Yes, by booking a diagnostic. The bot does not guess which exam fits a student because that decision needs real data. Instead, it surfaces your dual diagnostic (SAT vs ACT, GRE vs GMAT) and books that as the first step. After the diagnostic, course recommendations are grounded in score baseline rather than a hunch. Most services run this exact pattern because it converts diagnostic-takers into course enrollees at high rates.

If your score-guarantee terms are on a WordPress page, yes. The bot reads and quotes the terms exactly (must complete homework, must attend mocks, refund or free retake structure) without paraphrasing. Score guarantees are legally sensitive, so the bot should quote your published language rather than summarising. If you do not offer a guarantee, configure the prompt to say so when asked.

Yes. The bot recognises common test dates (March, May, June, August, October, December SAT; February, April, June, July, September, October, December ACT; LSAT and MCAT testing windows) and recommends course start dates that finish 2-4 weeks before the test, with mock-exam coverage built in. Recommendations are grounded in your published course schedule rather than aspirational timelines.

Yes. The bot asks about learning preferences and test-date pressure, then recommends group, semi-private, or private accordingly. Most services have a tier preference (group is cheaper and most popular; private is for high-stakes retakes or specific weak areas). The bot mirrors that hierarchy and quotes accurate pricing for each tier from your real pricing page.

Yes, if your roster includes LSAT, MCAT, GMAT, GRE, or specialty grad-school exam prep. The bot reads those tracks separately because the conversation pattern is different (longer prep arcs, higher price points, more research about instructor backgrounds). Most services run a separate landing page for graduate prep, and the bot adapts tone and recommendations when the visitor enters from there.

Yes. If mock-exam dates are published, the bot quotes them and explains how many mocks come with each course package. Mock-only packages (for students who want practice without the full course) are also quoted if you offer them. For families deciding between courses based on mock count, this is one of the most useful conversations the bot has.

Test prep enquiries spike before each major test date. The bot scales because it reads from WordPress and calls your model API, with no admin bottleneck. Conversation logs let you see which exams and course starts are filling up before they sell out. Most services use this signal to open additional sections or shift instructor hours before the cycle peaks rather than turning families away mid-rush.

Logs stay in your WordPress database. Bring-your-own-key sends prompts directly to OpenAI, Anthropic, or Google with no third-party processor in between. That matters when families share scores, struggles, or sensitive deadline pressure: you control retention and redaction, and the plugin does not add a new vendor between you and the model. Apply your existing WordPress security to the log table.