Chat qualification and booking for law firms

TaskChad sells the $250 Business Diagnostic Session and the $2,000 14-Day Implementation Sprint described on this page. This is provider-written implementation guidance from TaskChad's own product team, not independent research, a bar-association publication, or a customer case study. The inquiry-to-booking path below is a scoping hypothesis until a real law firm pays for a Session, accepts a scope, and TaskChad has terminal evidence for the result.

The expensive problem behind a law firm chat window with no conflicts gate

A chat bubble on a law firm's website exists to catch the visitor who won't call at 9 p.m. but will type. That part works. Missing is the gate that has to run before the conversation goes anywhere: a check for whether the person typing is adverse to an existing client, and a limit on how much of their story the chat collects before that check clears.

Most firms never wrote that gate down, so the chat behaves like a generic lead form with a personality, asking what happened, when, and where. Under ABA Model Rule 1.18, a person who discusses possible representation becomes a prospective client the moment that conversation starts. If that person is adverse to a current client and discloses something significantly harmful, the firm can be disqualified from continuing that representation. A chat collecting a full narrative before anyone checks who is talking is collecting exposure, not just leads.

The second failure runs the other direction: a visitor asks "do I have a case" or "what's this worth," and an unbounded chat answers from general knowledge about how similar matters resolve. That risks creating an unjustified expectation about results, the conduct ABA Model Rule 7.1 prohibits, and it is legal analysis a chatbot has no authority to perform. The event that matters is narrower than "make the chat smarter": did a visitor with a real matter end the conversation with a confirmed consultation, and did every fact collected before that point pass a conflicts screen first.

What "one source-grounded inquiry-to-booking path" means here

This lane builds exactly one thing: a single source-grounded inquiry-to-booking path scoped around one kind of chat conversation, not a rewrite of the firm's intake process. Source-grounded means informational replies come only from content the firm designated in advance — published practice-area pages, fee-range disclosures already public, office hours, service area — never from a model's general sense of how a case "usually" resolves. Inquiry-to-booking means the path has one acceptable ending: a confirmed consultation slot, an accepted callback window, or a live transfer to an attorney.

The realistic candidate list of chat conversations a law firm's site receives is short, and each already has a de facto owner today:

Chat scenario	Current owner today	Blocking exception
New-matter inquiry (accident, divorce, contract dispute)	Intake coordinator or whoever answers the widget	No conflicts screen before the chat collects a full narrative
Existing-client status question	Paralegal or case manager, once noticed	Shares a queue with new-matter chat
Case-merit or case-value question	Nobody — the chat often answers itself	Answer risks becoming legal advice
Message from someone adverse to a current client	Whoever's monitoring chat, if anyone	Ordinary intake risks disqualifying exposure
After-hours chat, no live monitor	Nobody until the next business day	No conflicts screen until the office opens

The Session scores these against the firm's actual chat volume and practice mix, then scopes the one path costing the firm the most exposure or business today.

Baseline and the KPI that decides whether this worked

Before any build starts, TaskChad writes down the baseline using evidence the firm can already produce, even a manually reviewed sample: how many chat conversations opened in a defined window, how many described a real matter, and how many ended with a confirmed next step before the Sprint began.

A conversation is qualified only when four conditions hold: the described matter falls inside a practice area the firm handles, its jurisdiction matches a state where an attorney is licensed, a conflicts screen against the adverse-party list clears or is not yet triggered, and a documented consent basis exists for follow-up contact. Conversations failing any test route to human review and do not count toward the KPI denominator.

The KPI for this lane is qualified conversations reaching a confirmed next step, measured as a rate over a stated window. It is a booking-completeness metric, not a case-outcome metric.

Signal	Source of truth
Chat session opened with a matter description	Chat platform transcript log
Practice area and jurisdiction match	Attorney roster, licensed-jurisdiction list
Conflicts screen result	Conflicts-check log or case-management module
Source content used for informational replies	Designated source-content library
Booking or handoff confirmed	Case-management calendar or disposition field

No percentage improvement gets published before that baseline is dated and written. A workflow that logged a transcript to the inbox is not the same as one that booked a qualified, conflict-cleared visitor.

Where the chat cannot go without a licensed attorney

Three roles carry standing approval authority: a scope owner who decides what gets built, a data owner who confirms which system is authoritative for source content and the adverse-party list, and an executive sponsor accountable for the outcome. A fourth role sits beside them: the managing attorney or a designated conflicts reviewer, who approves the source library and the conflicts-screen logic before either goes live.

That approval exists because both failure modes above are regulated. ABA Model Rule 1.18 protects information a prospective client discloses even when no representation follows, and can disqualify the firm from an existing matter if that information is significantly harmful to the discloser. ABA Model Rule 1.6 extends that duty to information relating to a representation "from whatever source." Together they are why this lane limits what the chat asks for before a conflicts screen clears.

The Florida Bar addressed intake chatbots directly in Ethics Opinion 24-1, issued January 19, 2024. It requires disclosure that a chatbot is not a lawyer or firm employee, warns that intake chatbots can create a prospective- or lawyer-client relationship without the lawyer's knowledge, and holds that a chatbot must not offer legal advice and must refer legal questions to a lawyer. This lane's Ground and Screen states are built around that guidance.

A firm's authority to take a matter is also geographic. ABA Model Rule 5.5 restricts practicing law where a lawyer is not admitted. A chat that books a consultation nobody at the firm can staff is not a scheduling convenience — it is why jurisdiction match is one of the four qualification conditions above.

The path from chat open to confirmed booking

State	What happens	Evidence required
Open	Source, channel, timestamp captured; bot identity and no-relationship notice disclosed	Logged event
Ground	Replies pull only from the source library; unsupported questions route to a booking offer	Reply linked to a source entry
Screen	Counterparty checked against the adverse-party list; narrative pauses until clear	Screen recorded clear, flagged, or escalated
Qualify	Practice area and jurisdiction checked against the roster	Eligible, not-eligible, or escalate flag
Consent	Contact information and consent basis recorded	Consent basis recorded
Offer	A specific consultation slot or callback window is selected	Slot logged with date and time
Confirm or escalate	Attorney accepts the booking, or the visitor is transferred live	Disposition recorded
Disposition	Terminal outcome compared to baseline	Baseline-to-outcome comparison

No state lets an AI system self-approve customer-facing content or clear its own conflicts screen. Screen and Qualify both clear before Offer runs. The working systems are the chat platform, the source-content library, the adverse-party list, the case-management system as booking record, and the licensed-jurisdiction roster.

Failure tests the path must survive before launch

A path is accepted because TaskChad tried to break it and watched it fail safely:

• Case-merit or case-value answer slips through. The chat must offer a booking, not synthesize an answer.
• Adverse-party message misrouted as ordinary intake. It must route to a conflicts hold, not the standard sequence.
• Narrative collected before the conflicts screen clears. Detail questions must gate behind a cleared screen.
• Jurisdiction or practice-area mismatch. The path halts to human review rather than booking a consult nobody can staff.
• Duplicate booking across channels. The same visitor opening chat twice must not create two calendar holds.
• Missing bot or relationship disclosure. The release is blocked until disclosure fires first.

Each test must produce a visible failure state, an untouched source record, and a named next action.

The 14-day Sprint scope for this law firm cell

Days	Phase	What happens
1–3	Preflight and baseline	Confirm the source library, practice areas, licensed jurisdictions, adverse-party list, and baseline counts
4–7	Build	Implement the one chosen path, using the systems in the agreed scope
8–11	Failure and approval tests	Run the tests above, plus disclosure and screening checks named during the Session
12–14	Release and handoff	Ship with a safe-disable switch, an operator runbook, the baseline receipt, and the KPI window

For this technical example, the working scope is one inquiry-to-booking path, at most two connected systems, one KPI, one owner, one release, one acceptance decision. A full case-management migration, a conflicts-database rebuild, round-the-clock live staffing, and any workflow letting AI evaluate case merit, quote a settlement value, or represent that an attorney-client relationship exists sit outside this technical example. When a real request exceeds that boundary, TaskChad narrows scope or declines rather than absorbing unpriced work into a fixed fee. The purchased Sprint is scoped to the agreed business result, which may address one big problem or several connected problems.

Fit conditions and wait conditions

This Session fits a firm that already runs, or is committed to running, a website chat widget, can point to published content usable as a source library, has a case-management system with a live calendar, can name which practice areas and jurisdictions the firm handles, and can describe even a lightweight adverse-party check it already runs somewhere in intake.

Waiting is right in a few cases. If the firm has no conflicts-screening process at all, even an informal one, that gap closes before Screen can run — building a conflicts process live during the Sprint is a different, larger scope. If nobody can approve chat scripts and source entries within a week, Ground and Screen have no owner. And if the actual request is for AI to evaluate whether a visitor has a case, estimate a settlement value, or confirm representation has begun, that sits outside every offer here; the Session names that boundary rather than delivering around it.

Terminal evidence: what "booked" is allowed to mean

A chat is not booked because a transcript exists. A contact form filled out mid-conversation is not booked. A promised callback with no calendar hold is not booked. The only terminal evidence is a case-management disposition of confirmed-consultation, retained, declined-conflict, declined-practice-area, or lost, tied to the originating chat session, inside the agreed window. A completed chat with contact information captured is a leading indicator, not a claim of value.

The three demonstrations and the Revenue Leak Score

TaskChad publishes three controlled demonstrations. The lead-to-booking demonstration shows the same capture, qualification, approval, and receipt sequence this path uses, from a different starting channel. The AI Workflow Audit demonstration shows how a candidate list like the five chat scenarios above gets scored for grounding risk and conflicts exposure before a Sprint is recommended. The SEO and GEO improvement loop demonstration is unrelated to this lane's build, but shows how TaskChad treats a measurement claim.

Before booking, a firm can run the Revenue Leak Score for law firms, a short directional diagnostic covering visibility, trust, capture, response, follow-up, and owner dependency. It is not a revenue forecast or a guarantee — a reasonable starting point for a managing attorney unsure whether an ungrounded or unscreened chat widget is the firm's biggest exposure.

Questions law firm intake owners ask before booking

Will the chat ever tell a visitor whether they have a case or what it's worth?

No. The chat can restate published practice-area information, confirm hours or service area, and offer a booking slot. It cannot evaluate a matter's merits or estimate a value. Under ABA Model Rule 7.1, a statement likely to create an unjustified expectation about results is a misleading communication about the firm's services, and that assessment stays with an attorney on the booked call.

What happens if the chat receives a message from someone adverse to an existing client?

That message routes to a conflicts hold in the Screen state, outside the standard qualification-to-booking sequence. ABA Model Rule 1.18 can disqualify a firm from an existing matter if a prospective-client conversation discloses information significantly harmful to that person, why narrative questions gate behind a cleared screen instead of collecting a full story from every visitor by default.

Does the chat have to disclose it isn't a lawyer, and are details typed into it confidential?

Yes to disclosure, by design. Florida Bar Ethics Opinion 24-1 requires a clear notice that a chatbot is not a lawyer or firm employee, plus a disclaimer that no attorney-client relationship exists until the firm confirms one. On confidentiality, ABA Model Rule 1.6 protects information relating to a representation "from whatever source" once received, also why this lane limits how much the chat asks for before a conflicts screen clears.

What happens if the chat books a consult in a practice area or state nobody at the firm can handle?

The path is built to prevent that outcome, not repair it after the fact. Qualify checks the described matter's practice area and jurisdiction against the firm's roster before Offer runs; a mismatch halts to human review instead of booking a consultation nobody can staff. ABA Model Rule 5.5 is why this check exists: taking a matter where no firm attorney is admitted risks the unauthorized practice of law, not just a scheduling error.

Sources

• ABA Model Rule 1.18, "Duties to Prospective Client" — defines a prospective client and the disqualification risk the conflicts-screen state avoids.
• Florida Bar Ethics Opinion 24-1, issued January 19, 2024 — bar guidance specific to intake chatbots: disclosure, no-legal-advice, and screening requirements this lane's Ground and Screen states follow.
• ABA Model Rule 7.1, "Communication Concerning a Lawyer's Services" — bars communications likely to create an unjustified expectation about results.
• ABA Model Rule 5.5, "Unauthorized Practice of Law; Multijurisdictional Practice of Law" — basis for the jurisdiction-mismatch failure test and the Qualify state.
• ABA Model Rule 1.6, "Confidentiality of Information" — protects information relating to a representation from whatever source.
• Federal Trade Commission, Advertising and Marketing — AI claims need substantiation before publication, why this page states a baseline and KPI instead of a promised result.

Book the Session for this exact cell

If available, bring one real chat scenario from the list above: a new-matter inquiry, an existing-client status question, a case-merit or value question, a message from someone adverse to an existing client, or after-hours chat with no live monitor. The $250 Business Diagnostic Session for this cell produces a written brief within two business days, covering the accepted inquiry-to-booking path, the baseline and KPI, the conflicts-screen and disclosure approval points, and one recommended Sprint. Paid Sessions are contacted within one business day to schedule; payment does not book a calendar slot automatically.

Book the $250 Business Diagnostic Session for law firms