# Post Analysis

## Overview

The **Post Analysis** page is your command center for deep-diving into raw post-level content across your project's dataset. Whether you're validating narratives, investigating specific authors, or identifying risk signals at the post level, this page gives you the granular visibility you need.

"Posts" refers to any ingested content in your project — social media posts, news articles, chat data, and more — all displayed in one unified view.

### What You Can Do

* **Explore** post-level data with a powerful, customizable data grid
* **Monitor** key metrics at a glance using the top-level stats bar
* **Filter** posts using the Segment Bar's boolean search and filter capabilities
* **Customize** your view by adding or removing columns to suit your analysis needs
* **Track** post and engagement activity over time with the Activity Timeline


## Navigating to the Post Analysis Page

1. Open your project in Constellation 3.0
2. From the left navigation menu, select **Posts**
3. The page will load with your project's post data, applying any active Segment Bar filters


## Page Layout

The Post Analysis page is organized into three main sections:

1. **Stats Bar** — High-level metrics summarizing the currently filtered dataset
2. **Activity Timeline** — Visual graphs showing post and engagement trends over time
3. **Data Grid** — A searchable, filterable, and customizable table of individual posts


## Stats Bar

At the top of the page, the Stats Bar provides instant insight into your currently filtered dataset. All metrics update dynamically as you apply or remove filters.

### Available Metrics

| Metric | Description |
|  --- | --- |
| **Post Count** | Total number of posts currently displayed based on active filters |
| **Engagement Count** | Aggregated engagement across displayed posts (excludes news sources) |
| **Top Cohorts** | Up to 3 cohorts ranked by unique author count within the filtered dataset |
| **Narratives** | Count of AI narratives represented by at least one post in the current view |


**Pro Tip:** Use the Stats Bar as a quick sanity check after applying filters — these numbers reflect exactly what's in your current view, not the full dataset.

## Activity Timeline

The Activity Timeline provides a visual representation of post activity and engagement over time for your current filtered dataset.

### Timeline Controls

* **View Mode**: Toggle between a **Cross-Platform** view (each platform as a separate series) and an **Aggregated** view (all platforms combined)
* **Chart Type**: Switch between a **Line Graph** and a **Bar Chart** depending on your preference
* **Data Series**: Choose to display **Post Activity** or **Engagement Activity**


## Data Grid

The Data Grid is the core of the Post Analysis page. It displays individual posts in a structured, sortable, and customizable table.

### Default Columns

Out of the box, the following columns are visible:

| Column | Description |
|  --- | --- |
| **Post Text** | The full text content of the post |
| **Author** | The handle or display name of the post's author |
| **Narrative** | The AI Narrative ID associated with the post, when one exists |
| **Platform** | The source platform (e.g., X/Twitter, Facebook, Reddit) |
| **Engagement** | Total engagement count for the post |
| **Cohorts** | Author cohort classifications associated with the post |
| **Risk Signals** | Flagged risk indicators — see Risk Signal Columns below |
| **Date** | The timestamp when the post was published |


### All Available Columns

The following columns can be added to the Data Grid via the Column Selector:

#### Post Information

| Column | Description |
|  --- | --- |
| **Post Text** | Full text content of the post |
| **Author Handle** | The author's @username |
| **Engagements** | Total engagement count for the post |
| **Author Cohorts** | Cohort classifications associated with the post's author |
| **Platform** | Source platform (e.g., Twitter, Facebook, Reddit) |
| **Post Date** | Timestamp of when the post was published |
| **Cluster ID** | Unique identifier for the post cluster the post belongs to |
| **Toxicity** | Flags posts containing toxic or harmful language |


#### Engagement Metrics

| Column | Description |
|  --- | --- |
| **Total Engagement** | Combined count of all engagement actions |
| **Likes** | Number of likes or reactions |
| **Shares / Retweets** | Number of shares or retweets |
| **Comments / Replies** | Number of comments or replies |
| **Quotes** | Number of quote-retweets or quote posts |


#### Author Information

| Column | Description |
|  --- | --- |
| **Author Handle** | The author's @username |
| **Cohorts** | Author cohort classifications |


#### Risk Signals

| Column | Description |
|  --- | --- |
| **Bot-like** | Flags posts from accounts exhibiting bot-like behavior |
| **Anomalous** | Flags posts showing anomalous amplification or behavior patterns |


### Adding and Removing Columns

**To add or remove columns:**

1. Click the **Column Selector** button (located in the top-right area of the Data Grid)
2. A panel will appear listing all available columns
3. Check or uncheck columns to show or hide them
4. Your configuration is automatically saved and will persist across sessions until you make another change


**Note:** Your column preferences are remembered across sessions. The next time you visit the Posts page, your custom layout will be restored automatically.

## Filtering Posts with the Segment Bar

The Segment Bar at the top of the page is a powerful tool for narrowing down the posts displayed in the Data Grid. All filtering happens in real time — both the Stats Bar and the Data Grid update immediately as you apply filters.

### Search

Use the query field to search post content using boolean operators:

| Operator | Purpose | Example |
|  --- | --- | --- |
| **AND** | Both terms must appear | `misinformation AND election` |
| **OR** | Either term can appear | `vaccine OR vax` |
| **NOT** | Exclude a term | `climate NOT hoax` |
| **"quotes"** | Exact phrase match | `"coordinated inauthentic behavior"` |
| **(parentheses)** | Group conditions | `(bot OR automated) AND risk` |


**Important:** Boolean operators must be typed in **UPPERCASE** to be recognized.

### Available Filters

* **Date Range** — Limit posts to a specific time window using presets or a custom range
* **Platform** — Filter by one or more social media platforms
* **Risk Signals** — Filter for posts flagged as Bot-like, Toxic, Harmful, or Manipulated
* **Cohorts** — Filter by author cohort classification
* **Additional Filters** — Depending on your project, you may also have access to entity filters, engagement ranges, and more


### Saving and Reusing Filters

You can save your current filter combination as a **Segment** for quick reuse. See the Segment Bar User Guide for full details.

## Tips & Best Practices

### For Efficient Analysis

* **Start with the Stats Bar** — Get a high-level read on your dataset before diving into individual posts
* **Use the Activity Timeline to identify spikes** — Click and drag on a spike in activity to zoom in on that time window in the Data Grid
* **Save effective filter combinations** — When you find a filter combination that yields good results, save it as a Segment for fast retrieval in future sessions


### For Column Management

* **Add Risk Signal columns for threat investigations** — Surface Bot-like, Toxic, and Anomalous flags together to quickly triage high-risk posts
* **Add Engagement breakdown columns** — Likes, Shares, Comments, and Quotes give you a fuller picture than Total Engagement alone
* **Remove noise** — Hide columns you don't need for a given workflow to keep your view clean and focused
* **Your layout is persistent** — Don't worry about reconfiguring columns every session; your setup is remembered automatically


### For Large Datasets

* **Filter first** — Use the Segment Bar to narrow your dataset before analyzing individual posts
* **Use date range selection on the timeline** — Isolate time-specific activity to focus your review
* **Sort by risk signal columns** — Sorting by Bot-like or Anomalous flags can quickly surface the most relevant posts for investigation


## Troubleshooting

### "No posts are showing"

* Check your active Segment Bar filters — they may be too restrictive
* Verify your date range includes the expected time window
* Confirm the correct project is selected


### "My column changes weren't saved"

* Column configurations are saved automatically — try refreshing the page
* If the issue persists, reconfigure your columns and navigate away and back to confirm persistence


### "The Narrative column is empty for some posts"

* Not all posts are associated with an AI Narrative — the field is only populated when the post has been matched to a narrative with sufficient confidence


### "Stats Bar numbers don't match my expectations"

* The Stats Bar reflects the **currently filtered dataset**, not the entire project
* Check for active filter pills beneath the Segment Bar and remove any unintended filters


## Summary

The Post Analysis page gives analysts the tools they need to work directly with raw post data:

* **Stats Bar** for instant, filter-aware metrics at the top of the page
* **Activity Timeline** for temporal trend analysis and time-range filtering
* **Segment Bar** for powerful boolean search and multi-dimensional filtering
* **Customizable Data Grid** with a full suite of columns spanning post content, engagement, author details, and risk signals