How WhizzMS works

The real setup flow, end to end - plus straight answers to the questions people actually ask before signing up.

Getting live

1

Create your account and first business

One account can own multiple businesses (each isolated - see the FAQ below). Pick a category on signup and WhizzMS pre-fills sensible starting prompts, widget defaults, and field-mapping hints for that industry, so you're not starting from a blank screen.

2

Connect a data source

Point at your existing REST API. WhizzMS samples one real record, discovers every field automatically (including nested and array fields), and lets you check off which ones to embed for search versus keep as metadata - no hand-typed schema.

3

Configure your AI provider and prompts

Bring your own API key (OpenAI live today; Gemini, Claude, OpenRouter, and Ollama cataloged) and choose a search mode - Vector Search Only, Hybrid, or Full RAG. Six prompt types (System, Search, Embedding, Query, Response, Follow-up) are versioned, so you can test a change and roll it back if it doesn't work out.

4

Customize and embed your chat widget

Pick a design template, then fine-tune colors, position, bubble size/shape, welcome message, and suggested questions from a live preview - then drop in one script tag, or call the headless API directly if you're building your own front end.

5

Go live

Your Billing page tracks items ingested, storage used, and businesses against your plan in real time, so you always know exactly how much room you have left before you need to upgrade.

Frequently asked questions

Plans are usage-based - each one caps how many items you can sync, how much storage your synced content can use, and how many businesses one account can run, not how many chat messages you send. Any of those caps can be set to unlimited on a plan. If your business needs something bigger than what's listed, contact us directly for a custom plan.
New syncs simply stop - your existing data stays exactly as it is and stays fully searchable, nothing gets deleted or disabled. Upgrading your plan unblocks syncing again immediately, picking up right where it left off.
Yes. Every business gets its own vector store collection - never a shared one - plus every database record is scoped to that business internally. One business can never see or search another's data, even on the same account.
Vector Search Only returns the closest matching records with no AI generation cost. Hybrid combines dense vector search with keyword search for better relevance, still with no generation cost. Full RAG additionally sends the retrieved results to your configured AI provider to generate a real written answer. Pick per business based on how much you want to spend on generation.
Yes - each business gets its own users, roles, AI configuration, and data, all switchable from the same account, up to whatever your plan's business limit allows.
Yes - pick a design template (Classic, Minimal, or Bold, each a different panel/message shape and shadow treatment), then independently set colors, position, bubble size and shape, animation style, and header title. Every visible piece of text - welcome message, suggested questions, header title - is editable too.
No. Connecting a data source, configuring a provider, and embedding the widget are all done from the dashboard - the field discovery, embedding, and retrieval pipeline underneath is handled for you.