As my friends know, I have been fascinated by Google’s Custom Search Engines (CSEs) for years. I have met several colleagues who have become as addicted to CSEs as I am; I feel as if we belong to a tribe. 🙂
I remain disappointed by the apparent CSEs’ low penetration into our industry tools. Part of the undeserved unpopularity is due to the lack of documentation from Google – or anybody else. (Google help no longer keeps the documentation of its advanced operators either.) The lack of info makes the operators like a medieval trade secret that is known to few (and, in our times, communicated via Messenger!)
David Galley and I contribute to covering the CSE documentation gap by blogging, hosting webinars, and preparing a book on the subject.
I have never seen a summary of unique CSEs’ advantages, so I came up with one. Many of the features have been there from the start (in 2006), but some unique semantic features are newer additions and deserve your attention. I have tried to make a full CSE feature list; let me know if you think of something important to add.
- (Invisibly for the user) include only given site(s) (e.g., linkedin.com/in, which will find only LinkedIn profiles)
- Exclude given site(s)
- Give priority to the given site(s) but search the entire web
- Give priority to pages with given keyword(s)
- Narrow to a language
- Narrow to a country
- Boost results by country
- Include multiple sites via patterns using an Asterisk (e.g., site:behance.net/*/resume)
- Automatically append a string to user’s search (for example – narrow the search to PDF documents by adding filetype:PDF)
- Define synonyms to process user’s input
- Use the Synonyms feature to run long OR statements (for example, search for common women’s names)
- Select pages with given Schema.org object(s) (like Person, Physician, or Organization)
- Search for the presence of Schema.org objects’ fields (for example, find pages that have the filed “email” in the Person object)
- Search within Schema.org objects’ fields (for example, search for Github profiles containing “love Python” in the bio or LinkedIn profiles containing “open to new opportunities” in the headline)
- Guide the search by selecting Knowledge Graph object(s) (for example, find pages that are “CVs”)
While the first 11 points have been there from the start, 12-15 are later additions and make CSE search truly semantic, quite a challenge for a global search engine, nicely solved. (The only things that are less nice about CSEs is the old-style UI and cryptic operators to write out).
Please join us for an online class on CSEs on Tuesday, May 19th – “Become A Custom Search Engines Expert”. The optional workshop is the next day, Thursday. Seating is limited, sign up now!