# How to Start a Legal Tech Startup Without VC (2026 Playbook)

> How to start a legal tech startup without VC: pick a law firm segment, avoid unauthorized practice of law, price per firm, and sell through Clio and MyCase.

- Canonical: https://zeroraised.com/blog/how-to-start-a-legal-tech-startup-without-vc
- Published: 2026-11-12T07:30:00Z
- Author: ZeroRaised Team (https://zeroraised.com/blog/authors/zeroraised)
- Publisher: ZeroRaised (https://zeroraised.com/)
- Section: Building
- Tags: legal-tech, bootstrapping, vertical-saas, compliance

**Short answer:** To start a legal tech startup without VC, build a narrow workflow tool that licensed lawyers pay for (intake, drafting, deadlines, billing or matter-specific automation) instead of a consumer product that gives legal advice. Price it the way small firms already buy software, per firm or per user, sell it through the practice management platforms lawyers already use, and treat client confidentiality and AI accuracy as product features from the first commit.

Every rule, price and figure below comes from a regulator, court, bar body or vendor page checked on 12 Nov 2026 and linked inline. The strategy around them is ours and is marked as such. Nothing here is legal advice.

## TL;DR

- **Sell to lawyers first.** Law firm software is a normal B2B SaaS problem with extra ethics. Consumer legal help is an access-to-justice opportunity with unauthorized practice of law (UPL) risk attached.
- **Firms already pay per user.** [MyCase](https://www.mycase.com/pricing/) lists $50, $100 and $130 per user per month billed annually; [PracticePanther](https://www.practicepanther.com/pricing/) lists $49 to $114 per user per month on annual billing; [Docketwise](https://www.docketwise.com/pricing/) lists $80 to $140 per user per month.
- **Confidentiality is a lawyer's duty you inherit.** ABA Model Rule 1.6(c) says "A lawyer shall make reasonable efforts to prevent the inadvertent or unauthorized disclosure of, or unauthorized access to, information relating to the representation of a client" ([ABA](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_1_6_confidentiality_of_information/)).
- **AI must be verifiable.** Mata v. Avianca ended with a $5,000 penalty for filing fake AI-generated cases ([opinion and order](https://www.nhd.uscourts.gov/sites/default/files/pdf/Mata-v-Avianca-sanctions-order.PDF)).
- **Distribution is integrations.** Clio publishes a listing process for its App Directory ([Clio developer docs](https://docs.developers.clio.com/handbook/launch-your-app/)), MyCase has a partner program, and Office add-ins run in Word on Windows, Mac, iPad and the web.
- **Regulatory sandboxes are not a shortcut right now.** Utah's sandbox is closed to new applications; Arizona licenses alternative business structures but requires an Arizona compliance lawyer.

**Disclosure:** ZeroRaised publishes this guide and is a launch site for bootstrapped products. ZeroRaised has no commercial relationship with any vendor, bar body or event named here and earns nothing if you buy from them. This is general information, not legal advice; talk to a lawyer licensed where you operate before you ship anything that touches client matters or consumer legal problems.

## Why legal tech looks like a VC game (and when it is not)

Most legal tech headlines are about large funding rounds for AI research assistants sold to big firms. That shape needs money: long enterprise security reviews, procurement committees, pilots that run for months and sales teams who can sit through them.

The other shape gets less press. There are thousands of small firms and solo lawyers who run their practice on a practice management system plus a mess of Word templates, spreadsheets and email. Their problems are narrow and expensive: intake that leaks leads, documents that take an hour to assemble, deadlines tracked in a calendar, invoices chased by hand. A solo lawyer can say yes to a $99 a month tool on a call. Nobody has to raise a round to sell that.

The general logic of picking a niche industry and owning it is in our [vertical SaaS without VC playbook](https://zeroraised.com/blog/how-to-build-a-vertical-saas-without-vc). Legal is a strong case for it, with three twists that this guide focuses on:

1. **Who you sell to changes your legal risk.** Selling to lawyers is mostly a contracts and security problem. Selling legal help to the public can be the practice of law.
2. **Your buyer has professional duties about data and technology.** Their ethics rules shape your product requirements.
3. **AI is both the biggest opportunity and the biggest liability.** Courts are sanctioning lawyers who file AI-invented citations, and your product will be blamed alongside them.

How big is the field? Stanford's CodeX TechIndex describes itself as "A curated dataset of 5,775 legal tech companies shaping the future of legal work" ([CodeX TechIndex](https://techindex.law.stanford.edu/)). Our read: that is crowded at the category level (practice management, e-discovery, contract review) and wide open at the level of "intake for small immigration firms in Texas" or "conflict checks for two-lawyer family law practices".

## Choose your segment

Legal tech buyers fall into four broad groups. Here is how we would rank them for a founder without investors.

| Segment | Who pays | Sales cycle | UPL risk | Our read for bootstrappers |
|---|---|---|---|---|
| Solo and small firms (1 to 20 lawyers) | Managing partner, often on a card | Days to weeks | Low (lawyer stays in the loop) | Best place to start |
| In-house legal teams | General counsel, via procurement | Months | Low | Good second market once you have security paperwork |
| Legal operations at mid and large firms | Ops, IT, knowledge management | Months to a year | Low | Hard without a sales team; sell via partners later |
| Consumers and small businesses (access to justice) | The person with the problem | Minutes, but low trust | High | Possible, but needs lawyer involvement and state-by-state care |

### Solo and small firms

This is where we would start. The buyer is usually the person doing the work, so the pain is personal. Small firms already pay for software per user (see the pricing section below), and they compare your price with billable time. If your tool saves one hour a month for a lawyer who bills $250 an hour, a $99 subscription is an easy decision. That rate is our illustration, not a survey figure; ask your interviewees what they bill.

Good problems in small firms tend to sit around the practice, not the legal reasoning: intake and conflict checks, document assembly for repeat work (wills, leases, demand letters, immigration forms), court deadline tracking, client communication, trust accounting helpers and collections.

### In-house legal teams

Corporate legal departments buy contract lifecycle tools, matter intake, outside counsel spend tracking and NDA automation. The money is better but the process is slower. Expect a security questionnaire before a pilot. Our [SOC 2 guide for bootstrapped startups](https://zeroraised.com/blog/soc-2-for-bootstrapped-startups) covers when you need a report and how to get there cheaply. Our read: do not start here unless you came from an in-house team and can sell to former colleagues.

### Legal operations at larger firms

Big firms have knowledge management and innovation teams that run pilots. They are a fine customer for year three, often through an integration partner. As a first market, the timelines will burn your savings.

### Consumers and access to justice

Self-help tools for tenants, debtors, small claims, wills or immigration have a real social case and huge demand. They are also where the unauthorized practice of law lives. We cover this next. Our read: if consumer legal help is your mission, partner with a legal aid organisation or a law firm and let the lawyer own the advice, or build tools for legal aid staff rather than for the public directly.

## Unauthorized practice of law: the rule that shapes your product

### United States

In the US, the practice of law is regulated state by state, mostly by state supreme courts and bars. The ABA's Model Rule 5.5 bars a lawyer from practicing law in a jurisdiction in violation of that jurisdiction's regulation of the legal profession, or assisting another in doing so ([ABA Model Rule 5.5](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_5_5_unauthorized_practice_of_law_multijurisdictional_practice_of_law/)). Two things follow for founders:

- **The definition is local.** What counts as "practicing law" depends on each state's rules and case law, so a consumer product that is acceptable in one state can be challenged in another.
- **Lawyers who work with you are exposed too.** A law firm that lets your tool give advice to its clients without proper supervision may be "assisting" unauthorized practice. Expect careful lawyers to ask about this.

Our rough risk ladder for software (our read, not a legal test):

| What the tool does | Risk |
|---|---|
| Helps a licensed lawyer draft, research, manage or bill | Lowest |
| Publishes general legal information (guides, explainers, checklists) | Low |
| Fills in official forms with a user's own answers, no recommendations | Medium, varies by state |
| Applies law to a person's specific facts and tells them what to do | Highest |
| Represents someone in court or negotiates for them | Do not, without lawyers |

If you are building anything in the last three rows, get an opinion from a lawyer in each state you will serve before launch, and keep a licensed lawyer in the workflow. Our read: designs that route a consumer to a lawyer who reviews and signs off are slower but far easier to defend.

### Regulatory sandboxes: Utah and Arizona

Two US states opened formal paths for non-traditional legal services. Neither is a quick win for a bootstrapped founder in late 2026, but both are worth understanding.

**Utah.** The Utah Supreme Court's Office of Legal Services Innovation describes the sandbox as "a pilot project and experiment launched by the Utah Supreme Court designed to test whether changing the way the practice of law is regulated in Utah will increase access to justice without increasing consumer harm" ([Utah Office of Legal Services Innovation](https://utahinnovationoffice.org/)). Its authorized-entities page says it "was launched by the Utah Supreme Court in 2020 and is a seven-year regulatory pilot project", and that "Types of providers authorized in the Sandbox include community justice advocates, other non-lawyer providers, and technology-based services such as AI" ([authorized entities](https://utahinnovationoffice.org/authorized-entities/)). The catch is on the homepage: "The Sandbox is closed to new applications", the Court closed the ABS-only portion at the end of 2024, and the sandbox "is set to sunset on August 14, 2027."

**Arizona.** Arizona licenses alternative business structures (ABS), defined as "a business entity that includes nonlawyers who have an economic interest or decision-making authority" ([Arizona Supreme Court](https://www.azcourts.gov/cld/Alternative-Business-Structure)). Its FAQ says the changes "came into effect January 1, 2021", that "Nonlawyers may own, have an economic interest in, manage, or make decisions in, an Alternative Business Structure that provides legal services", and that "At least one lawyer licensed to practice law in Arizona must be appointed by the ABS to serve as its compliance lawyer" ([Arizona ABS FAQ](https://www.azcourts.gov/accesstolegalservices/Questions-and-Answers/abs)). Fees are set out in ACJA 7-209(J), which we did not price here.

Our read: an ABS licence makes sense when you want to deliver legal services (with lawyers) and own the business as a non-lawyer. If you sell software to law firms, you usually do not need one.

### United Kingdom

England and Wales take a different approach. Most legal work is not reserved, but six activities are. Section 12(1) of the Legal Services Act 2007 defines "reserved legal activity" as the exercise of a right of audience, the conduct of litigation, reserved instrument activities, probate activities, notarial activities and the administration of oaths ([Legal Services Act 2007, s.12](https://www.legislation.gov.uk/ukpga/2007/29/section/12)). Under section 14, "It is an offence for a person to carry on an activity" that is a reserved legal activity "unless that person is entitled to carry on the relevant activity" ([s.14](https://www.legislation.gov.uk/ukpga/2007/29/section/14)).

That leaves a lot of room for unregulated legal software and services (general advice, document tools, many will-writing and contract tools) as long as you stay out of reserved activities. The Solicitors Regulation Authority is openly interested in the area: its Innovate resources say "The SRA welcomes queries from startups or lawtech providers needing support and guidance through our regulations" ([SRA Innovate](https://www.sra.org.uk/solicitors/resources/topic/innovate/)). Its consumer guidance notes that "Solicitor firms are no different with more and more of them offering their services through apps, chatbots or other online tools" ([SRA: using technology](https://www.sra.org.uk/consumers/instructing/using-technology/)), and points out that if you use a regulated solicitor "you can be sure that they are qualified". Our read: UK consumers get regulatory protections from regulated firms that they do not get from unregulated tools, so say clearly which one you are.

Scotland and Northern Ireland have their own rules. Check them separately.

## Confidentiality, competence and data: your buyer's duties become your spec

Lawyers do not just prefer secure software. Their professional rules require them to think about it, which means your product either helps them meet those duties or becomes a reason to say no.

### Rule 1.6(c): reasonable efforts

ABA Model Rule 1.6(c) reads: "A lawyer shall make reasonable efforts to prevent the inadvertent or unauthorized disclosure of, or unauthorized access to, information relating to the representation of a client" ([ABA Model Rule 1.6](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_1_6_confidentiality_of_information/)). Note the scope: "information relating to the representation", not only privileged material. Almost everything a lawyer puts into your tool qualifies.

What that means in product terms (our checklist):

- Encryption in transit and at rest, single sign-on or at least strong MFA, and role-based access inside a firm.
- A clear data processing agreement and a list of subprocessors, including any AI model provider.
- No training of models on customer data, stated in the contract and not just in a blog post.
- Data residency options if you sell outside the US. For the EU and UK, start with our [GDPR checklist for solo SaaS founders](https://zeroraised.com/blog/gdpr-checklist-for-solo-saas-founders).
- Export and deletion on request, because lawyers have file retention duties and clients change firms.

### Rule 1.1 comment 8: technology competence

The comment to Model Rule 1.1 on competence says lawyers should keep up with changes in practice, including "the benefits and risks associated with relevant technology" ([ABA Model Rule 1.1 comment](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_1_1_competence/comment_on_rule_1_1/)). Our read: this is why good lawyers ask hard questions in demos. Give them a one-page security and AI explainer they can put in their own file to show they did their homework. It shortens sales cycles.

### ABA Formal Opinion 512: generative AI

On 29 July 2024 the ABA issued Formal Opinion 512, its first formal ethics guidance on lawyers' use of generative AI tools ([ABA news release](https://www.americanbar.org/news/abanews/aba-news-archives/2024/07/aba-issues-first-ethics-guidance-ai-tools/)). In summary (paraphrased, not quoted):

- Lawyers need a reasonable understanding of the capabilities and limits of the specific AI tool they use, not expert knowledge.
- Before putting information relating to a representation into a self-learning AI tool, lawyers should get the client's informed consent, and boilerplate language in an engagement letter is not enough.
- Lawyers must review AI output, including citations, before relying on it.
- On fees, lawyers billing hourly can charge only for time actually spent, and generally should not bill clients for time learning a tool they will use across matters.

Product implications we would build in (our read): make it obvious whether your model "learns" from customer inputs (ideally it does not), give firms a plain-English description they can paste into client consent letters, and log which outputs were reviewed and by whom.

## AI in legal tech: design for verification

Generative AI is the reason many lawyers are suddenly open to new software. It is also the fastest way to embarrass your customers in front of a judge.

The leading example is Mata v. Avianca in the Southern District of New York. In the opinion and order on sanctions dated 22 June 2023, Judge P. Kevin Castel wrote: "Technological advances are commonplace and there is nothing inherently improper about using a reliable artificial intelligence tool for assistance. But existing rules impose a gatekeeping role on attorneys to ensure the accuracy of their filings." The court found the respondents had "submitted non-existent judicial opinions with fake quotes and citations created by the artificial intelligence tool ChatGPT", and ordered that "A penalty of $5,000 is jointly and severally imposed on Respondents" ([Mata v. Avianca opinion and order](https://www.nhd.uscourts.gov/sites/default/files/pdf/Mata-v-Avianca-sanctions-order.PDF)).

It was not a one-off. Damien Charlotin's AI Hallucination Cases database says "2288 cases identified so far", last updated 10 November 2026 ([AI Hallucination Cases](https://www.damiencharlotin.com/hallucinations/)). That is the author's own count, not an official court statistic, but it shows why lawyers are wary.

On the demand side, Clio says its 2026 Legal Trends Report found 85% of legal professionals report some level of AI adoption at their firms, up from 79% in 2025 ([Clio press release](https://www.clio.com/about/press/2026-legal-trends-report/)). Our read: the question for buyers is no longer "AI or not" but "which AI can I trust".

Design rules we would follow:

1. **Ground every legal claim.** Retrieve from a defined source (the firm's own documents, a statute set, a licensed case law database) and link every citation to the source text. If you cannot link it, do not show it.
2. **Never fabricate a citation format.** Post-check outputs for anything shaped like a case citation and verify it exists in your source before display.
3. **Show uncertainty.** Flag answers based on thin retrieval. Lawyers prefer "I could not find this" to a confident guess.
4. **Keep the lawyer as the author.** Draft, do not file. Require a human review step before export for anything going to a court.
5. **Contract for no-training.** Use model providers' business terms that exclude your customers' data from training, and say so in your DPA.

For implementation patterns (retrieval, evals, cost control) see our guide to [adding AI features to a SaaS](https://zeroraised.com/blog/how-to-add-ai-features-to-a-saas).

## Bootstrappable legal tech ideas

Here are product shapes that fit a founder without investors. These are our suggestions, not market data.

| Idea | Buyer | Why it is bootstrappable |
|---|---|---|
| Intake and conflict-check forms for one practice area | Small firms | Narrow, measurable (leads captured), sells on a demo |
| Document assembly for repeat work (estate plans, leases, demand letters) | Solo and small firms | Saves hours per matter; templates become your moat |
| Court deadline and docket calculators for one state | Litigators | Clear value; accuracy is the product |
| Immigration form prep add-ons | Immigration firms | Per-matter value; existing vendors show willingness to pay |
| Trust accounting and collections helpers | Small firms | Money problems sell; integrates with payments |
| NDA and vendor contract triage | In-house teams | Repetitive, low-risk, quick ROI |
| Tools for legal aid staff | Nonprofits and courts | Grant-funded buyers; mission plus revenue |
| Word add-in for clause libraries or citation checks | Any lawyer who drafts | Lives where lawyers work |

Open source can help you move fast. Docassemble, for example, describes itself as "A free, open-source expert system for guided interviews and document assembly, based on Python, YAML, and Markdown", with code licensed under MIT ([docassemble](https://docassemble.org/)). Our read: building a hosted, supported product on an open engine is a legitimate bootstrapped model, as long as you add the onboarding, templates and support that firms actually pay for.

Before building any of these, run the interview and pre-sale process in our [guide to validating a SaaS idea before building](https://zeroraised.com/blog/how-to-validate-a-saas-idea-before-building). In legal, ask to see the lawyer's current template folder and calendar. The mess tells you more than the interview.

## A bootstrapped legal tech example (and what it does and does not prove)

Primary-source stories of fully bootstrapped legal tech companies are rarer than funding announcements, so we will only use what founders have said on the record.

**BriefCatch.** In an August 2023 ABA Journal column by Ari Kaplan, BriefCatch founder Ross Guberman said: "I found bootstrapping appealing and maybe even romantic. I was lucky that the product was profitable early on." He described customers ranging "from individual practitioners to the world's biggest firms", and said that years later he faced "the Hamlet choice of coasting or going big and exploring funding options. I chose the latter to scale and see where it takes us", adding "We just raised $3.5 million" ([ABA Journal](https://www.abajournal.com/columns/article/3-legal-tech-founders-share-their-perspectives-on-raising-money-in-the-current-environment)).

Our read: BriefCatch is a useful model because it shows the order of operations that suits a bootstrapper. Build a writing tool lawyers pay for, reach profitability, and only then decide whether outside money buys anything you need. The same column features two founders who raised from the start, which is a reminder that both paths exist. If you need a cash bridge without giving up control, our list of [ways to fund a product without venture capital](https://zeroraised.com/blog/ways-to-fund-a-product-without-venture-capital) covers revenue-based financing, grants and pre-sales.

## Pricing models law firms understand

Lawyers think in hours, matters and users. Your pricing should map to at least one of them. Here is what real legal software vendors publish (checked 12 Nov 2026).

| Vendor | Category | Published price | Unit | Source |
|---|---|---|---|---|
| MyCase | Practice management | $50 / $100 / $130 billed annually ($60 / $120 / $150 monthly) | Per user per month | [MyCase pricing](https://www.mycase.com/pricing/) |
| PracticePanther | Practice management | $49 to $114 on annual billing; $59 to $124 monthly | Per user per month | [PracticePanther pricing](https://www.practicepanther.com/pricing/) |
| Docketwise | Immigration forms and case management | $80 to $140, depending on plan and billing period | Per user per month | [Docketwise pricing](https://www.docketwise.com/pricing/) |
| Gavel | Document automation | Lite $83, Standard $165, Pro $290; Scale from $417 annual only | Per plan per month | [Gavel pricing](https://www.gavel.io/pricing) |
| LawPay | Legal payments | $19 monthly; cards 2.99% + $0.30, eCheck 1% | Per account plus per transaction | [LawPay pricing](https://www.lawpay.com/pricing/) |
| Lawmatics | Intake and CRM | Custom quote; 3-user minimum on Essential and Premium | Per user, quoted | [Lawmatics pricing](https://www.lawmatics.com/pricing) |
| Smokeball | Practice management | Quote only | Quoted | [Smokeball pricing](https://www.smokeball.com/pricing) |

A few details worth noticing. MyCase offers a 10-day free trial with no credit card. PracticePanther offers 7 days and says annual billing saves "$120/user/year". Gavel lists a 7-day trial with no card on its lower plans, says annual plans get "2 months free", and charges extra builder seats on Pro and Scale. LawPay adds card brand pass-through fees that "may apply". Clio's pricing page refused our automated check today, so we have left its prices out.

### Which model to use

Our read on the options:

- **Flat per firm (small firm tier).** Easiest to sell to solos and two-lawyer firms, no seat counting. Cap it by users or matters to protect yourself from larger firms on a small plan.
- **Per user.** Matches practice management norms, scales with firm size. Works when every lawyer and paralegal uses the tool.
- **Per matter or per document.** Matches value for transactional work (each immigration case, each estate plan). Good as an add-on to a base fee.
- **Hybrid.** A base fee per firm including a few users, plus per-user or per-matter beyond that. This is where most bootstrapped legal tools land.

Avoid "free forever" for law firms. Lawyers are used to paying for tools and a free tier tends to attract the least committed users. A short trial, a paid pilot, or a money-back guarantee works better.

## Integrations as distribution

Lawyers rarely buy a standalone tool that does not talk to their practice management system. That is a constraint and a channel.

### Clio

Clio publishes how to get into its App Directory in its developer handbook. Building "starts with creating a developer account", and the docs note that "Clio does not provide a test/sandbox environment at this time" ([Clio: build your app](https://docs.developers.clio.com/handbook/build-your-app/)). To launch, Clio says it "will send you our security and compliance questionnaire"; after approval you send the API Partnerships team "a pre-recorded demo", they send a listing form and draft page, and then "the API Partnerships team will make it publicly available" ([Clio: launch your app](https://docs.developers.clio.com/handbook/launch-your-app/)).

Clio also runs a Certified App Program. The first requirement for Silver is to "Have at least 100 active Clio customer accounts", alongside at least 99.5% uptime and a one business day support turnaround, and "The Gold App tier is invite-only" ([Clio Certified App Program](https://docs.developers.clio.com/handbook/grow-your-app/clio-certified-app-program/)). Our read: aim for a plain directory listing in your first six months and treat Silver as a goal once you have traction.

### MyCase and PracticePanther

MyCase says "19,000+ law firms choose MyCase" ([MyCase](https://www.mycase.com/)). Its integrations page describes an API ("Connect MyCase to your existing internal or third party platforms") and lists 62 integrations, and its partner page invites developers to "Become an Integration Partner" ([MyCase integrations](https://www.mycase.com/integrations/), [MyCase partners](https://www.mycase.com/partners/)). PracticePanther's partnerships page says "Our KISS API makes it easy to integrate with us" and tells developers to "Reach out to our support team and request API access" ([PracticePanther partnerships](https://practicepanther.com/partnerships/)).

### Microsoft Word add-ins

Lawyers live in Word. Microsoft says Office Add-ins "run in Office on the web, Windows, Mac, and iPad" ([Microsoft Learn](https://learn.microsoft.com/en-us/office/dev/add-ins/overview/office-add-ins)). Publishing goes through Partner Center, where Microsoft says the review "can take 3-5 working days, depending on the volume of submissions in the queue" ([Microsoft Learn: submit](https://learn.microsoft.com/en-us/partner-center/marketplace-offers/submit-to-appsource-via-partner-center)). Our read: a Word add-in is often the fastest way to get a drafting tool in front of a lawyer without asking them to change systems.

### Payments

If your product touches billing, expect firms to ask about trust accounting and legal-specific processors like LawPay. Integrate rather than compete: payments are a regulated, trust-heavy business you do not need to own.

## Selling channels

### Direct outreach to small firms

The most reliable early channel is still one-to-one. Pick one practice area and one state, list 200 firms from the state bar directory and court records, and reach out with a specific problem you solve. Our full process is in [how to sell B2B SaaS as a solo founder](https://zeroraised.com/blog/how-to-sell-b2b-saas-as-a-solo-founder). In legal, add three things: a short security and AI explainer, a reference from a lawyer who uses it, and a clear statement of who owns the data.

### Bar associations and practice groups

State and local bars run sections, member benefit programs and CLE events. Getting in front of a practice section (immigration, family law, estate planning) is worth more than a general booth. Teaching a useful CLE session on the workflow you automate, without a hard pitch, is the classic way in. Requirements and costs vary by bar, so ask each one directly.

### Conferences

Two big US events in early 2027:

- **ABA TECHSHOW 2027:** "March 3-6, 2027 | McCormick Place, Chicago, IL" ([ABA TECHSHOW](https://www.techshow.com/)). Registration prices were not on the homepage when we checked. TECHSHOW has run a Startup Alley competition for early companies in past years.
- **Legalweek New York 2027:** "March 1 - 3, 2027 | North Javits Center | New York, NY" ([Legalweek](https://www.event.law.com/legalweek/)). Its 2027 pricing page lists passes from $245 for expo access and from $2,295 for a full conference pass, with prices rising closer to the event, and a $4,995 non-exhibiting vendor pass ([Legalweek 2027 pricing](https://www.event.law.com/legalweek/2027-pricing)).

Note that last line. Our read: Legalweek is built for large firms and corporate legal teams, and vendors who walk the floor selling without exhibiting are charged accordingly. For a bootstrapper selling to small firms, a state bar conference or a practice-area event is usually better value than either national show.

### Directories and launch sites

Get listed in the CodeX TechIndex (Stanford's footer notes its info "may be out of date", so check your entry), in your integration partners' directories, and on launch sites. Our [product launch checklist for bootstrapped makers](https://zeroraised.com/blog/product-launch-checklist-for-bootstrapped-makers) covers the general launch sequence; for legal, line up two or three lawyer testimonials first, because lawyers trust other lawyers more than they trust upvotes.

## Worked example: an intake and conflict-check tool for small immigration firms

This is an illustrative model with our own numbers. It is not data from a real company.

**Product:** a web intake form and conflict check for immigration firms, syncing new contacts and matters into Clio or MyCase.

**Pricing (ours):**

- Solo plan: $99 per firm per month, up to 2 users.
- Firm plan: $199 per firm per month, up to 6 users, plus $25 per extra user.
- Annual plans: two months free.

**Assumptions for month 12 (ours):**

| Item | Number |
|---|---|
| Solo plan firms | 45 |
| Firm plan firms | 20 |
| Extra users across Firm plans | 15 |
| Share paying annually | 30% |

**Revenue arithmetic:**

- Solo: 45 × $99 = $4,455 per month
- Firm: 20 × $199 = $3,980 per month
- Extra users: 15 × $25 = $375 per month
- List MRR: $4,455 + $3,980 + $375 = **$8,810**

Annual customers pay 10 months for 12, so their effective monthly price is about 83% of list. If 30% of revenue is annual:

- Effective MRR ≈ $8,810 × (0.70 + 0.30 × 10/12) = $8,810 × 0.95 = **about $8,370**

**Monthly costs (ours, illustrative):**

| Cost | Per month |
|---|---|
| Hosting, database, backups | $150 |
| Email, auth, monitoring | $100 |
| AI API usage (if used for intake summaries) | $150 |
| Insurance (cyber and tech E&O, amortised) | $250 |
| Lawyer advisor retainer | $500 |
| Tools, domain, misc | $100 |
| **Total** | **$1,250** |

That leaves roughly $7,100 a month before your own salary and tax. It needs 65 paying firms. If you can close 6 firms a month after a three-month build and pilot phase, you reach that in about 11 months of selling. Change any of these numbers and the timeline moves; the point is that a narrow legal tool can pay a founder without a funding round if the sales motion is steady.

What would break this model: building for every practice area, a free tier that never converts, an AI feature that produces one bad citation in a filing, or a security incident. Those are the risks to manage, more than competition.

## Decision table

| Your situation | Our recommendation |
|---|---|
| You are a lawyer with a workflow pain from your own practice | Build that tool for firms like yours; you are the first user and the first reference |
| You are a developer with no legal background | Partner with a practising lawyer (equity or paid advisor) before writing code |
| Your idea gives legal answers to consumers | Redesign so a lawyer reviews advice, or sell to legal aid staff; get state-by-state UPL advice |
| Your idea needs non-lawyer ownership of a law practice | Look at Arizona ABS; Utah's sandbox is closed to new applications |
| You are in England and Wales | Stay outside reserved legal activities, or partner with an SRA-regulated firm |
| Your product uses generative AI on client matters | Ground outputs in sources, no training on customer data, human review before filing |
| Your buyer is an in-house legal team | Prepare a security pack first; consider SOC 2 once deals depend on it |
| You need distribution fast | Build the Clio or MyCase integration before your second feature |
| You want to raise later | Get to profit first; BriefCatch raised after years of profitable bootstrapping |

## A 90-day plan

### Days 1 to 30: segment and evidence

1. Choose one practice area, one firm size and one jurisdiction.
2. Interview 15 to 20 lawyers and paralegals. Ask to see their templates, intake forms and calendars.
3. Write down the one workflow that costs the most hours or loses the most money.
4. If your idea touches consumers, book an hour with a lawyer on UPL in your target state, or on reserved activities in England and Wales.
5. Ask 5 firms for a paid pilot or a pre-payment. If nobody pays, change the problem, not the pitch.

### Days 31 to 60: build the narrow version

1. Build the smallest tool that removes that workflow pain for one firm type.
2. Set up the basics lawyers will ask about: MFA, encryption, audit log, export and deletion, and a DPA. Have a lawyer review your terms of service.
3. If you use AI, implement source-linked outputs and a review step from the first version.
4. Apply for a developer account with your main integration partner and start the integration.
5. Onboard pilot firms yourself, on calls, and watch them use it.

### Days 61 to 90: sell and integrate

1. Convert pilots to paid plans and ask each happy firm for a written testimonial and an intro.
2. Request Clio's security questionnaire or contact MyCase or PracticePanther about partnership.
3. Write one practical guide for your practice area and offer it to a bar section or practice group.
4. Start outbound to 200 firms in your segment with a short, specific message.
5. Launch publicly once you have 3 testimonials, and track which channel each paying firm came from.

## Common mistakes

- **Building for "lawyers" in general.** A family lawyer and an M&A partner have nothing in common except the bar exam.
- **Consumer advice without a lawyer in the loop.** It is the fastest way to a cease-and-desist from a bar regulator.
- **AI demos that work on stage and invent citations in production.** Test with adversarial questions and real filings.
- **Ignoring the integration.** If your tool does not sync with their practice management system, it becomes extra admin.
- **Underpricing.** Lawyers equate low prices with risk. Price against hours saved, not against consumer apps.
- **Vague data terms.** "We take security seriously" is not an answer to a lawyer's question about who can read client files.

## Frequently asked questions

### Can you start a legal tech startup without VC funding?

Yes, if you build a narrow tool that law firms pay for from month one. Small and solo firms already pay per user for practice management software (MyCase lists $50 to $130 per user per month billed annually), so a focused tool for intake, document drafting, deadlines or billing can be funded by subscriptions. BriefCatch founder Ross Guberman told the ABA Journal in 2023 that he found bootstrapping appealing and that his product was profitable early on. What usually needs outside money is consumer legal services, which run into unauthorized practice rules, and large-firm enterprise sales.

### Do I need to be a lawyer to start a legal tech company?

No, but you need lawyers close to the product. Software that helps licensed lawyers do their work can be built by anyone. Software that gives legal advice to the public can be treated as the unauthorized practice of law in many US states, and in England and Wales certain reserved legal activities, such as conducting litigation, can only be carried out by authorised persons under the Legal Services Act 2007. A lawyer co-founder or a paid advisory relationship with one is the practical answer. This is not legal advice.

### What is the unauthorized practice of law risk for legal tech?

In the US, each state defines and polices the practice of law, and ABA Model Rule 5.5 bars lawyers from practicing in violation of a jurisdiction's rules or helping others do so. Consumer tools that apply law to a person's specific facts and tell them what to do carry the most risk. Tools sold to lawyers, general legal information, and form-filling under a lawyer's supervision carry less. Utah's regulatory sandbox allowed some technology-based services, but its own site says it is closed to new applications and set to sunset on 14 August 2027. Get advice from a lawyer in each state you target. This is not legal advice.

### How should I price legal tech software for law firms?

Match how firms already buy. Practice management tools charge per user per month (MyCase $50 to $130 billed annually, PracticePanther $49 to $114 on annual billing, Docketwise $80 to $140), while document automation vendor Gavel lists per-plan pricing from $83 a month. Our read: for a bootstrapped tool, a flat per-firm price for a small firm tier plus a per-user price for larger firms is easiest to sell, and per-matter pricing works when value maps to cases, such as immigration filings.

### Is it safe to build AI legal tech after the fake citation cases?

It is safe if you design for verification. In Mata v. Avianca the court said there is nothing inherently improper about using a reliable AI tool, but that existing rules give attorneys a gatekeeping role over the accuracy of their filings. ABA Formal Opinion 512 says lawyers must review generative AI output and protect client information. Build features that link every citation to a real source, show uncertainty, keep client data out of model training, and leave the final judgment to the lawyer.

## Launch your legal tech product without investors

ZeroRaised is a launch directory for products built without institutional VC. When your first firms are paying and you have a couple of lawyer testimonials, [submit your product](https://zeroraised.com/submit) for free and put it in front of other independent makers. For more playbooks on building in regulated niches without investors, [join the ZeroRaised newsletter](https://zeroraised.com/newsletter).

## Sources

- [ABA Model Rule 5.5: Unauthorized Practice of Law](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_5_5_unauthorized_practice_of_law_multijurisdictional_practice_of_law/), last checked 12 Nov 2026
- [ABA Model Rule 1.6: Confidentiality of Information](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_1_6_confidentiality_of_information/), last checked 12 Nov 2026
- [ABA Model Rule 1.1: Comment](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_1_1_competence/comment_on_rule_1_1/), last checked 12 Nov 2026
- [ABA news release on Formal Opinion 512](https://www.americanbar.org/news/abanews/aba-news-archives/2024/07/aba-issues-first-ethics-guidance-ai-tools/), last checked 12 Nov 2026
- [Mata v. Avianca, Opinion and Order on Sanctions (S.D.N.Y., 22 June 2023)](https://www.nhd.uscourts.gov/sites/default/files/pdf/Mata-v-Avianca-sanctions-order.PDF), last checked 12 Nov 2026
- [AI Hallucination Cases database (Damien Charlotin)](https://www.damiencharlotin.com/hallucinations/), last checked 12 Nov 2026
- [Legal Services Act 2007, section 12](https://www.legislation.gov.uk/ukpga/2007/29/section/12), last checked 12 Nov 2026
- [Legal Services Act 2007, section 14](https://www.legislation.gov.uk/ukpga/2007/29/section/14), last checked 12 Nov 2026
- [SRA Innovate resources](https://www.sra.org.uk/solicitors/resources/topic/innovate/), last checked 12 Nov 2026
- [SRA: Using technology for legal services](https://www.sra.org.uk/consumers/instructing/using-technology/), last checked 12 Nov 2026
- [Utah Office of Legal Services Innovation](https://utahinnovationoffice.org/), last checked 12 Nov 2026
- [Utah sandbox authorized entities](https://utahinnovationoffice.org/authorized-entities/), last checked 12 Nov 2026
- [Arizona Supreme Court: Alternative Business Structures](https://www.azcourts.gov/cld/Alternative-Business-Structure), last checked 12 Nov 2026
- [Arizona ABS questions and answers](https://www.azcourts.gov/accesstolegalservices/Questions-and-Answers/abs), last checked 12 Nov 2026
- [Stanford CodeX TechIndex](https://techindex.law.stanford.edu/), last checked 12 Nov 2026
- [Clio 2026 Legal Trends Report press release](https://www.clio.com/about/press/2026-legal-trends-report/), last checked 12 Nov 2026
- [Clio developer handbook: build your app](https://docs.developers.clio.com/handbook/build-your-app/), last checked 12 Nov 2026
- [Clio developer handbook: launch your app](https://docs.developers.clio.com/handbook/launch-your-app/), last checked 12 Nov 2026
- [Clio Certified App Program](https://docs.developers.clio.com/handbook/grow-your-app/clio-certified-app-program/), last checked 12 Nov 2026
- [MyCase pricing](https://www.mycase.com/pricing/), [MyCase homepage](https://www.mycase.com/), [MyCase integrations](https://www.mycase.com/integrations/) and [MyCase partners](https://www.mycase.com/partners/), last checked 12 Nov 2026
- [PracticePanther pricing](https://www.practicepanther.com/pricing/) and [PracticePanther partnerships](https://practicepanther.com/partnerships/), last checked 12 Nov 2026
- [Docketwise pricing](https://www.docketwise.com/pricing/), last checked 12 Nov 2026
- [Gavel pricing](https://www.gavel.io/pricing), last checked 12 Nov 2026
- [LawPay pricing](https://www.lawpay.com/pricing/), last checked 12 Nov 2026
- [Lawmatics pricing](https://www.lawmatics.com/pricing), last checked 12 Nov 2026
- [Smokeball pricing](https://www.smokeball.com/pricing), last checked 12 Nov 2026
- [docassemble](https://docassemble.org/), last checked 12 Nov 2026
- [ABA Journal: 3 legal tech founders share their perspectives on raising money (Ari Kaplan, 28 Aug 2023)](https://www.abajournal.com/columns/article/3-legal-tech-founders-share-their-perspectives-on-raising-money-in-the-current-environment), last checked 12 Nov 2026
- [Microsoft Learn: Office Add-ins platform overview](https://learn.microsoft.com/en-us/office/dev/add-ins/overview/office-add-ins), last checked 12 Nov 2026
- [Microsoft Learn: submit an Office add-in via Partner Center](https://learn.microsoft.com/en-us/partner-center/marketplace-offers/submit-to-appsource-via-partner-center), last checked 12 Nov 2026
- [ABA TECHSHOW](https://www.techshow.com/), last checked 12 Nov 2026
- [Legalweek New York](https://www.event.law.com/legalweek/) and [Legalweek 2027 pricing](https://www.event.law.com/legalweek/2027-pricing), last checked 12 Nov 2026

## Frequently asked questions

### Can you start a legal tech startup without VC funding?

Yes, if you build a narrow tool that law firms pay for from month one. Small and solo firms already pay per user for practice management software (MyCase lists $50 to $130 per user per month billed annually), so a focused tool for intake, document drafting, deadlines or billing can be funded by subscriptions. BriefCatch founder Ross Guberman told the ABA Journal in 2023 that he found bootstrapping appealing and that his product was profitable early on. What usually needs outside money is consumer legal services, which run into unauthorized practice rules, and large-firm enterprise sales.

### Do I need to be a lawyer to start a legal tech company?

No, but you need lawyers close to the product. Software that helps licensed lawyers do their work can be built by anyone. Software that gives legal advice to the public can be treated as the unauthorized practice of law in many US states, and in England and Wales certain reserved legal activities, such as conducting litigation, can only be carried out by authorised persons under the Legal Services Act 2007. A lawyer co-founder or a paid advisory relationship with one is the practical answer. This is not legal advice.

### What is the unauthorized practice of law risk for legal tech?

In the US, each state defines and polices the practice of law, and ABA Model Rule 5.5 bars lawyers from practicing in violation of a jurisdiction's rules or helping others do so. Consumer tools that apply law to a person's specific facts and tell them what to do carry the most risk. Tools sold to lawyers, general legal information, and form-filling under a lawyer's supervision carry less. Utah's regulatory sandbox allowed some technology-based services, but its own site says it is closed to new applications and set to sunset on 14 August 2027. Get advice from a lawyer in each state you target. This is not legal advice.

### How should I price legal tech software for law firms?

Match how firms already buy. Practice management tools charge per user per month (MyCase $50 to $130 billed annually, PracticePanther $49 to $114 on annual billing, Docketwise $80 to $140), while document automation vendor Gavel lists per-plan pricing from $83 a month. Our read: for a bootstrapped tool, a flat per-firm price for a small firm tier plus a per-user price for larger firms is easiest to sell, and per-matter pricing works when value maps to cases, such as immigration filings.

### Is it safe to build AI legal tech after the fake citation cases?

It is safe if you design for verification. In Mata v. Avianca the court said there is nothing inherently improper about using a reliable AI tool, but that existing rules give attorneys a gatekeeping role over the accuracy of their filings. ABA Formal Opinion 512 says lawyers must review generative AI output and protect client information. Build features that link every citation to a real source, show uncertainty, keep client data out of model training, and leave the final judgment to the lawyer.