Articles in this section

Guide to Data Lab

  • Updated

For editorial teams looking to streamline team workflows and drive informed content strategy with historical reporting, Chartbeat Data Lab integrates flexible report- and visualization-building, scheduling, and exporting capabilities within a secure platform, leveraging a robust set of trusted media-centric metrics to empower your team to build and share customized dashboards and reports.

Templates

Chartbeat Data Lab introduces a streamlined approach to reporting through its customizable templates. These templates serve as starting points for users to build insightful reports tailored to their specific needs.

Currently, Data Labs offers nine core templates:

  • Story Analysis: This template highlights your top-performing articles, ranked by key metrics like page views, engaged time, and more. Gain insights into which content resonates most with your audience and drives engagement.
  • Multi-Site Analysis: This template provides a unified view of performance metrics across all your company’s sites. Easily track, compare, and analyze data from multiple sites to identify trends and optimize overall performance.
  • Author Analysis: This template ranks your authors based on key performance metrics. It provides a clear view of individual author contributions to your content strategy, helping you recognize top performers and uncover opportunities for improvement.
  • Section Analysis: This template provides a comprehensive view of section-level performance. Designed to track key metrics, it empowers publishers to analyze how different sections of their publication are performing.
  • Referrer Analysis: This template provides a comprehensive overview of your canonical referrers and referrer types, including search engines, social platforms, and other traffic sources. Use this dashboard to analyze where your audience is coming from, identify trends, and optimize your strategies for different traffic channels.
  • Visitors Analysis: The Visitors Analysis template provides a comprehensive overview of your audience's demographics, behaviors, and engagement patterns. This template allows you to analyze key metrics such as page views and engaged time across various dimensions.
  • Home Page Analysis: The Home Page Analysis template is a view-only tool that helps users track click-through performance on landing pages over time. It provides insights into how long content has been on each page and identifies first-click and last-click interactions to help understand user engagement better.
  • Scroll Depth Analysis: The Scroll Depth Analysis template provides valuable insights into how far visitors scroll on individual stories compared to site-wide averages. By utilizing the word count filter in conjunction with this template, users can further refine their analysis and compare stories of similar lengths.
  • Free Form : The Free Form template allows users to explore and analyze their data with full flexibility. After selecting a site and date range, all data loads, enabling users to group, filter, export, and build charts directly within the page.

Each template provides a snapshot of key metrics over a 7-day period. The “DataModel” page contains the full set of data available based on your selected sites and date range.

While these templates offer a foundation for analysis, users have the flexibility to enhance their reports by adding multiple dimensions and expanding on the metrics they wish to explore further.

Main Metrics

There are seven primary metrics in Chartbeat Data Lab and a long list of available groups and filter parameters to access the data you're looking for.

  • Average Scroll: The average maximum depth that visitors have scrolled to, in pixels, from the top of the page.
  • Average Engaged Time: The average amount of time in seconds visitors actively spent on a page.
  • Scroll Starts: The ratio of page views where a scroll event occurred, expressed as a decimal.
  • Total Engaged Time: The total amount of time in seconds spent actively on a page.
  • Pageviews: The number of pages viewed.
  • Loyal Visitor Page Views: Pageviews from visitors who visit your site an average of every other day, or a minimum of eight unique days in a 16 day period.
  • Quality Page Views: Pageviews that received at least 15 seconds of engaged time.

Note: The Uniques metric is not available because our hosting system does not support the format in which it is calculated (HyperLogLog); however, Uniques are still accessible in other areas of the product, and we are exploring ways to add it to Data Lab in a reliable way.

Columns

Use the dimensions to select the grouping criteria for which your metrics are returned, such as title, referrer, geographic location, and page specific data.

  • Author: The author of a page.
  • Browser: The web browser used by a visitor to access the site (e.g., Chrome, Firefox, Safari).
  • Canonical Referrer: A rolled-up view of the referrer domain. For instance, traffic from m.facebook.com and facebook.com both have a canonical referrer of Facebook.
  • Character Count: The total number of characters in an article. (scraper)
  • Content Type: The content type (e.g. gallery, article) as tagged on your site or AMP page.
  • Country: The visitor's country.
  • Device: The device used (mobile/desktop/tablet).
  • Flesch Reading Ease: The Flesch Reading Ease gives a text a score between 1 and 100, with 100 being the highest readability score. (scraper)
  • Html Page Type: (e.g., NewsArticle, VideoObject, BlogPosting, LiveBlogPosting) (scraper)
  • Internal Path: The page path that users clicked away from when they opened a new page. Useful for learning about historical click-through data.
  • Language: Language code or locale of the article's content. (scraper)
  • Loyalty Type (AKA Visitor Frequency): Returns New, Returning, or Loyal.
  • Os: The operating system used.
  • Page Type: The type of page, returned as either Article, LandingPage, or Unknown.
  • Path: The URL/path of the page.
  • Pinger Source aka Distribution: The distribution channel used (site/app).
  • Publish Date: Date the article was published. (scraper)
  • Referrer: The referring domain.
  • Referrer Type: The page referrer type: social, search, direct, internal, and external (aka Links).
  • Region: The visitor's region (returns states for US).
  • Scroll Depth Histogram: Histogram of the number of pixels scrolled (100 pixel bins).
  • Scrolled Page Views: Number of pageviews where the user scrolled at all
  • Section: The section of a page visited.
  • Site: Specific site that is being tracked.
  • Site Experience: The site experience for the visitor (AMP/Standard). An NA value indicates that the visitor comes from an off-site distribution channel.
  • Subdomain: The specific subdomain (e.g. blog.chartbeat.com, chartbeat.com).
  • Subscriber Type: The visitor's subscriber status (Guest, Registered, Subscribed, Unspecified).
  • Title: Title of the article.
  • Total Scroll: Sum of the total number of pixels scrolled.
  • Tz Day (Timezone Day): The calendar day in your time zone.
  • Utm Campaign: Identifies a specific promotion/campaign.
  • Utm Content: Identifies what specifically was clicked.
  • Utm Medium: Identifies the type of link, 'email' for example.
  • Utm Source: Identifies to which site a link was posted.
  • Utm Term: Identifies search term/terms.
  • Visit Frequency: Number of days out of the last 15 that this user has visited; min 0, max 15. 
  • Word Count: Number of words in the full text of the article. (scraper)

You can also create new columns in Data Lab using any of the data points above for your own custom calculations.

How Chartbeat Collects Page Information

We rely on your integration of our JavaScript configuration variables to populate key page properties in your Chartbeat dashboards, like page sections, authors, paths, and titles.

Chartbeat also collects key details by visiting the page and pulling important information from the page’s content and structure. This data is then linked to visitor engagement metrics to provide deeper insights in Data Lab.

To make this process as smooth as possible, we follow common web standards. The best way to provide this information is using JSON-LD.

How to Format Your Metadata for Chartbeat

  • Use our recommended format, JSON-LD: This involves adding a small piece of structured data inside a <script> tag in the <head> section of your webpage. Learn more here.
  • Use schema.org standards: This ensures consistency and makes your content easier to process.

Why does this matter? Providing metadata in this structured way helps Chartbeat extract information accurately and consistently, improving how we track and analyze engagement on your site. 

The "Columns" values above marked with 'scraper' are collected in Chartbeat via our web scraper. If you see any issues with these values, this can indicate an issue with your JSON-LD structured data or with our scraper accessing the web page. Read more about whitelisting our scraper here.

Create and Manipulate a Report

Once a template is selected, if you have “Explorer” access you'll have the ability to manipulate the data by choosing which fields to add to the Rows, Columns, and Values of your table.

  1. Scroll down and select “Explore” in the bottom right corner
  2. Drag fields to Rows or Columns: These will be your pivot table's row or column headers.
  3. Drag fields to Values: These fields will be calculated or summarized. 
  4. Hide Columns: To hide columns you are not interested in seeing, simply select the columns, right click and select "hide."

Filter your data

Filters let you only focus on a particular layer of data, for example if you only want to see the metrics for a particular section, referrer type, or page path.

We support 7 different filter types:

  • Include
  • Exclude
  • Range
  • Date Range
  • Text Match
  • Boolean
  • Top N

Filters can be accessed from the filter button in the element's inline toolbar:

Build a Visualization

  1. Open a Template or Document
    • Go to Data Lab and open an existing template or document or create a new one
  2. Toggle the “Explore” slider on
    • The “Explore” slide toggle is located on the bottom right corner of the page: 
  3. Add a New Visualization
    • Click on the "+ Add Element" button.
    • Select "Visualization" from the dropdown menu.
    • Select the dataset with “Explore” in the title
  4. Choose Visualization Chart
    • Example: In the visualization type selector, choose "Bar Chart".
  5. Configure the Data
    • Drag and drop the fields you want to use into the "X-Axis" and "Y-Axis" sections.
      • For the X-Axis, choose the categorical data you want to group by (e.g., days/ TZDay).
      • For the Y-Axis, choose the numerical data to measure (e.g., pageviews).
  6. Customize Your Chart
    • Use the "Settings" panel to customize the appearance of your chart.
    • Change colors, labels, and other display options to fit your needs.
  7. Apply Filters (Optional)
    • If you need to filter your data, drag fields into the "Filters" section and set the criteria.
  8. Save and Share
    • Once your visualization looks good, save your workbook.
    • Share it with your team or export it as needed. 

Send and schedule exports

Immediately send an element or workbook to one more destinations:

  • Direct download
  • An email list
  • Webhook

Set up a scheduled export of an element or workbook to one or more destinations:

  • An email list
  • Webhook

Data can be exported as a PDF, Excel doc, or PNG. You can export the entire worksheet, or only certain tables/visualizations as needed.

Save Document

Use the “Save As” function located at the bottom right of the page to save your document for future use.

  1. Click on “Save As”
  2. Give your document an identifiable name
  3. Click “Save”

Share a Document within Data Lab

In order to share a document with others within Data Lab, use the “Save As” function and select a “Workspace” to save to.

  1. Click on “Save As”
  2. Select Workspaces
  3. Share with Explorers or Viewers
  4. Click “Save”

Related articles