A family law firm in Atlanta spent eighteen months investing in content. They added forty resource pages. They published a weekly blog. They ranked on the first page for "atlanta divorce lawyer," "child custody attorney atlanta," and "georgia divorce process." Their organic traffic went from 2,200 monthly visits to 4,600 monthly visits over the period. A real win.
Their consultation form submissions over the same eighteen months: 31 per month at the start, 32 per month at the end. Their phone call volume from the website was 12 per month at the start, 11 per month at the end. They had doubled the size of the funnel and produced no additional leads.
The traffic was real. The visitors were qualified (we checked: 78% from Atlanta metro, 92% on legal-intent queries). The problem was downstream of traffic. Their site was sending every new visitor back to Google to pick a different firm.
We audited the site and found five specific design choices, each individually small, that combined to gut conversion. None of them were the kind of thing you'd flag in a quick visual review. All five together explained why doubled traffic produced flat leads.
Five small choices, one flat funnel
A gavel-on-books stock shot reads as a category, not a firm. Swap in the real office and the real attorney. A $300 to $600 shoot covers a homepage and a few service pages.
"Compassion. Experience. Results." fits every firm in the state. Take a position instead, and name the actual practice and approach. Some visitors bounce; the rest convert better.
About a third of law firm conversions start with a call. In the footer, that path is buried. Put the number in the nav, sticky on mobile, tap to call.
Four badges that read as one block on desktop stack into three scrolls of isolated claims on the phone, where most visitors actually are. A 2x2 grid keeps them cohesive.
"Dissolution of marriage" is what the court calls it. "Divorce" is what the client typed at 11pm. Write the pages in the words the audience already uses. Engines match those words too.
Mistake one: stock photography that signals "interchangeable"
The Atlanta firm's homepage led with a stock photo of a gavel resting on a leather-bound law book, lit from above with warm golden lighting. It was on the same stock photo library that probably four hundred other law firm sites were using.
The visitor's brain processes this in roughly 200 milliseconds. The pattern recognition runs: "law firm website," "looks like the others," "lacks distinctive identity." Trust signal: low. The visitor scrolls or leaves, almost involuntarily.
This isn't subtle. It's the strongest known visual trust signal on law firm sites, and most firms fail it. Stock photography says "we are not the firm; we are a category." Real photography (the actual office, the actual attorneys, the actual conference room) says "we are a specific firm."
The fix isn't expensive. A two-hour professional photo shoot, $300 to $600 depending on the market, produces enough imagery for a homepage and three or four service pages. The Atlanta firm replaced the gavel photo with a wide shot of the senior partner sitting in the firm's actual conference room with documents on the table. Bounce rate on the homepage dropped from 71% to 54% in the first month. Nothing else changed on the page.
The five homepage elements that build trust piece covers the trust-signal hierarchy in detail. Photography is element one.
Mistake two: headlines that could apply to any firm
The Atlanta firm's H1 read "Atlanta Family Law Attorneys. Compassion. Experience. Results." That headline could apply to every family law firm in Georgia. It wasn't false. It wasn't bad. It was simply interchangeable.
Interchangeable headlines don't differentiate. They normalize. The visitor reads them and assumes the firm is, well, normal. They keep looking for something distinguishing. Usually they find it on a competitor's site that opened with a specific position instead of a generic credential.
Compare two H1s for the same firm.
Generic: "Atlanta Family Law Attorneys. Compassion. Experience. Results."
Specific: "Atlanta divorce, custody, and post-decree help, without the courthouse drama in most cases."
The second one takes a position. It implies a specific approach (settlement over litigation), names the actual practice areas, and uses a tone that distinguishes the firm from corporate-feeling competitors. Some visitors will read that and bounce because they specifically want a litigator. That's fine. The remaining visitors will be more qualified and substantially more likely to convert.
Headlines are the highest-impact hundred characters on the site. Most firms spend more time on Slack channel names than on their homepage H1. Worth fixing first. For the brand-positioning logic underneath this, the E-E-A-T guide walks through how specific positioning compounds into trust signals AI engines also recognize.
Mistake three: phone number buried below the fold
The Atlanta firm's phone number lived in the footer. Visitors had to scroll to find it on desktop. On mobile, they had to tap the hamburger menu, look at the footer link, and scroll there.
This is a small UX choice with a measurable conversion impact. Roughly 35% of law firm consultations start with a phone call, not a form submission. That ratio is higher for personal injury and criminal defense (closer to 55%) and lower for estate planning and business law (closer to 20%). Family law sits in the middle.
For 35% of the firm's potential conversions, the phone number's location is the conversion path. A buried phone number is a buried conversion path.
The fix: phone number in the primary nav as a tap-to-call link, sticky on mobile, visible from any page in any scroll position. The Atlanta firm shipped this on a Tuesday. Phone call volume from the site went from 11 per month to 24 per month in the following month. Nothing else changed.
The behavioral case for why this works is in the psychology of law firm CTAs piece: friction kills calls more than poor headline copy does.
Mistake four: trust signals that collapse on mobile
The Atlanta firm had four real trust signals on the desktop homepage: a Martindale-Hubbell "AV Preeminent" badge, a Super Lawyers logo, a Georgia State Bar membership badge, and a row of three testimonial cards with real client photos. All four lived in a horizontal strip below the hero.
On desktop this strip read as one cohesive trust block. The visitor saw all four signals in a single glance.
On mobile the strip stacked into a vertical column that took three full screen heights to scroll through. The visitor saw one signal at a time, isolated, often surrounded by whitespace. The cumulative trust effect that worked on desktop disappeared on mobile. Each badge read as a small isolated claim rather than as part of a credible whole.
This is the most under-diagnosed conversion problem on law firm sites. 64% of legal-marketing decision-maker traffic on competitor sites is mobile (we pulled this from a sample of seven sites we audit). The trust hierarchy that worked when the desktop comp was approved gets quietly destroyed on the device most visitors actually use.
The fix is design-specific. Trust signals need to read as a coherent block on mobile too, which usually means consolidating them into a horizontal scrolling carousel, or a 2x2 grid, or a single combined "credentials" card with all four logos in one frame. The Atlanta firm went with the 2x2 grid. Mobile conversion rate on the homepage climbed from 0.8% to 1.6% in the four weeks after.
Mistake five: lawyer-speak in the body copy
This one is harder to fix because it requires the firm to break a habit.
The Atlanta firm's practice area pages opened with sentences like "Our firm's practice encompasses the representation of clients in domestic relations matters, including dissolution of marriage, post-decree modifications, and matters of parental rights and responsibilities." That sentence is grammatically correct, legally precise, and conversionally dead.
The visitor isn't a lawyer. The visitor is a parent who's been served with divorce papers and is googling at 11 PM. They don't speak the language of "dissolution of marriage." They speak the language of "divorce." They don't search for "matters of parental rights and responsibilities." They search for "child custody."
Every sentence written in lawyer-speak instead of client-speak loses some fraction of the visitors who can't quickly map the legal terminology back to their actual situation. Most visitors don't ask for a translation. They leave.
The fix: rewrite the body copy of every public-facing page in the language clients actually use. "We handle divorces, child custody disputes, and modifications of existing court orders." Not "our practice encompasses." This isn't dumbing down content. It's translating legal accuracy into the words the audience already uses. Court documents stay in lawyer-speak. Marketing copy doesn't.
The bonus effect: AI engines also do better with client-speak. ChatGPT cites pages that match the language of the query. Visitors search "child custody," not "parental rights and responsibilities." Pages written in the client's language match more queries than pages written in the lawyer's language. Entity reinforcement works the same way: consistent, plain-language naming compounds.
For the Atlanta firm, we rewrote eleven practice area and resource pages over two weeks. Average time on page across the rewrites increased by 47%. Form completion rate on the divorce page went from 2.1% to 3.8%. Same traffic, same firm, different language.
What changed in 90 days
The Atlanta firm shipped all five fixes over a 60-day window. Photography first (the easiest), then phone number (the cheapest), then the trust strip and the lawyer-speak rewrite, then finally the headlines and the harder copy work.
The 90-day after-snapshot looked like this. Organic traffic: unchanged. Bounce rate on the homepage: down 24%. Time on practice area pages: up 38%. Consultation form submissions: from 32 per month to 71 per month. Phone calls from the website: from 11 per month to 31 per month. Total inbound from the website: from 43 per month to 102 per month.
The traffic that had been flowing for the previous eighteen months was suddenly converting. The visitors hadn't changed. The site was finally letting them buy.
The four mistakes worth not making in the first place
Three of the five mistakes are expensive to fix late and cheap to avoid up front. Worth flagging for any firm considering a new site or a redesign.
Stock photography. Don't ship a site without real photos. If a photo shoot can't happen before launch, ship with no photos rather than stock photos. The negative-space version reads as deliberate. The stock-photo version reads as generic.
Generic headlines. Write the H1 as the last step in the design process, not the first. Headlines drafted before the firm has a clear positioning end up generic. Headlines drafted after the positioning is settled end up specific.
Hidden phone numbers. Phone number in primary nav, sticky on mobile, tap-to-call enabled. Non-negotiable.
Lawyer-speak. Hire a non-lawyer to read every page of the site out loud before launch. If they stumble on a sentence, rewrite it. The senior partner shouldn't be the final reader on marketing copy. The senior partner already speaks lawyer-speak.
For how all of this fits into the broader site architecture (where your resource content lives, how visitors navigate between practice areas and education), see how to design a law firm knowledge hub. Conversion fixes on individual pages compound when the site as a whole is built right.
What this looks like as a real engagement
For a firm whose traffic is fine and whose conversions aren't, the conversion-recovery work usually fits inside a 30-day rebuild at our Launch tier ($3,500). Photography, headline rewrite, primary nav redesign, mobile trust block reconstruction, and a 10-to-15 page body-copy rewrite.
For a firm that needs the full site rebuilt because the underlying structure is also broken, this work folds into a larger Launch + Grow engagement (Launch tier plus monthly retainer).
The honest answer is that nobody can quote conversion fixes accurately without seeing the site. Send us your URL. We'll audit the five mistakes, identify which ones are costing you the most, and come back with a real quote for what it would take to fix them. Free, no card required, no obligation to hire us afterward.
