Yes. The system prompt can ingest target programme tiers from a WordPress page (top-15 econ PhD, mid-tier MBA, humanities master's) and recommend a section emphasis accordingly. A STEM-PhD applicant gets steered toward Quant-heavy plans; a humanities applicant gets a Verbal-and-AWA emphasis. Update the programme tier page and the bot reflects the new framing on the next conversation, no retraining needed.

It hands off to TutorBird, Calendly, Acuity, or any tool with a public booking URL via deep link or webhook. The bot collects preferred trial slot, target section, and test date in conversation, then either passes the prospect to your booking link with the slot pre-selected or fires a webhook that creates the appointment server-side. Most services use deep links.

If your hour-bundle pricing is on the pricing page, yes. The bot surfaces trial-session discounts in context and explains when the 20-hour package is cheaper than singles plus separate mock fees. It won't invent discounts; if you want a specific code honoured, add the rule to the prompt. Otherwise it quotes only what's published.

No, and that's intentional. AWA scoring requires a human reviewer who has trained on the ETS rubric for both Argument and Issue tasks. The bot can describe your AWA review process, queue an essay submission to your reviewer, and quote turnaround time. Treat the chat as triage and intake, not grading. Several prep services use the bot to capture the essay text and email it to the AWA reviewer with the prospect's target score.

Yes. The 28+ display conditions let you run a GRE bot on /gre/ pages and a GMAT bot on /gmat/ pages, each with their own tone, source pages, and tutor roster. Conversation logs stay separate so you can review GRE enquiries without scrolling past GMAT ones. Most graduate prep services run three to four bots: GRE, GMAT, LSAT, and an admissions consulting bot.

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 because graduate applicants share sensitive details: target schools, prior application history, work background. You control retention and redaction, and the plugin doesn't add a new vendor between you and the model.

The system prompt should forbid specific score promises (no 'we guarantee +10 Quant'). The bot describes typical practices: 20-30 hours for a 6-point Quant gain, weekly missed-item review, four full-length mocks. It also flags that score guarantees usually come with strings (mock attendance, homework completion), and recommends the consult call rather than the chat for any guarantee terms.

The bot handles non-traditional cases explicitly. Returning-to-academia after 10 years, career-switchers from law to data science, international applicants needing AWA help, each gets a different cadence recommendation. The bot reads your published faculty-fit prose and routes accordingly. For very unusual cases, it offers the consult call as the next step rather than guessing a plan in chat.