Frequently asked
questions.
If your question isn't here, reach us at support@fayebot.com
Getting started
How does the embed work?
You copy a single script tag from your FayeBot dashboard and paste it before </body> on your website. The script loads asynchronously — it won't slow down your page. Once loaded, Faye's chat widget appears at the bottom-right corner of your site.
Do I need a backend or any server setup?
No. FayeBot handles everything server-side. Your embed script connects to our API using your API key. You don't need to write any backend code or configure any infrastructure.
How long does it take to go live?
Under 10 minutes from signup to live widget. Create your account, create an app in the dashboard, copy the embed script, paste it before </body> on your site, and Faye is live.
Do I need to train Faye on my product?
Not upfront. Faye starts from your configured greeting and system prompt. You write a short description of what your product does and who your customers are — Faye uses that as her persona. As she accumulates conversations, she learns from them automatically via RAG.
Faye's behavior
What does Faye actually say to visitors?
Faye starts with your configured greeting, then responds naturally to whatever the visitor says. She tries to understand what they need, asks clarifying questions when relevant, and keeps the conversation moving toward a captured lead or a resolved question.
How does lead extraction work?
Faye parses the conversation in real time and extracts structured data — email address, name, company, intent (e.g. 'pricing inquiry', 'support request', 'demo request'), and sentiment. This happens automatically; you don't need to configure extraction rules.
What triggers an escalation?
Four things trigger an escalation: (1) The visitor explicitly asks to talk to a human. (2) Faye detects frustration, urgency, or anger in the conversation. (3) A configurable rule is triggered (e.g. 'competitor mentioned', 'billing issue'). (4) The conversation has been open for more than 10 minutes without resolution. All escalations are flagged with a reason code.
What happens when an escalation is triggered?
The conversation is flagged in your dashboard under the Escalations tab with a reason and the full conversation thread. You or a team member can take ownership, respond to the visitor directly (outside of Faye), and mark it resolved in the dashboard.
Does Faye get smarter over time?
Yes. Every conversation is embedded using Gemini and stored as a vector. When Faye responds to new visitors, she retrieves semantically similar past conversations as context. The more conversations she has, the more contextually accurate her responses become.
Privacy & data
What data does FayeBot collect from visitors?
FayeBot collects the contents of conversations (messages typed by the visitor), extracted lead data (email, name, company, intent), and the page URL where the widget is loaded. No cookies are set on visitors without your configuration. No third-party tracking.
Where is conversation data stored?
Conversation data is stored in a managed PostgreSQL database (Neon), hosted on AWS in the ap-southeast-1 region, encrypted at rest using AES-256. Embeddings (vectors for RAG) are stored alongside in pgvector. Data is encrypted at rest and in transit.
Can visitors request their data be deleted?
Yes. Under GDPR and similar regulations, you are the data controller for your visitors' data. You can delete any conversation or lead record from the dashboard at any time. We provide a data deletion API for automated compliance workflows.
Does FayeBot share visitor data with third parties?
No. Conversation data is used only to operate the service for your account. We use Gemini API for AI responses and embeddings — conversation content is sent to Google for inference — but is not used to train Google's models.
Plans & billing
What are the differences between the plans?
Free gives you 1 app and 10 conversations per day — enough to evaluate Faye. Pro (₹2,499/mo) unlocks 3 apps, 500 conversations/day, escalations, widget customization, and API access. Max (₹6,999/mo) goes to 10 apps and unlimited conversations, suited for agencies or teams running multiple products. Enterprise is a custom quote for unlimited apps and volume.
Can I cancel anytime?
Yes. Cancel from your account settings at any time. You keep access to your paid plan until the end of the billing period. No cancellation fees.
Do you offer refunds?
If you cancel within 7 days of your first Pro or Max payment, we'll refund it fully. After 7 days, we don't offer refunds for partial billing periods — but you can cancel to avoid the next charge.
What payment methods are accepted?
Major credit and debit cards via Razorpay. UPI and net banking available for Indian customers. All payments are processed securely by Razorpay — we never store card details.