AI Tagging

What is AI Tagging?

AI tagging is sometimes described as auto tagging, automatic tagging, or automated tagging. The important distinction is that AI tagging does more than capture activity: It helps interpret product structure so teams can decide what is worth tracking and how those tags should be organized.

In product analytics and digital adoption tools, tags are what make behavior understandable. A page tag might track visits to a dashboard, while a feature tag might track use of a button, filter, or workflow step. When tags are complete and well maintained, teams can trust their analytics, target in-app guides more precisely, build more accurate segments, and give AI systems a better understanding of the product.

How does AI tagging work?

AI tagging typically analyzes a product's application structure, navigation, page patterns, and UI elements to determine which parts of the product may be worth tracking. It can then recommend tags with useful metadata, such as a proposed name, description, product area, rule, or page-to-feature relationship.

A strong AI tagging workflow should include:

• Suggested page and feature tags
• Proposed tag names and descriptions
• Suggested product areas or groupings
• Matching rules that define what each tag tracks
• Confidence or quality signals for suggested tags
• Recommendations for improving, fixing, or removing existing tags
• Human review before changes are applied

That last point matters. AI tagging can reduce the manual work of finding and defining tags, but it should not remove human control. Product teams still need to decide which suggestions matter, how granular their tagging strategy should be, and whether the resulting structure reflects how the business understands the product.

Why is AI tagging important?

Manual tagging creates a persistent instrumentation backlog. Teams often need product analytics quickly, but first they have to decide which pages and features to track, define rules, name everything consistently, and keep those tags updated as the product changes.

When tagging coverage is incomplete or stale, the impact shows up across the product organization. Analytics reflect only what someone remembered to tag. In-app guides can target the wrong location or miss the intended audience. Segments are built from incomplete behavioral data. AI-generated insights are limited by gaps in the product roadmap. Teams spend time maintaining instrumentation instead of acting on insights.

AI tagging helps reduce that burden by turning tagging from a fully manual setup task into a guided review workflow. The goal is not to tag everything automatically. The goal is to build and maintain coverage of the right things, with less manual effort.

Every analytics report, guide target, behavioral segment, and AI-generated insight depends on a clear product map. If the right pages and features are not tagged, teams are working from an incomplete picture of how users actually move through the product.

AI tagging helps build that map by using artificial intelligence to identify pages, features, product areas, and interactions in a digital product, then suggest structured tags that teams can review and apply. These tags help turn raw product usage into organized, meaningful data by giving product activity clear names, rules, descriptions, and relationships.

AI tagging vs. auto tagging

Auto tagging is a broader term for automatically applying or suggesting tags. AI tagging is a specific form of auto tagging that uses artificial intelligence to understand product structure, identify meaningful pages and features, and recommend structured tags.

AI tagging vs. autocapture

AI tagging is related to autocapture, but the two are not the same.

Autocapture records raw user activity, such as clicks, page views, and other events, often without requiring teams to define each event in advance. That can create broad behavioral coverage, but raw captured events may lack meaningful names, hierarchy, product areas, or business context.

AI tagging adds structure. It helps identify which captured interactions matter, names them in a way teams can understand, and organizes them into useful pages, features, and product areas.

In short: autocapture collects activity; AI tagging helps turn activity into an organized product map.

Does AI tagging replace manual tagging?

AI tagging can reduce the manual effort required to create and maintain tags, but it does not replace product judgment. Teams still need to decide which suggestions to accept, what naming conventions to follow, how detailed their tag structure should be, and which parts of the product matter most for analytics, guide targeting, segmentation, and AI.

For example, an AI system might identify several buttons, filters, pages, or navigation items. A product manager may decide to track only the elements tied to meaningful workflows or adoption goals. AI tagging accelerates discovery and setup, while human reviewers keep the tagging strategy aligned to business context.

What makes an AI-generated tag reliable?

A reliable tag depends on a stable rule. If a tag is based on a strong identifier, such as a custom HTML attribute or element ID, it is more likely to keep working when the product changes. If a tag relies on weaker selectors, such as visible text or an element's position on the page, it may break when copy, layout, or code changes.

Good AI tagging systems should flag weaker rules so teams can review them. In some cases, product and engineering teams may choose to add more stable identifiers to key UI elements so future tag suggestions are stronger.

How does AI tagging improve product analytics?

Product analytics depends on a clear understanding of what users are doing inside the product. If important pages or features are not tagged, teams may miss adoption patterns, drop-off points, workflow friction, or opportunities for improvement.

AI tagging helps improve product analytics by making it easier to establish and maintain tag coverage. With better coverage, teams can answer questions like:

• Which features are users actually adopting?
• Where do users drop off in a workflow?
• Which accounts are engaging with high-value areas of the product?
• Which parts of the product need better onboarding or guidance?
• How does behavior differ by role, segment, plan, or lifecycle stage?

The more complete and structured the product map, the more trustworthy the analysis.

Does AI tagging support in-app guides, segments, and AI?

Yes. Tags are not only used for reporting. They also help teams target in-app guides, define behavioral segments, and give AI tools a structured understanding of the product.

For in-app guides, accurate tags help teams deliver messages in the right place, to the right users, at the right moment. For segmentation, complete tags help teams define audiences based on real product behavior. For AI, structured tags provide cleaner context about how the product is organized and how users interact with it.

AI tagging helps strengthen the product map that analytics, guides, segments, and AI systems all depend on.

AI tagging in Pendo

In Pendo, AI tagging runs inside the Visual Design Studio. It reads an application's routing configuration and DOM, then suggests page and feature tags for review.

Each suggestion can include a proposed name, rule, description, product area, and page-to-feature relationship. Pendo AI tagging can also surface new tag suggestions, improvements to existing tags, and fixes or removals for broken or unused tags.

AI tagging is not a standalone Pendo product. It is a feature that helps multiple Pendo capabilities work from a cleaner product map. Once pages and features are tagged, teams can use that structure in Product Analytics, in-app guides, segments, and AI-powered workflows such as Leo.

Pendo AI tagging is review-based. Nothing is created, changed, or deleted until a user reviews the suggestions and accepts the selected changes. Pendo also includes safeguards for tag quality and data handling, including client-side sanitization before AI analysis and selector preferences that prioritize stronger rules such as custom HTML attributes and element IDs.

Common Questions About AI Tagging

Is AI tagging the same as auto tagging?

AI tagging is a type of auto tagging. Auto tagging is the broader idea of automatically applying or suggesting tags, while AI tagging uses artificial intelligence to understand product structure and recommend meaningful tags.

Is AI tagging the same as autocapture?

No. Autocapture records raw activity, while AI tagging helps organize activity into named pages, features, and product areas. Autocapture creates coverage; AI tagging adds structure and meaning.

Can AI tagging change or delete tags automatically?

It depends on the system, but a review-based workflow is safest. In Pendo AI tagging, changes are not applied until a user selects which suggestions to accept.

Does AI tagging work with existing tags?

Yes, AI tagging can support existing tags by suggesting improvements, identifying broken tags, or recommending removals for tags that no longer match the product structure.

Does AI tagging replace manual tagging?

AI tagging reduces the manual legwork involved in creating and maintaining tags, but teams still need to review suggestions and decide what should be tracked.

Is AI tagging the same as feedback auto-tagging?

No. Feedback auto-tagging categorizes qualitative feedback, such as requests or comments. Product AI tagging identifies pages and features in an application for analytics, guides, segmentation, and AI insights.

Does AI tagging cover track events?

Not always. In Pendo's current AI tagging release, AI tagging focuses on pages and features from the application structure. Custom-coded track events are outside the current release scope.