Native variable maps exist for LearnDash (sfwd-courses), LifterLMS (llms_course), and Tutor LMS. Custom LMS or course CPT setups work via a manual variable map: list the CPT slug and the meta keys for level, duration, prereq, and price. The bot reads these at conversation time, so any plugin storing course data in wp_postmeta can be wired up in an afternoon.

By asking. The bot does not assume. The standard intake asks the learner to describe a specific task they can already do and one they cannot, which is more reliable than a self-rated beginner-intermediate-advanced slider. The answers are written to the user profile for future conversations and any sales follow-up.

Yes. If a course has a prereq meta field, the bot will check the learner's stated level against the prereq before recommending. When the prereq is not met, the bot recommends the prereq course first and explains the path. This avoids the worst LMS experience: enrolling in a course you cannot follow and silently dropping out.

The bot reads course price meta and any bundle product associations from WooCommerce or EDD. When a learner is recommended two courses, the bot can quote the individual prices and the bundle price if one exists. The final message includes a direct link to the checkout for the recommended path.

The bot does not replace LMS quizzes. It complements them. For visitors who have not enrolled yet, the bot handles the recommendation conversation. For enrolled learners, your LMS's native quizzes and progress tracking continue to handle assessment. Some teams use the bot as an in-course study companion as well, scoped via display conditions.

The stated goal, current level, and time budget are written to the WordPress user meta (or to a guest record for non-logged-in visitors). Subsequent conversations pick up where the last one ended, and your sales or success team can see the same data in the admin. Exports to CSV are available for any cohort analysis.

Indirectly. The bot itself runs only when the visitor opens it, but the recommendation conversation can fire a webhook to your CRM or ESP with the goal and recommended course. From there you can run a tailored sequence: "You were looking at Technical SEO for Agencies, here is a free chapter to get started." Open rates on these warm sequences run notably above generic newsletter follow-ups.

No, not when the prereq meta is populated. The system prompt explicitly forbids advancing past stated level without acknowledging the gap, and the bot is instructed to recommend a foundational course first when the gap is significant. Conversation logs let you spot any drift and tighten the prompt or fix a missing prereq in the course CPT.