Legal services platform

An AI will advisor that answers questions and captures leads for a legal services platform.

Published

Console inbox with a conversation marked needs a call, a sales brief and attribution for a sample visitor
The challenge

Most visitors arrive on a guide about wills, read, and leave. A generic chatbot risked wrong legal information, quoted fees, promises of outcomes, and leads handed to sales with no context. Some of the platform's own guide content also contained legal errors, so the assistant could not simply trust whatever it retrieved.

The Quellix mandate

Answer a reader's first questions accurately from reviewed facts, never quote a price or guarantee an outcome, move to a human specialist at the right moment, capture contact details inside the conversation, and give the business full visibility of every answer, source and decision.

What shipped

01

A helpful first conversation on the guide page

A single script tag adds a launcher to any page. It records where the visitor came from and what they read, offers a nudge and starter questions that fit the guide, and opens the chat in an isolated frame. Readers can type, or hold a button and speak in English, Hindi or mixed speech.

02

Reviewed facts first, sources second

The assistant always has a reviewed set of will-law and service facts in front of it, each with a citation and a review field for a lawyer to sign off. When a question needs more, it retrieves passages from statutes, service pages and guides. Reviewed facts outrank guides, so a guide that gets the law wrong cannot override them.

03

Check the answer before the reader sees it

One step classifies the question, writes the answer and judges whether it is time to offer a specialist. A verifier then checks that every claim is supported, that no price or guarantee appears, and that the answer is not personal legal advice. A failed draft is rewritten once, and if it fails again the reader gets a safe reply that offers a specialist.

04

Hand off to a person, conversationally

Pricing questions are never answered with a number, and disputes, urgency or an explicit request for a call move to a specialist faster. From the third message the assistant can offer a specialist, and by the fourth it always does; the visitor replies with a name, then a number or email. There is no form and no live takeover: a specialist calls back with a short AI brief and a suggested next step.

05

Give the business a console, not a black box

An inbox shows who needs a call and who is only browsing. Each conversation shows the transcript, the sources and guardrail results under every answer, the page journey, attribution, the brief and the CRM status. Insights, a knowledge screen, a sales-system view and a hash-chained audit log round it out.

How the agents are wired together
Reader on a guidepage · sourceTyped questionor voiceReviewed factsalways firstStatutes · guidesretrieved passagesAnswer stepclassify · answer · hand-offVerifiergrounded · no price · no promiseSafe fallbackspecialist offeredContact in chatname, then numberSales briefsummary · next stepSpecialistcalls backWHAT THE VISITOR BRINGSANSWER AND CHECKHAND-OFF

What it looked like in action

Representative mockup using anonymized sample data. The interaction patterns reflect the production flows; names, amounts, IDs, and dates are illustrative.

Conversation review
Reader of a registration guide · needs a call
3 messages · contact shared in chat
A relative says a registered will was forged. What can I do?
I am sorry you are dealing with this. In general, a will can be challenged on grounds such as forgery or undue influence, even when it is registered. Because this involves a dispute, I would like a will specialist to look at your documents. Could I take your name?
Grounded in sourcesNo pricingNo guaranteesNo personal advice
2 reviewed facts cited · 1 statute passage retrieved
Brief for the specialist
Suspected forged will, family dispute
Arrived from a paid social ad · read the challenge guide · gave a phone number
Next step: call today, explain the general position, then discuss a lawyer-led review.
Sources · guardrail results · attribution · audit trail

Inside the product

Screens captured from the working system with illustrative sample data. Names, contacts, companies and figures are placeholders.

Will guide page with a chat launcher and an ask a question panel
The assistant sits on a guide page and offers a page-aware nudge and starter questions.
Chat panel open on a guide page showing a hand-off to a specialist and contact details shared in conversation
Contact details are shared inside the conversation, name first, with no form.
Answer detail showing guardrail badges and cited reviewed facts
Under each answer, admins see the cited reviewed facts, retrieved passages and guardrail results.
Answer that was regenerated once after the verifier rejected an unsupported claim
When the verifier rejects a draft, the answer is rewritten once and the rejection stays visible.
Voice question in a mixed Hindi and English style answered with cited sources
Voice questions in English, Hindi or mixed speech are transcribed and answered in kind.
Visitor page journey with time on page and scroll depth next to the sales brief
The visitor page journey and attribution travel with the lead.
Insights dashboard with conversations, leads, cost per lead, latency and guardrail results
Insights: leads by channel and guide, intents, cost per lead, latency and guardrail outcomes.
Knowledge screen listing reviewed facts with legal basis, citation counts and review status
The reviewed knowledge pack, with legal basis, how often each fact is cited and its review status.
Sales system records created for captured leads with brief, channel and owner
Leads pushed to the sales system with the brief, questions asked and attribution.
Append-only audit log with a chain intact notice and recent events
An append-only, hash-chained audit log of every visit, message, escalation and lead.

Business outcomes

Answers with sources
Every answer can be traced to reviewed facts or statute passages.
No quoted prices
Pricing and guarantee language is handled by design and checked by tests.
Context-rich leads
Sales receives a brief, the guide the reader came from, the channel and the transcript.

Technical work demonstrated

The systems and controls behind the story above.

Source-priority retrieval

Statutes are chunked by section, guides are de-duplicated and stripped of prices, and passages are embedded into a vector store; results are re-ranked so reviewed facts beat statutes, statutes beat service pages, and service pages beat guides.

Single-call classify, answer and judge

Merging routing and answering into one call cut median latency substantially; the static prompt prefix is identical on every call so most input tokens are served from cache.

Code-computed verifier verdicts

The verifier checks faithfulness, price, guarantee and personal-advice rules with the whole reviewed set as cached context, and pass or fail is computed in code from its findings rather than from its own flag.

Deterministic contact capture

Name and number replies are handled by rules without a model call, including names given mid-sentence and replies in Hindi, with a policy for how many asks and how far apart.

Evaluation harness and live scores

More than a hundred scripted cases, including whole conversations, lead capture and an AI grader for legal accuracy, run through the real pipeline and fail on any price leak, guarantee or prompt leak; each live answer is also scored for grounding and reply rules.

Hash-chained audit and PII masking

Audit rows each include the previous hash, so edits or deletions are detectable, and phone numbers and emails are masked before traces leave the application.

Architecture

Answer pipeline

Retrieval over statutes, service pages and guides is re-ranked by source priority, then one model call classifies, answers and judges the hand-off moment, a verifier checks the draft, and a lead policy decides when to ask for contact details.

Conversation and lead records

Visitors, visits, page journeys, conversations, messages with sources and guardrail results, leads, sales-system records and the audit chain are stored in a managed document database with vector search for retrieval.

Console and tracing

The admin console reads the same records, and every turn is traced with retrieval, answer and verifier steps, token counts, cost and latency, so engineers and sales leaders see the same story.

Orchestration flow
Reader question
Answer and verify
Specialist hand-off
Briefed lead
Security · compliance · governance

Information, not advice

The assistant explains general law and services; personal situations, disputes and urgency go to a specialist.

No prices, no promises

Pricing is always handed to a specialist and guarantee language is blocked by the verifier.

Reviewed before trusted

Facts carry a citation and a reviewer field, and unreviewed facts are flagged in the console until a lawyer signs them off.

Everything is inspectable

Sources, guardrail outcomes, attribution, briefs and an append-only audit trail are available for every conversation.

Related buying paths

Where this proof applies

This case study supports the solution and service pages below because the same workflow, controls, and handoff model are visible in the implementation.

More case studies