Tag with AI (beta)

Last updated:

Use AI tagging to capture, review, and apply tag suggestions directly in the Visual Design Studio (the designer). If you're new to AI tagging, start with Overview of AI tagging (beta) for an overview of how it works.

Important: AI tagging requires a paid Pendo subscription. If your subscription is on the Pendo Free plan, you can only tag manually.

Prerequisites

To use AI tagging, you must:

  • Be in a subscription where an admin has enabled AI tagging in Settings > Subscription settings > AI access, with the AI tagging box checked.
  • Have permission to tag the application. AI tagging uses the same permissions as manual tagging, so it isn't limited to admins.

Access AI tagging

  1. Go to Product > Pages > Tag Pages or Product > Features > Tag Features. Both entry points open the same AI tagging experience.
  2. Select Launch Designer to open your application in the designer.
  3. AI tagging opens by default.
aitagging-fristscreen

Select More options (...) to change the panel layout between Popover (default), New Window, or Fullscreen, and to view available keyboard shortcuts. Reset AI tagging resets any suggestions already generated.

Tip: To switch between AI tagging and classic tagging, select the AI tagging dropdown at the top of the panel and choose Classic tagging, or select More options (...) and choose Switch to classic mode. To return to AI tagging from classic mode, select AI Enriched Tagging.

Configure your tagging setup

When the designer opens, two recommended actions appear before you start tagging. These are optional but will improve the quality of AI tagging's suggestions. Both are app-level settings that apply across your subscription. For the full list of configuration options, see Additional settings below.

aitagging-configsettings

  1. Set naming convention (admin only). Describe your team's naming conventions in the text box. This helps AI tagging generate suggestions that match your existing structure and terminology. If you already have a lot of manual tags, setting this first is especially recommended. This setting is controlled by subscription admins and applies to all tagging sessions in the subscription.
  2. Set tag rule preference (admin only). This sets the order AI tagging follows when choosing the selector for each suggested tag's rule. AI tagging works down the list and uses the first selector that matches the element. Selectors near the top of the list are more stable and keep working when your application's code changes. Selectors lower down are more likely to break when your application changes. Pendo provides a recommended default order. You can reorder selectors by dragging them, or turn off any you don't want AI tagging to use. This setting is controlled by subscription admins and applies to all tagging sessions in the subscription.

    The list distinguishes two kinds of selector:
    • Selectors marked as recommended but without retroactive data are custom HTML attributes that haven't been added to your Pendo subscription yet. When you accept a tag using one of these, Pendo automatically adds it to your subscription. However, these selectors only capture data if the attribute is already present in your application's code. If it isn't, no data will be collected. For more information, see Custom HTML attributes.
    • Selectors marked as not recommended, such as text :contains() and positional :nth-child(), rely on the wording or position of an element rather than a stable identifier. Because wording and layout change often, rules built on these selectors break easily, and a broken rule can stop collecting data without warning. AI tagging leaves them off by default, but you can turn them on if no stronger selector is available. If you find AI tagging regularly falling back to weak selectors, working with your engineering team to add stable custom HTML attributes is the most effective fix. See Strengthen tagging with your engineering team below for more information.
  3. If your application serves content in multiple languages, enable Localization aware to prevent AI tagging from relying on text-content selectors, since visible text changes per locale.
  4. Optionally, select Last resort to let AI tagging fall back to matching an element by its visible text or position on the page. This only applies when nothing more reliable is available. These rules are more likely to break if the page's content or layout changes, which is why the setting is off by default.
After you've set up AI tagging, you can begin to capture tags.

Tag pages and features

To capture pages and features:
  1. Navigate to the page in your application that you want to tag.
aitagging-taggingoptions
  1. Review your session settings at the bottom of the modal. To edit these, select Adjust to set the following:
    • Tag method. Create, Update, and Maintain are all enabled by default.
    • Sitewide. Tag UI elements that exist across your app.
    • Only visible. Don't tag UI elements that are hidden or closed.
    • Auto-scan async content. Tag UI elements that can but don't appear during the initial scan for tags.
  2. Select Done save your session settings.
  3. Select Start tagging. You see different options for how to tag. Select one of the following:
    • This page. Tag the page and features on this page only.
    • Multiple pages. Capture several pages in one session before reviewing suggestions. Browse through the pages you want to tag. AI tagging captures each page's DOM as you navigate. When you're done, select Stop and tag with AI to generate suggestions for all captured pages at once. Suggestions may take a while to load.
    • A container on this page. Select an area of the UI and to tag features within it.
    • Manual tagging. Switch to classic tagging.
  4. AI tagging captures the DOM of the current page and generates suggestions in the designer for the page tag and its features. Large pages are captured in chunks, so their suggestions can take longer to load. Suggestions may take a few moments to load.

    Review suggested tags

    AI tagging identifies and suggests page tags automatically alongside feature tags. You don't need to tag pages separately. Suggestions are sorted into four tabs:

    • Create. These are new pages and feature tags suggested for you.
    • Update. These are existing tags that require changes. For example, you can change the description, product area, and rule.
    • Maintain. These tags are broken or unused and may need to be removed for good tag hygiene.
    • Existing. The rest of the tags associated with your app, which don't require updating.

    In any given tab, the page tag appears first, followed by features. Feature tags are grouped based on a theme, usually location in the app, such as Sitewide, or a UI element like Navigation.

    suggested-tags

    Tip: Use the Layer button to view suggested tags and a heatmap of popular features in your app.

    After tags have been surfaced, review the suggestions:

    1. Expand a tag to see the suggested name, element type, rule, description, and product area.

      You can refine the tag suggestions displayed using the Filters or by typing in the tag name in the search bar.

    2. Select the checkbox next to any page or feature you want to tag. Use Select all to select everything in the list, or deselect individual items to skip them. Nothing is committed to your subscription until you select Create.

      If you're in the Update tab, you can also use Select all or select individual tags to update. When you're ready to make the changes, select Update.

    3. To start over, select Exit at the top of the designer to end the session and start tagging again.
    update-tag

    When you accept tag suggestions, AI tagging creates the new tags and applies the updates in your subscription, and a confirmation message appears. The suggestion list refreshes, and you can continue tagging other pages in the same session.

    Review tags for deletion

    AI tagging surfaces broken or unused tags as deletion recommendations in the Maintain tab. These appear because AI tagging builds on tag maintenance: when it detects a tag with no recent DOM match, it flags it for review rather than leaving it silently broken.

    1. Review deletion recommendations in the designer.
    2. To accept a deletion, select the tag, then select Remove.

    Deleting a tag removes it from your subscription and any reports or guides that reference it. Review deletion suggestions carefully before accepting.

    Strengthen tagging with your engineering team

    AI tagging uses a prioritized rule set to choose the selector for each suggested tag. When it can't find a strong selector, such as a custom HTML attribute or element ID, the suggestion is flagged. Flagged suggestions still work, but they're more likely to break when your application's code changes.

    The fix is to add a pendo-analytics attribute to key interactive elements in your codebase. This gives AI tagging a stable, human-readable anchor for each element, so its suggestions are stronger and more durable. It also creates a flywheel: AI tagging does the work of identifying and suggesting tags, and where selectors are too weak, engineering can strengthen them so the next session produces better results automatically.

    Pendo has created an AI coding skill for engineers. Share it with your engineering team and they can use it with Claude Code or another AI coding tool to audit your React codebase and add pendo-analytics attributes in a single session.

    Download the Pendo Analytics Tagging skill

    Once the attributes are in place and deployed, re-run AI tagging on the same pages. The suggestions will be significantly stronger.

    Additional settings

    Select More options > Manage settings at any time to access the full settings panel. Settings are organized into the sections below.

    Each setting applies at one of two levels:

    • App level: Shared across your subscription. A change affects everyone who tags this application.
    • User level: Applies only to your own tagging session. A change doesn't affect anyone else.

    Capture

    • Suggest product areas (app level): AI tagging suggests product areas for tagged features and pages. On by default.
    • Sitewide mode (user level): Suggests which tags should be applied sitewide, if they're repeated multiple times and/or across multiple pages. On by default.
    • Only visible: Ignore UI elements that are hidden by default.

      Auto-scan async content: Continue scanning after the initial capture of tags.

      Methods: Select whether to create, update, or maintain tags.

    Naming convention (app level)

    Describe how AI tagging should name pages and features in your application. You can set the naming convention in two places: as a recommended action when you first open AI tagging (see Configure your tagging setup), or here in Manage settings at any time. Both update the same setting. Your naming convention can be up to 1,000 characters.

    Updating your naming convention doesn't automatically change existing tag rules. AI tagging gives you the opportunity to review and apply updates to existing tags before accepting any changes.

    Tagging context (app level)

    Add any extra context to help AI tag accurately, for example, product terminology, areas to ignore, or what matters most. This is distinct from the naming convention, which controls format and structure. Tagging context controls scope and intent.

    For example, Ignore the admin debug bar. Treat the left rail as navigation, not features.

    Selector preference (app level)

    Set the order AI tagging follows when choosing the selector for each tag's rule. You can set this in two places: as a recommended action when you first open AI tagging (see Configure your tagging setup), or here in Manage settings at any time. Drag to reorder selectors, placing the most stable ones (least likely to break when your application changes) at the top. AI tagging works down the list and uses the first selector that matches the element. Turn off any selectors you don't want AI tagging to use.

    You can also add a custom selector using the text field at the top of the list.

    Advanced

    • Dry run (user level): Always on. No tags are created or modified until you confirm your selections. Full control will be available in a future release.
    • Debug (user level): Shows verbose logging to help troubleshoot tagging issues. Off by default.

    Frequently asked questions

    Can I choose which suggestions to apply?

    Yes. Use the checkboxes next to each suggestion to select what to accept or reject, or use Select all to apply everything. Nothing is committed to your subscription until you select Accept selected.

Was this article helpful?
0 out of 0 found this helpful