How to Sell AI Builds to Law Firms (High Budgets, Narrow Doors)

Why Law Firms Are Worth Selling To

The economics are unusually favourable. A law firm converts time directly into revenue and knows precisely what an hour is worth, because it invoices in fractions of one. That means you never have to explain why saving time has value. You are talking to a buyer who already runs that calculation professionally, every single day, and who will do the maths on your proposal before you finish describing it.

The work itself is also unusually well suited to automation. A great deal of what happens in a small firm is repetitive, structured, and administrative: taking down the same intake details, chasing the same documents, scheduling around the same court calendars, writing the same client update, reconstructing what happened on a matter that has been quiet for a month. None of that is lawyering, all of it is expensive, and most of it is being done by somebody who is paid a great deal to do something else.

The Confidentiality Conversation, Handled Properly

Raise it first, before they do. A lawyer who has to ask you about confidentiality has already decided you had not thought about it. A builder who opens with here is where client information does and does not go has done something almost nobody else selling to them has bothered to do, and the meeting changes tone immediately.

Their duty of confidentiality is not a preference or a policy, it is a professional obligation with real consequences attached, and it applies to client information regardless of how carefully you handle it technically. That is why the winning move is not a better security argument. It is scoping the first build so that the question does not arise at all.

1. Design the first engagement to touch no client matter whatsoever. Marketing, intake before a matter is opened, internal scheduling, billing administration, internal process documentation. There is plenty of expensive work here.
2. Be able to state exactly where data goes and what is retained. Not reassuringly, specifically. Which service, which region, what is stored, for how long, and who could access it. Write it down before the meeting.
3. Ask them what their professional obligations require, and take notes. This is genuinely their expertise and not yours, and treating it that way is both correct and disarming.
4. Expect to sign a confidentiality agreement, and expect it to be theirs rather than yours. Read it, do not negotiate it line by line over a small first engagement, and do not act surprised by it.
5. Accept that the answer to some proposals will be no for reasons unrelated to the technology. Arguing with a professional obligation is a losing position, and gracefully accepting it is what gets you the next opportunity.

What Small Firms Actually Buy

Lead with intake. It is the highest-value, lowest-sensitivity problem in almost every small firm, and the money leaking out of it is visible to the person you are selling to.

The intake build is the one to lead with because the loss is so easy to quantify with them on the call. Ask what an average matter is worth to the firm, then ask how many enquiries come in outside office hours, then say nothing. They will do the arithmetic themselves and they will be more persuaded by their own number than by anything you could have told them.

Note where document assembly sits. It is the thing most builders want to sell first because it is the most technically interesting, and it is exactly the wrong place to start. It touches matter content, it invites the advice question, and it requires trust you have not built yet. Earn it on intake, then have the conversation.

How to Price It

Price against the firm's own hourly rates and against matter value, never against what software costs. This is a buyer who understands professional fees, expects to pay for expertise, and is suspicious of anything unusually cheap.

1. Get their numbers on the call. What a typical matter is worth, roughly what a fee earner's hour bills at, how many enquiries arrive in a week. Three questions, and every one of them makes your proposal easier to write.
2. Frame the build fee against recovered fee-earning time. Not this costs X, but this returns roughly N hours a month to people who bill at their rate. That comparison is one they run instinctively.
3. Always attach a monthly. Firms genuinely want somebody responsible when a system misbehaves, and they are far more comfortable with an ongoing relationship than with a one-off supplier who disappears.
4. Do not discount to win the first one. In this vertical a low price reads as inexperience rather than as value, which is the opposite of the effect it has in most markets.
5. Charge for a paid scoping engagement before quoting anything substantial. Their processes are more particular than they look from outside, and quoting blind is how a good engagement turns into a bad one. The general case is covered in [the audit offer](/blog/the-audit-offer-foot-in-the-door).

Getting the First Three Firms

Go narrow. One practice area, one geography, small independent firms rather than anything with a head office. Two to fifteen people is the sweet spot: large enough to feel the admin pain and have a budget, small enough that the person you are talking to can decide.

1. Pick one practice area and learn its specific rhythm. Family, conveyancing, immigration, personal injury and commercial work have completely different intake patterns and completely different bottlenecks. Naming theirs correctly is most of your credibility.
2. Test their own enquiry process before you contact them. Send a genuine enquiry through their website and time the response. That result is your entire opening line, and it is a fact rather than a claim.
3. Lead with the number, not the technology. I sent an enquiry through your site on Tuesday evening and heard back Thursday afternoon, here is what that is probably costing you.
4. Bring a working artifact. A short recording of the intake flow responding sensibly to an enquiry of the kind their firm receives. Built from public information, no access required.
5. Ask for twenty minutes, not a demo call. Lawyers are precise about time and a specific short request is far more likely to be accepted than an open-ended one.

Expect a slower cycle than any other vertical you have sold to. There will often be a partner meeting, a look at the professional obligations, and a delay you cannot influence. That is not disinterest and chasing it weekly will read badly. Set a follow-up rhythm you agreed with them on the call and hold to it, exactly as you would want a supplier to do with you.

The Failure Modes in This Vertical

Four mistakes account for most of the disappointing outcomes here, and all four are decided before any code gets written.

1. Starting with a build that touches client matter. It turns a purchasing decision into a compliance review, adds months, and frequently ends in no. Take the boring intake job first.
2. Positioning anything as producing advice. Even implying it moves the conversation to professional risk, where you have no standing and no answer.
3. Underpricing to get in the door. It signals inexperience to a buyer who associates low fees with low competence, and it anchors every future engagement with that firm.
4. Building something that requires them to change how they work. A system that needs a new habit from a fee earner who is already behind will be abandoned inside a month. Deliver into the tools already open on their screen.

The first one is the expensive mistake because it is the most tempting. Document work is where the impressive demos live and it is the thing builders are proudest of. Doing it second, for a firm that already trusts you, is worth several times doing it first for a firm that does not.