Irrelevant results

We sometimes face a problem that users do not get what they really want. In this section, we'll try to figure out what could be the reason.

Keyword phrases

When we specify a keyword which consist of two or more words (keyword phrase), we usually want to get the exact match in the same strict order.
For example if we add smm tool:

But in the timeline we see results such as "I've started doing SMM. Suggest any tool, please." As you might've noticed, both words of our keyword are in the same post, but they are located in different sentences and are separated by other words.

That is why when we use keyword phrases - keywords which consist of two or more words - we should add quotes (" ") - a search operator used when you are searching for a specific word combination or an exact phrase (learn more in the Keyword formats article):

Note, that quotation marks allow punctuation marks between words of the keyword (e.g., SMM, tool).

So in case you need 100% exact match without any characters between words, you should also add the plus (+) operator:

+” ” - Exact data search takes into account special characters and punctuation of your keyword (this is normally ignored) but ignores lower and upper cases.

General keywords

Another frequent problem concerns the results for general keywords. Often people receive lots of mentions which are not relevant to their keyword. For example, your brand name is "Daddy & Sons", and your company specializes on producing wooden furniture. But, unfortunatly, the brand name is not unique and there is also a company called "Daddy & Sons", but they are own a pizza restaurant. As a result we receive lots of irrelevant mentions, which we should sort out manually. What can be done in this case?

First of all, I would suggest to use negative keywords, so that the results containing any of the specified negative keywords next to your main keywords are not collected. Obviously, we should use "pizza", "restaurant" as negative keywords in the example mentioned above:

Unfortunately, we still can get a lot of "trash", e.g., posts from social accounts that promote that company instead of yours. Then you can add those accounts to the Blacklist.

To be honest, for the first time you may not create a perfect search query and we always recommend to follow these simple steps:

1. Create the most specific request you can.

2. Analyse the first resutls and blacklist unnecessary sources.

3. Return to the alert settings and modify the search to exclude irrelevant results.

Bonus, learn how to create requests in Boolean Search and get the most out of this tool! ;)