The ultimate guide to Boolean Search for social listening
Today we are going to talk about one of the most useful features that could ever be created for social listening tools - Boolean search. Because everyone can type a keyword and wait for the result. But in this article I will show how to do it better ;)
Do you know this gentleman in a black suit? His name is George Boole, an English mathematician from the 19th century. He developed an algebraic method described in “The Mathematical Analysis of Logic" and “An Investigation of the Laws of Thought" which contains Boolean algebra. It set Boolean logic in motion and made it fundamental to all major programming languages. Boolean concepts also take place in search engines.
What is Boolean search?
Boolean search is an advanced and effective way to work with almost all search engines. It allows users to combine keywords with Boolean operators (AND, OR, AND NOT and etc.) and find exactly what you’re looking for. With Boolean operators you can configure more specific queries, create multiple combinations, modify existing requests and do many other crazy things that are not available in a simpler search mode. Moreover, you can even use it for lead generation!
Do all social listening tools come with a Boolean search mode?
Boolean search can be found in some social listening tools (though not too many of them). Usually this feature is available only for Enterprise users, but here at Awario we offer it to all our customers no matter what plan you have and even including the free trial!
Free 14-day trial, no credit card required.
Let’s start with theory and take a detailed look at each of the commonly used Boolean operators. Please keep in mind that I will talk about the syntax which is used in Awario. It can differ from other similar tools.
AND means you are searching for documents that contain BOTH keywords.
Cheeseburger AND McDonald’s
OR means you are searching for documents that contain any of your keywords.
Pepsi OR Coca-Cola
Near/n specifies the closeness of your keywords to each other. This means the program will show you only the documents where the keywords appear within n words away from each other. The word order is ignored by this operator.
Onear/n specifies the closeness of your keywords to each other, while taking the word order into account. For example we have the name William Johnson and need to exclude any mentions of Johnson William, but still allow all second names and nicknames:
AND NOT excludes documents that contain a particular word(s). In this case you’re telling the program to include all documents that have the X word in them AND don’t have the Y word in them.
Brackets are used to group terms together, so that operators like AND, OR and AND NOT can be applied to all the terms in the brackets.
Tesla AND (electric vehicle OR energy company OR automotive company)
Exact match search takes into account special characters and punctuation of your keyword (this is normally ignored) but ignores lower and upper cases. This operator should be used for keywords which consist of two or more words.
Case sensitive exact match search takes into account special characters, punctuation, and lower and upper cases of your keyword.
FROM specifies the source(s) you want your mentions to come from. If no source is specified, you'll get mentions from all sources. You should type the sources in the following way (in brackets the shortened variants are provided): twitter (tw); facebook (fb); google + (g+); instagram (insta), youtube, reddit, web, news-blogs. Note that this parameter should always be at the end of the query.
BMW FROM tw,fb
The Language operator specifies the language of your results. Ignore this parameter to get mentions irrespective of their language.
The Country operator specifies the location of your results. Note that some social media profiles don't have information about the location and some platforms like Reddit don't even show such info! So when you choose a specific location you might be missing relevant mentions if you don't have the UNKNOWN setting turned on.
Search for any document that contains a link to a website.This operator has 4 options:
link:"*.site.com/*" - search for any document that contains a link to site(dot)com, including all its subdomains, pages, etc.
link:"site.com/" - search for any document that contains a link to site(dot)com, excluding the subdomains.
link:"site.com/page" - search for any document that contains a link to the exact pages site(dot)com/page.
link:"site.com/?p=\"123\"" - if your link contains quotation marks, include a backslash before each quotation mark that belongs to the link, as in the example.
That’s been a list of some Boolean operators, but there are more operators which can be used for uncommon cases - check out the link.
Let me give you some examples of the real cases which are used by our customers or for our own needs.
1. Find linkless brand mentions for link building
With this query, we'll find exact mentions of either of Link Assistant or SEO PowerSuite. But according to our scenario we want to exclude any document that contains a link to linkassistant(dot)com, including all its subdomains, pages, etc. And we need mentions only in English and from the US. In the end let’s also specify necessary sources: web and news-blogs.
By the way, thus you can find unlinked mentions of your brand on the web so you can turn them into links.
2. Find mentions of your brand's specific product
And here comes a tricky one ;) I want to find documents which contain the words Tesla and Model 3, but exclude results with any (or all) of the negative keywords model x, model s or roadster. Pay attention that I grouped keywords and negative keywords with brackets to apply AND NOT to the entire group.
3. Generate new leads
Awario already uses a unique algorithm for lead generation. Leads provides impressive results, even though it’s still in beta. The advantage of the manual search type is that you can adjust a query the way you need even for such needs as lead generation.
Keep in mind that Awario searches for the exact match of the keywords or keyword phrases which you specify in the request. Therefore, it is not enough to input a list of services or goods you offer when you use Boolean search - the results will be too broad for lead generation purposes. You should use such keyword phrases as your prospects would.
In Example 3 we will consider how a web designer can use Boolean search to find potential clients. As you can see, I entered requests like looking for a web designer, need a web designer, recommend a web designer, suggest a web designer. But definitely some people may skip articles, use different word order, and so on. That is why I left only the meaningful words and used them with near/n. Mind that the more words are used in a key phrase, the wider should be the word range.
Of course there is no true fun without spam posts. That is why we should exclude posts which contain offers or ads. Before adding negative keywords, I looked through the first results and noted the most used phrases which corrupt my search. You might have noticed I specified tool OR tools. I always recommend to use plural and singular forms of your keywords not to miss any of them in the search.
Note that lead generation is not an easy task! From the very start you will likely need to modify search requests, add new keyword phrases to get more results, specify more negative terms to exclude irrelevant results, block spammers, and filter found mentions. But the final result is worth the effort!
Conclusion (sort of)
Boolean search is an advanced search mode with its pros and cons. Yeah, I suppose after reading all this stuff, you may think “Are you serious ?” or “I do not have time for this”. And maybe you are right, but those 30% of rock ‘n’ rollas who decided to deal with it “know da wae” to success. So why not join them? :D
That was not easy, but we made it! Leave your questions about Boolean search and I will reply to all of them ;)