Every price on this page was read directly from the vendor's own pricing page on 30 August 2026. Two of the five publish a real number. The rest want a demo call first — and one of them would not load at all.
Why this page is not another "best intake chatbot" list
Search for an AI chatbot for your law firm and you will find a dozen listicles, most of them published by companies that sell one. They rank the same six or seven vendors, describe each as "powerful" and "intuitive," and almost never tell you what anything costs — because the vendors themselves mostly don't say.
That silence is the single most useful thing to know before you start booking demos, so it is what this page is organised around. Not which tool is best; which tools will tell you the price before they have your phone number, and what you are actually buying when they do.
The second useful thing: "AI intake chatbot" describes at least four different products, sold to different buyers, at prices that differ by a factor of twenty. Firms routinely sit through three demos before realising they were comparing a $99 website widget against a $2,100-a-month answering service. Sorting that out first will save you more money than any negotiation.
"Intake chatbot" describes four different products
1. The website widget
A chat box on your site that greets visitors, asks a few qualifying questions, and captures a name and number before the visitor leaves. It does not answer legal questions and it does not open a matter. It exists to stop anonymous traffic from bouncing. Intaker is the clearest example of a product built for only this job.
2. The conversational intake builder
A tool you configure yourself: a decision tree or LLM-driven conversation that replaces a long intake form, collects structured answers, and often generates a document — a retainer, an engagement letter, a questionnaire — at the end. LawDroid Builder and Gideon sit here. You are buying a build tool, which means the real cost includes the hours somebody at your firm spends building it.
3. The CRM with intake attached
A full client-acquisition system: intake forms feed a pipeline, which feeds automated email and text follow-up, e-signature, and attribution reporting that tells you which marketing spend produced which signed matter. Clio Grow and Lawmatics are the two most established. The chatbot is a feature here, not the product. If your problem is that leads go cold after first contact rather than that they never make contact, this is the category you actually want.
4. The answered-for-you service
Not software you operate. A service that answers your calls and web chats — with AI, with humans, or with both — screens the caller, and hands you a qualified lead or a booked appointment. Smith.ai is the best-documented example. You are buying capacity, not a tool, and it is priced accordingly.
Those four categories fail in different ways too. A widget cannot fix slow follow-up. A CRM cannot answer the phone at 9pm. Be specific about which failure is costing you matters before you look at a single price.
The pricing transparency table
| Vendor | Category | Entry price | Publishes a number? |
|---|---|---|---|
| Smith.ai | Answered-for-you service | $300/month (30 calls) | Yes |
| LawDroid | Conversational intake builder | $99 per user/month (Builder) | Yes |
| Lawmatics | CRM with intake | Not published — "Get a demo" | No |
| Intaker | Website widget | Not published — no figure on the pricing page | No |
| Clio Grow | CRM with intake | Not verified — see methodology | Could not confirm |
| Gideon | Conversational intake builder | Not verified — site did not respond | Could not confirm |
The two that publish real numbers
Smith.ai
Smith.ai publishes a full price list, with overage rates, on a public page. As of 30 August 2026:
| Plan | Calls included | Price | Effective cost per call | Overage |
|---|---|---|---|---|
| Starter | 30 | $300/month | $10.00 | $11.50/call |
| Basic | 90 | $810/month | $9.00 | $10.50/call |
| Pro | 300 | $2,100/month | $7.00 | $8.50/call |
| Enterprise | Custom | "Ask about enterprise pricing" | — | — |
Lead screening, qualification and intake are included at every tier rather than sold as an upgrade, and add-ons carry published prices too — appointment booking runs $1.50 per call.
The detail worth pausing on: Smith.ai offers an "AI-first" and a "human-first" approach, and both cost exactly the same. That is a straightforward answer to a question a lot of firms are quietly asking, which is whether AI answering should be cheaper than a person. At this vendor, today, it is not. You are buying answered calls; how they get answered is a service-design choice, not a discount.
Do the arithmetic before you assume this is expensive. Thirty calls a month at $300 is $10 a call. If one in fifteen of those calls becomes a signed matter, and your average matter is worth four figures, the maths is not close. If you take four calls a month, it is dreadful value and you should not be looking at this category at all.
LawDroid
LawDroid publishes per-seat pricing with no contract requirement:
| Plan | Price | What it is |
|---|---|---|
| Copilot | $25 per user/month | An AI assistant — research, document summarising, drafting. Not an intake product. |
| Builder | $99 per user/month, no contract | One chatbot, visual editor, templates, document automation, online payments |
| Ultra | $99 per user/month, annual contract | Builder on an annual commitment |
Builder is the intake-relevant plan. The $25 Copilot tier gets quoted in roundups as though it were an intake chatbot; it is a different product and will not run your intake.
What the $99 does not include is your time. A build-it-yourself chatbot is only as good as the conversation you design, and designing a good intake conversation — one that qualifies without interrogating, and knows when to stop and hand off to a human — is real work. Budget the hours honestly. A tool nobody finished configuring is the most common way firms waste money in this category.
The three that won't tell you
Lawmatics lists three tiers — Essential, Premium, Enterprise — and every one of them is a "Get a demo" or "Contact Sales" button. Its AI add-on modules are the same: Merlin Qualify and Merlin Engage are described as outcome-based pricing, Merlin Copilot as user-based pricing, all quote-only. The page says plainly that pricing is tailor-made and asks you to help them get to know you better for a custom quote.
Intaker's pricing page carries no dollar figure at all.
Clio Grow we could not verify. Clio's site returns a 403 to automated requests from every route we tried, so we could not read the price off the page ourselves and will not reproduce a third-party number as fact. Third-party sources describe Grow as quote-only and available as an add-on to a Clio subscription; treat that as unconfirmed. Our separate review of Clio Duo covers what the AI layer on Clio's practice-management side costs and whether it earns it.
Gideon did not respond on repeated attempts from two different clients on 30 August 2026. That may be temporary. We mention it because a vendor you are about to route client intake through is a vendor whose uptime is now your uptime, and it is a fair thing to ask about on the call.
What quote-based pricing actually costs you
The reason to care about this is not principle, it is leverage. In a market with published prices, the list price sets a ceiling and your negotiation is about the discount. In a quote-based market, the vendor sets the opening number after they have learned your firm size, your practice area, and how much pain you are in. You negotiate without an anchor.
This is the same pattern we found across contract review pricing, and the practical response is the same: get a written quote from at least two vendors in the same category before you take any call seriously, and ask what the number does at renewal.
The part the listicles skip: this is a professional-responsibility question
An intake chatbot is the first thing a prospective client talks to, and prospective clients are not nobody. Duties of confidentiality attach to prospective-client information in most jurisdictions well before anyone signs anything. That has three practical consequences.
- Conflicts. A bot that collects detailed facts from an adverse party before your conflicts check runs can create a problem you did not have. Ask where the conflict check sits in the flow — before detailed intake, or after.
- Disclosure. A prospective client should know they are talking to software. Under the EU AI Act's Article 50 transparency obligations this is now a legal requirement rather than a courtesy for tools in scope; see our guide to what applies after August 2026.
- Unauthorised practice and advice. A bot that answers "do I have a case?" has given legal advice. Most vendors handle this with scripted refusals, but you own the output, not the vendor.
Our guide to the state bar AI ethics opinions covers what the ABA and nine states have said, and the two-page AI policy template gives you something to hand a vendor when they ask what your requirements are.
Eight questions to ask on the demo call
- What is the list price, in writing, before we go further?
- What does it cost at renewal, and what is the cap on the increase?
- Where does the conflicts check sit in the intake flow?
- Is the prospective client told they are talking to AI, and in what words?
- Where is intake data stored, for how long, and is it used to train any model?
- What happens when the bot cannot answer — hand-off to a human, a callback, or a dead end?
- How many hours of configuration does a firm our size typically need before go-live?
- What is your uptime commitment, and what happens to inbound leads during an outage?
How to choose, honestly
Work out which failure is costing you matters, then buy for that one:
- Leads arrive but go cold. Your problem is follow-up, not capture. Look at the CRM category — Lawmatics or Clio Grow — and accept that you will have to sit through a quote process.
- Nobody answers after hours. Your problem is capacity. An answered-for-you service is the only category that genuinely fixes this, and Smith.ai will tell you the price today.
- Visitors leave without making contact. Your problem is the top of the funnel. A widget is the cheapest test, and the cheapest to abandon if it does not work.
- Intake takes too long and the forms are too painful. Your problem is the process. A conversational builder like LawDroid Builder at $99 per user per month is a low-risk way to find out whether a conversation converts better than your form — provided somebody actually builds it.
And a genuine option worth naming: do nothing yet. If you take a handful of enquiries a month, a same-day callback from a human beats every product on this page, costs nothing, and converts better. The category earns its keep on volume.
Frequently Asked Questions
How much does an AI intake chatbot for a law firm cost? It depends entirely on which of the four products you mean. A self-built conversational chatbot starts around $99 per user per month at LawDroid. An answered-for-you service starts at $300 a month for 30 calls at Smith.ai. CRM platforms with intake attached, such as Lawmatics and Clio Grow, do not publish prices and quote per firm. Anyone giving you a single number for "an intake chatbot" is describing one product and calling it a category.
Do AI intake chatbots cost less than a human answering service? Not necessarily, and at least one major vendor charges the same either way. Smith.ai prices its AI-first and human-first approaches identically — $300, $810 and $2,100 a month for 30, 90 and 300 calls. You are buying answered calls, not a specific technology, and the saving from automation is not being passed to the customer at that vendor today.
Can a chatbot give legal advice to a prospective client? It should not, and if it does, you own the consequences rather than the vendor. A bot that answers "do I have a case?" has given legal advice and may raise unauthorised-practice issues. Most vendors handle this with scripted refusals and escalation to a human. Confirm the exact wording before go-live rather than after.
Do we have to tell people they are talking to an AI? Increasingly yes as a matter of law, not just courtesy. The EU AI Act's Article 50 transparency obligations took effect on 2 August 2026 and require AI systems to disclose that a user is interacting with AI. Several US state bar opinions point the same direction on candour. Assume disclosure is required and write it into the first message.
Does an intake chatbot create a conflicts problem? It can. Duties to prospective clients generally attach before any engagement is signed, so a bot that collects detailed matter facts from someone adverse to an existing client can create a conflict you did not previously have. Ask every vendor where the conflicts check sits in the flow, and prefer designs that run it before detailed intake rather than after.
Methodology, and what we could not verify
Prices were read directly from each vendor's own pricing page on 30 August 2026. Where a vendor publishes no price, we have written "not published" rather than reproducing a third-party estimate as fact. Where we could not reach the vendor's page at all, we have said so and labelled the third-party figure unconfirmed rather than quietly passing it off as read.
Specifically: Clio's site returned HTTP 403 to automated requests on every route attempted, so Clio Grow's price is recorded as unverified rather than "not published" — those are different claims, and Clio may well publish it to a human browser. Gideon's site did not respond on repeated attempts from two clients on the same day.
Being explicit about the other limits of this page: we have not run these products against one another on live intake traffic, and no conversion figures here are ours. Nobody should claim a conversion ranking across intake tools without a controlled test, and we have not run one. What this page is good for is knowing what things cost, which category you are actually shopping in, and what to ask before you sign.