In Rival IQ, you'll find two powerful listening tools for analyzing conversations on the social web—Instant Search and Saved Searches. These tools are powered by the Quid platform and bring you insights from millions of posts and documents from Twitter, Reddit, blogs, and more.

In this article, we'll cover how to use these features along with best practices for building searches.

You'll find the Instant Search and Saved Searches dashboards under Social Listening > Social Web on the left navigation bar.

The Instant Search and Saved Searches dashboards analyze a collection of posts and documents from the social web using the Quid platform.

The data sources you can analyze in these dashboards include posts published in the last 30 days from the following sources:

The data sources you can analyze in our social web listening tools do not include the following sources:

If you're interested in listening capabilities for any of these sources through full access to the Quid platform, please contact our team.

The Instant Search dashboard is the easiest way to get started analyzing conversations from the social web.

An easy way to start your analysis of a brand or an organization in your landscape is by clicking one of the recommended brand searches at the bottom of the dashboard. The recommendations are based on the current landscape.

For example, clicking "Starbucks" in the example above uses Quid AI to generate keywords for conversations and mentions of Starbucks across the social web.

Rival IQ then runs the analysis using the brand query and returns results, including key metrics, popular terms, sentiment drivers, and more.

To analyze a brand or topic that isn't in your landscape, start your search by specifying primary terms—keywords or phrases related to your subject of interest. The Instant Search tool returns posts containing any of the terms or phrases you include.

The example search below matches any document that mentions Starbucks, @starbucks, or #starbucks.

Click Search to start the analysis.

Rival IQ runs the query and returns results, including key metrics, popular terms, sentiment drivers, and more.

After you run a search, you can edit the query to refine the keywords, sources, domains, and authors being searched. Click Edit at the upper left to access the query editor.

The following sections provide more information on the Query Editor.

To help you quickly refine your analysis, you can use 3 filters to narrow your analysis by source, sentiment, or language.

Saving a search allows you to:

There are 2 ways to create a saved search in Rival IQ.

One way to get started on creating a saved search is to start from the Saved Searches Dashboard.

Click Create New Search to begin.

From there, choose the keywords, a language, and sources. To save your search and view the results, click Save Search.

You can save a search in the query editor by clicking Save Search.

You can also edit an existing search and use the Duplicate Search button in the query editor. Using this path does not affect the existing saved search.

To edit the definition of an existing search, click Edit Search at the right of a row in the Saved Searches Dashboard.

Alternatively, you can edit the search query from the fast switcher in the dropdown menu located in the title bar.

To edit the name of an existing search, click Edit Search Name on the right of a row in the Saved Searches Dashboard.

The social web query editor is an easy-to-use way to define your listening searches in Rival IQ.

There are three primary sections in the query editor that help you define your analysis.

In this section, you'll define the words and phrases that you're looking for (and looking to avoid) in your query. This section contains the following fields.

The first field enables you to find any document in our search index that matches any of the words, phrases, hashtags, or social handles you specify.

The example below matches any document/post that contains the word Nordstrom, the handle @Nordstrom, or #nordstrom.

Primary terms also serve as the objects of sentiment analysis. For example, consider the following sentence:

"I love shopping at Nordstrom, but I hate when my children scream the whole time."

Rival IQ classifies this sentence as positive because of the sentiment expressed towards Nordstrom and ignores the negative sentiment in the second phrase.

The second field in this section allows you to refine and narrow the query to include only documents that also contain one of the specified words or phrases.

The example below uses two phrases to limit the search to mentions of Nordstrom that also include mentions of San Francisco (or San Fran for short).

Adding terms to this INCLUDE field always narrows the results set because it limits the results found in the initial FIND step of the search.

This third field helps you further refine and narrow your query by excluding posts that match any of the words or phrases you specify.

The example below narrows the query to omit documents/posts that mention either Gavin or @GavinNewsom.

The Language field allows you to filter content authored in a specific language or "All" languages to find all mentions regardless of language.

This panel enables you to select any combination of available sources to include in your search.

The Reduce Noise panel includes additional options to improve the relevance of your search:

Rival IQ AI-generated queries include additional terms when the brand you're analyzing might be ambiguous (its name is the same as a common word, such as Tide, Apple, or Dove). These additional include terms improve the precision of your analysis. Note that queries are starting points that you can edit and refine further to refine your search.

For primary terms that do not need to be disambiguated from common words, like social handles or branded hashtags, Rival IQ includes the primary terms in the INCLUDE posts field to ensure we include all relevant results.

The Quid platform assigns sentiment to mentions of the primary terms using advanced natural-language processing in multiple languages. In Rival IQ, it isn't possible to change the assigned sentiment of phrases.

A net sentiment score of +100 means that all opinions are positive; a score of -100 means that all opinions are negative. To analyze your brand's net sentiment in more detail, compare its net sentiment score to brands in the same category. Based on an analysis of more than 200 brands, we found that most brands have a net sentiment score of 50. Brands with the highest degree of negative sentiment have scores of 0 to slightly negative, such as -10. A score below 50 means that the net sentiment for the search is lower than most brands.

The net sentiment is a ratio of positive to negative opinions about your primary terms:

​( ( Positives - Negatives) / ( Positives + Negatives ) ) * 100

A single post may have multiple opinions. For example, the fictional post below has two positive opinions and one negative about ice cream:

"I love ice cream. Chocolate is my favorite, but all ice cream tastes delicious. Unfortunately, I also hate ice cream because I have a dairy intolerance."

This fictional post has a Net Sentiment score of 33%.

​( ( 2 - 1) / ( 1 + 2 ) ) * 100 = 33.33

The results Rival IQ returns for Twitter are sampled and estimated due to the large volume of posts. As a result, sample Tweets for an author, phrases, emotion, or other item may be fewer than the estimated post volume.

For tips on how to use keyword fields and filters to dial in your social listening analysis, go to Best Practices for Building Searches in Social Listening.