Law firm website design: costs, musts, judging bids

A law firm website has one job that matters: turn someone with a legal problem into a conversation with a lawyer who can help them.

Most firm sites are built for a different audience — the partners who approve them. That's why so many open with a photograph of a courthouse and a paragraph about a century of excellence, and why so many bury the one thing the visitor came for, which is do you handle exactly my situation, and what happens if I call you.

This is what we've learned building and rebuilding legal sites: what the site has to do, what it costs, what to insist on in a proposal, and the specific mistakes that quietly cost firms enquiries.

What a law firm website actually has to do

Four jobs, in order of how much they matter:

1. Answer the visitor's actual question. Not "what does this firm do" but "does this firm handle my problem, in my jurisdiction, for someone in my situation." That answer belongs in the first screen of the relevant page, not three clicks in.
2. Establish that the firm is real and credible. Named lawyers with real photographs, actual results where the rules permit, offices, years, memberships. Legal buyers are risk-averse by definition — they are hiring insurance against a bad outcome.
3. Make contact frictionless for a qualified person. Short intake, obvious next step, no jargon, and a clear statement of what happens after they hit send.
4. Be findable when someone searches the problem rather than the firm. Which is a structural question about how practice areas are organised, and it's the one most firms get wrong.

Everything else — the video hero, the animated statistics, the scrolling logo wall — is decoration. Decoration is fine. It just isn't the job.

What a law firm website costs

The honest answer is that it depends on how much content and how much structure, not on how it looks. Design is rarely the expensive part.

Here's the range we see in the market, and what actually separates the bands:

Two things move a quote far more than firms expect.

Content volume. A firm with 400 blog posts and 60 lawyer bios is a migration project before it is a design project. Every one of those URLs has to be mapped, redirected and checked, or the firm loses rankings it spent years building. This is invisible on a proposal and it is often the single largest line of real work.

Practice-area depth. Ten practice areas written properly is ten pieces of persuasive, specific writing that a lawyer has to review. That's not a design cost, it's a content cost, and it's usually the thing that determines whether the site earns anything.

If you want a real number for a specific scope rather than a band, our website cost calculator prices it out directly.

The features that matter, and the ones that don't

Proposals are full of line items. Very few of them change outcomes.

The pattern: anything that helps a worried person confirm they're in the right place earns its budget. Anything that exists to look modern does not.

Intake is the product

The site's job ends at the moment someone submits an enquiry. What happens next determines whether any of it was worth paying for, and this is where most firms lose the work they've already paid to attract.

The specifics that make a measurable difference:

• Ask for less. Name, contact, one sentence about the situation. You do not need their address, their file number, or a dropdown of twelve practice areas. Every field is a place to abandon.
• Say what happens next. "We'll reply within one business day" converts better than silence, and it sets an expectation the firm then has to meet.
• Route it to a human who is expecting it. An intake that lands in a shared inbox nobody owns is worse than no intake, because the visitor believes they've been heard.
• Never make an urgent problem wait for a form. If someone is arrested, injured, served, or facing a deadline, the fastest route to a person should be visible without scrolling.
• Test it after every change. Forms break silently and constantly. We audit the whole chain rather than assume, because the failure mode is months of silence.

Response time beats design. A plain site with a two-hour reply will outperform a beautiful one with a three-day reply, every time, and no amount of budget changes that.

Practice-area pages are the SEO

Here's the structural mistake, and it's nearly universal.

A firm lists its services on one page — "Our Practice Areas" — with a paragraph each. That page targets nothing in particular, so it ranks for nothing in particular. Meanwhile the searches people actually run are specific: a description of a situation, usually with a place attached.

The fix is architectural. One page per practice area, per jurisdiction you actually serve. Each of those pages should:

• Name the problem in the words a non-lawyer would use, not the statutory term
• Say plainly who it's for and who it isn't
• Explain what the process looks like and roughly how long it takes
• Answer the questions the firm gets asked on every intake call
• Name the lawyers who do this work, with links to their profiles
• End with the same short intake

That last set of questions is the most valuable and least-used content a firm owns. Every practice group already knows the ten questions clients ask before they retain. Written down, those answers are exactly what people search — and increasingly, exactly what AI assistants quote when someone asks them a legal question. If your page is the one that answers it plainly, you're the one being cited.

The internal linking matters too: practice pages should link to related practice pages and to the lawyers who staff them, so the site reads as a connected body of expertise rather than a stack of unconnected leaves. We've written about how that structure works in more detail.

Being the page an AI assistant quotes

A growing share of legal research now starts in an assistant rather than a search box. Someone types a description of their situation into ChatGPT, Gemini or Google's AI results and gets a summary, sometimes with sources attached. Firms are not chosen from that summary directly — but they are shortlisted from it.

What gets quoted is not what ranks highest. It's what answers most cleanly. In practice that means:

• A direct answer immediately under a clear question. Assistants extract passages. A heading phrased as the question a person would ask, followed by two or three sentences that actually answer it, is far more quotable than the same information buried in the middle of a long paragraph about the firm's philosophy.
• Specifics that can be verified. Jurisdictions, timelines, process steps, what documents are needed. Vague reassurance summarises to nothing.
• Structure a machine can parse. Real headings, short paragraphs, tables for comparisons, lists for sequences. The same structure that helps a worried person skim helps a model extract.
• Plain language over statutory language. People describe situations, not causes of action. A page that uses both — the plain description first, the legal term second — matches the question and demonstrates the expertise.

The practical version of this for a firm: take the ten questions your intake staff answer every week, give each one a heading on the relevant practice-area page, and answer it in three sentences before elaborating. That single change makes a page more useful to a human in distress and more quotable to a machine at the same time, which is the rare case where the two audiences want exactly the same thing.

We've written up how to measure this channel properly, because the numbers in most dashboards are not what people think. It's worth being sober about the limits. This traffic is small today relative to search, it's harder to measure, and nobody can guarantee a citation. But the work it requires — clear questions, direct answers, verifiable specifics — is the same work that makes a practice-area page convert. It's not a separate optimisation, it's the same one, which is why it's worth doing before anyone can prove the return.

Compliance is a design constraint, not an afterthought

Legal advertising is regulated, and the rules differ by jurisdiction. In Canada, provincial law societies set them; in the US, state bars do; in England and Wales it's the SRA. Check your own regulator before publishing anything on this list, because the specifics genuinely vary:

• Claims about being the "best" or "top" firm
• Testimonials and client reviews
• Case results and settlement figures, and what disclaimer must accompany them
• Specialist or expert designations, which are often formally certified terms
• Comparative advertising against named competitors
• Fee advertising, and whether "no win, no fee" phrasing is permitted

The practical approach: decide these before design starts. Retrofitting a disclaimer into a results section designed without one is how firms end up with a page that satisfies neither the regulator nor the reader.

Rebuilding an existing site without losing rankings

If the firm already ranks for anything, a rebuild is a risk-management exercise before it's a design exercise.

The rebuild checklist we run:

1. Crawl the existing site first. Get the full URL inventory before anything is touched. A firm with a decade of publications usually has hundreds of pages nobody remembers.
2. Find what actually earns. Which pages have traffic, rankings and backlinks. Those are the ones that must survive with their content intact.
3. Map every old URL to a new one. One-to-one where possible. A redirect to the homepage is functionally a deletion.
4. Keep the content that ranks. Rewriting a page that already ranks well, purely because it doesn't match the new design, is the most common way firms lose traffic in a redesign.
5. Launch redirects with the site, not after. The gap is where rankings go.
6. Verify in the wild. Crawl the new site, check every redirect resolves, and watch Search Console for a month.

The single biggest avoidable loss in a legal rebuild is a redirect map nobody made. It is boring, it is unglamorous, and it is the difference between a redesign that grows the practice and one that quietly halves its enquiries.

How long it takes

The build is rarely the bottleneck. Approvals and content are.

A small firm site can ship in a month. A multi-practice rebuild with a decade of content is a quarter, and most of the elapsed time is waiting for busy people to approve words about themselves.

The single most effective thing a firm can do to shorten a website project is decide, before it starts, who has final sign-off on copy. Projects that stall almost always stall because three partners each have opinions and no one has authority. That is a governance decision, not a design one, and it costs more weeks than any technical choice on the project.

One practical trick that works: interview each practice lead for thirty minutes, draft from the transcript, and send them something to correct rather than something to write. Lawyers are fast, accurate editors and slow, reluctant authors. Asking for edits instead of drafts routinely takes weeks out of a schedule.

How to judge a proposal

You do not need to evaluate the technology. You need to evaluate whether the people writing the proposal understand the four jobs at the top of this article.

The last two matter most. A proposal that never says what isn't included is a change-order engine, and "we're always here" is not a maintenance agreement.

One more: ask who will actually do the work. Agencies that pitch with a senior team and deliver with a junior one are common enough that the question is fair, and the answer tells you what you're buying.

What we'd do first with a limited budget

Before spending anything, it is worth seeing what separates a firm site that books consultations from one that does not — law firm website design examples walks through nine patterns and scores which of them actually need a rebuild.

If a firm can only fund part of this, the order that produces the most return:

1. The three practice-area pages that represent most of your revenue. Written properly, with the client questions answered.
2. Working, short intake with a response-time promise the firm can keep.
3. Lawyer profiles with real photographs.
4. Mobile speed. Not a score — actual load behaviour on a phone on mobile data.
5. Everything else.

That order is deliberate. It concentrates spend on the pages that both rank and convert, and it defers everything that only makes the site look finished.

If you're weighing a rebuild against fixing what you have, tell us the URL — the redirect inventory alone usually settles the question, and it takes about a day to answer properly.

Keep reading

• Claude Code 2026: AI website development from the CLI
• OLVR ring scroll: Rolex 3D via Shopify custom development

Work with Rough Works

Need a partner for your next site? We're a Vancouver-based digital agency building websites and AI-enhanced experiences for brands across Canada and the United States. Start a project →