American Chiropractic Association, American Camp Association, Adult Children of Alcoholics, Association of Canadian Advertisers, Air Combat Command [the military seems to have an acronym for most everything], American Counseling Association, Atlantic City Alliance, and many, many more.
Key point: Searching for acronyms requires a Boolean query. The query includes the acronym plus a set of terms that describe the organization of its industry.
ACA as three letters in succession (really what the search engine looks for) also appears in many words including macadamia, academy, academic, acai and Acadia. Some search engines will deliver those words in a search for “ACA” if the search does not specify capitalization or spaces before/after the three letters.
Unique corporate and brand names are easy to monitor – and produce very few irrelevant clips. It’s much more difficult to accurately monitor media for names of organizations that contain common terms or if the names are not unique, or include an acronym. To achieve any level of clip accuracy for difficult-to-search organization names, it’s necessary to use a Boolean query that includes the name or acronym plus appropriate industry terms. Forcing capitalization on initial letters of names or all letters of the acronym vastly improves the accuracy of search results.
Here’s an example:
“Atlantic City Alliance” OR doatlanticcity OR “do atlantic city” OR ACA AND (tourism OR travel OR day-trip OR hotel OR casino OR boardwalk OR marina OR beach OR “business meeting” OR convention OR “Miss America” OR “light and sound show” OR atlantic OR “Do AC”) AND NOT (chiropractic OR “camp association” OR “Canadian Advertisers”) within 40 characters and with ACA capitalized.
As clip results are delivered, the query can be tweaked with search terms added, deleted or modified to improve accuracy.
Bottom Line: Most any company or organization can achieve greater than 95% clip accuracy by using well-structured Boolean queries combined with use of advanced search technologies such as specification of capitalization. Organizations dissatisfied with their clip results should work closely with their monitoring service to develop more refined keywords and Boolean queries – or test other monitoring services that offer more advanced search technologies.
William J. Comcowich founded and served as CEO of CyberAlert LLC, the predecessor of Glean.info. He is currently serving as Interim CEO and member of the Board of Directors. Glean.info provides customized media monitoring, media measurement and analytics solutions across all types of traditional and social media.