Getting Started

Glypse is a social intelligence platform that analyzes public conversation and turns it into structured data: sentiment scores, emotion breakdowns, topic classifications, volume trends, and entity discovery for any query. The dashboard is where this analysis comes together. You enter a query, choose a time range, and the platform presents an analytical breakdown of the surrounding discourse: what people are expressing, which emotions and topics dominate, and how the conversation is shifting over time.

The dashboard

When you sign in, the dashboard is the first screen you see. Three areas make up the layout:

• The query bar at the top is where you enter what you want to analyze and set the time range.
• The results area in the center is where charts, stats, and analytical breakdowns appear after you run a query.
• The feed panel on the right side (or below the results on smaller screens) shows the individual posts that matched your query.

The sidebar on the left provides navigation between the different analytical views. Each view focuses on a specific dimension of the conversation, and all of them apply to the query that is currently active.

Queries

Running a query

Time range. The time range button in the top-left of the query bar opens a popover with preset options such as 3 days, 7 days, and 14 days. The default is 7 days. The time range you choose affects more than just which posts are included. It also determines the window over which averages, trends, and scores are computed. A 3-day range gives you a snapshot of the most recent conversation. A longer range smooths out day-to-day noise and shows broader patterns. If sentiment dropped sharply yesterday, a short window will surface that clearly, while a wider window might absorb it into a stable average.

Query. The text input in the query bar is where you type what you want to analyze. This can be any publicly discussed subject: a company, a person, a product, a hashtag, a political issue, a cultural moment, or an abstract concept. The query matches keywords across social media posts. If you type electric vehicles, the platform returns posts containing both words. You can refine queries with operators covered in the next section, but a plain keyword or name is sufficient to begin.

Analyze. Pressing Enter or clicking the Analyze button runs the query. The results area populates with a stats bar, charts, a word cloud, and a feed of matching posts. If the query returns no results, you can try using fewer or more general terms, or extending the time range.

Query syntax

The query bar supports several operators that control which posts are matched. You can use them individually or combine them in a single query.

Matching multiple terms. When you enter more than one word, the query requires all of them to appear in the same post. AI energy matches posts that contain both "AI" and "energy," in any order, anywhere in the text. This is the default behavior.

Either/or queries. The pipe character lets you match posts containing any of the listed terms. vegetarian | vegan returns posts that mention vegetarian, vegan, or both. This is useful when a subject goes by multiple names, or when you want to capture related terms that are used interchangeably.

Exact phrases. Wrapping terms in double quotes matches them as a contiguous phrase. "White House" only returns posts where those two words appear next to each other in that order. Without quotes, the query would match any post containing both "white" and "house" independently, regardless of how far apart they are or what they refer to.

Excluding terms. A minus sign directly before a word removes posts containing that term from the results. crypto -Bitcoin matches posts about crypto that do not mention Bitcoin. Exclusions are useful when a dominant subtopic overshadows the rest of the conversation and you want to isolate what remains.

Combining operators. These operators work together in a single query. "machine learning" | AI -regulation matches posts that mention either the phrase "machine learning" or the term "AI," while filtering out any post that contains "regulation."

Multiple queries

The Add query button below the query bar creates another input row. Each query gets a distinct color and shape icon that carries through to every chart, stats row, and legend on the page.

You can optionally give each query a display name to make charts easier to read. The overflow menu (...) on each query row includes an Edit display name option. Display names appear in chart legends and tables instead of the raw query text, which is helpful when queries are long or technical.

When two or more queries are active, a view toggle appears at the top of the results with three options:

• Compare: all queries appear together in the same charts. Bar charts show one row per query. Time series overlay multiple lines or stacked segments. This is the default view and the most direct way to see how queries differ across every dimension.
• Aggregate: treats all queries as a single combined set, merging their results into one unified analysis. This is useful when you want the overall picture across several related terms rather than the differences between them.
• Select: focuses on one query at a time. A dropdown lets you pick which one, and the page renders the full single-query layout for just that query.

The feed panel also responds to the active view. In Compare and Aggregate modes, posts from all queries appear together. In Select mode, only posts matching the selected query are shown.

History and bookmarks

The dashboard automatically saves your recent queries. The history icon in the query bar (or Ctrl+K / Cmd+K) opens the history dialog, which lists your recent queries and any bookmarked entries at the top. A filter field at the top of the dialog lets you narrow the list by query text or display name. Selecting an entry restores those exact parameters into the query bar, including all query strings, time range, and filters.

The bookmark icon in the query bar saves the current query for easy access. Bookmarked queries are pinned to the top of the history dialog, separated from the recents list. This is useful for queries you return to regularly or want to keep as reference points.

Results

Stats bar and charts

The results area is organized into two panels that work together at different levels of detail: a stats bar for headline metrics, and a charts section for the distributional breakdown behind them.

The stats bar at the top provides headline metrics: total volume, overall sentiment, the dominant emotion, and the leading topic. This is the summary layer. It tells you the shape of the conversation in four numbers before you examine the charts in detail.

The charts below the stats bar show the distributional detail behind those headlines. Where the stats bar reports a single sentiment score, for example, the charts reveal the full breakdown of positive, neutral, and negative posts, how emotions are distributed across categories, and how the conversation changes over time. Each dashboard tab presents its own set of charts tailored to that dimension. The Overview tab consolidates the most important charts from every dimension into one view, which is why it appears by default.

Dashboard tabs

Each tab in the sidebar gives you a dedicated view into one analytical dimension. The active query and time range persist across tabs, so you can move between them freely.

The tabs are organized into two groups:

Measure tabs provide quantitative analysis:

Explore tabs help you discover what is inside the conversation as a word cloud, treemap, or ranked table:

In the Explore tabs, every item in the word cloud, treemap, and leaderboard table is interactive. You can click any term to add it to your query or open an action menu to refine your analysis with it.

The feed

The right side of the dashboard shows a scrollable list of posts matching your query. Each post displays a timestamp, the post text with hashtags and mentions styled as distinct elements, and a set of colored badges.

The badges correspond to the active dashboard tab. On the Sentiment tab, for example, posts show whether they were classified as positive, neutral, or negative. On the Emotions tab, the badges reflect which emotions were detected. On the Topics tab, they indicate which subject categories the post was classified under. This contextual labeling makes it easy to scan individual posts alongside the aggregate charts for that dimension.

You can click Show more at the bottom of the feed to load additional posts.

Detail view

Clicking any item in the feed opens a detail panel that slides in from the right. The detail view shows the full post with the author's display name and handle (linked to their profile on the source platform), any embedded images or link previews, engagement stats (replies, reposts, and likes), and a complete set of classification labels: sentiment, emotions, and topics all displayed together as colored badges. A link at the bottom opens the original post on its source platform.

Further reading

For a deeper look into the analytical techniques used in the dashboard, see the guides below.