The Science of Law Firm Website Conversions: What Makes Visitors Call?

A criminal defense website has one job that matters more than the rest: getting the right person to pick up the phone. Rankings bring visitors to the page, but whether they call depends on what they see in the next few seconds. That part can be studied. You can form a guess about what stops people from calling, change one thing, and measure what happens.

This guide treats conversions the way a careful researcher would. It covers how to define and count a conversion on a defense firm site, why most firms can’t run textbook A/B tests, which page elements are worth testing first, and how to read results without fooling yourself.

Start with a question you can measure

Every useful experiment starts with a clear outcome. For a defense firm, that outcome is usually a phone call, with contact form submissions and live chats close behind. Decide which of those you count, and write the definition down before you change anything.

Calls are the hard part to count. A visitor who taps your number on a phone leaves your website and opens the dialer, so your analytics only sees the tap. Set up an event for clicks on any “tel:” link, then in Google Analytics 4 flag it as one of your key events. Google renamed GA4 “conversions” to “key events” in 2024, which confuses people reading older tutorials.

A tap is still an attempt rather than a conversation. Call tracking software fills the gap by showing a different phone number to visitors from different sources and logging each call that comes through. That tells you whether a call lasted thirty seconds or twenty minutes, which matters more than the raw count. If you add a tracking number to your Google Business Profile, a common approach is to make it the primary number and keep your main office line as an additional number, so the profile still matches your website and directory listings.

Why most defense firms can’t run a classic A/B test

In a true A/B test, half your visitors see version A and half see version B, and software tells you which one produced more conversions. That works well for an online store with thousands of orders a week. It works poorly for a solo defense practice whose DUI page gets a few calls a month.

The problem is sample size. With only a handful of conversions per version, random variation swamps any real difference. One busy weekend of arrests can make a worse page look better. Run the test long enough to get a reliable answer and you may be waiting most of a year, while court calendars, holidays and local news shift the baseline underneath you.

The tooling changed too. Google Optimize, the free testing tool many small sites relied on, shut down on September 30, 2023. Paid tools exist, but they can’t fix low traffic.

So most defense firms should run a simpler kind of experiment: a before-and-after comparison on one page, with one change, measured over a fixed period and checked against the same period before. It is weaker evidence than a randomized test. Done carefully, it is still far better than redesigning on instinct.

What a visitor needs to see on the first screen

The first screen on a phone is the most valuable space on the site, and the place to look for your first hypothesis. Someone who searched “assault lawyer” plus a county name at midnight wants four answers before they scroll:

1. Does this firm handle my kind of charge?
2. Does it work in my county or court?
3. Can I reach someone now?
4. How do I contact them with one tap?

Open each of your main charge pages on your own phone and check them against that list. Common failures include a large photo of the city skyline that pushes everything else down, a headline that names the firm but not the charge, and a phone number that only appears in the footer.

Each failure suggests a testable change. “Moving the phone number and hours into the first screen of the DUI page will increase tap-to-call events” is a proper hypothesis. It names one change, one page and one measurable result.

The phone number is the main experiment

Because calls drive most defense cases, small changes around the phone number deserve the most attention. Things worth testing one at a time:

• A sticky call button that stays at the bottom of the screen on mobile
• Button wording, such as “Call now” against “Talk to a lawyer tonight” when someone really does answer at night
• Showing office hours, or stating clearly that calls are answered after hours
• Placing the number beside the main headline instead of below the first section

The website is only half of this experiment. A tap that reaches voicemail at 11 p.m. counts as a click in your analytics and a lost case in real life. Call tracking reports often show missed calls clustered at nights and weekends, the hours when arrests happen. Fixing who answers the phone may move your results more than any design change.

If you record calls for quality review, check your state’s consent rules first. Some states, including California, require every party to agree before a call is recorded.

Forms, chat and friction

Not everyone wants to call. Some people are at work, some are in a shared space where they can’t talk about a charge, and some are researching for a family member. A short contact form serves them.

Each extra field is a small cost to the visitor. Name, phone, email and a short box for “What happened?” is usually enough to start. Questions about case number, arresting agency and court date can wait for the consultation. If you want to test this, remove fields one at a time and compare completion rates.

Add a sentence near the form explaining what happens next and how quickly someone replies. A line noting that contacting the firm does not create an attorney-client relationship is common, and some state bars expect similar language.

Chat widgets are worth testing with care. They can capture people who won’t call, but some load heavy scripts that slow the page, which may cost you more visitors than the chat brings in.

Trust signals and the rules around them

Visitors call firms they trust, and defense clients have more reason than most to be wary. Real attorney photos, a short bio listing bar admission and the courts where the lawyer appears, and genuine client reviews all give a nervous visitor something to go on.

Here the experiment has limits set by ethics rules. ABA Model Rule 7.1 bars false or misleading statements about a lawyer’s services, and a statement can mislead by creating unjustified expectations even when each word is true. Many states add their own rules on case results, testimonials and words like “specialist.” Some require disclaimers next to past results. Before you test a page that adds case outcomes or reviews, check your state bar’s advertising rules, and keep the compliant version as your baseline rather than the thing you’re testing against.

Speed is a variable you can measure

Page speed is easy to measure and easy to forget. Google’s Core Web Vitals track how fast the main content appears (Largest Contentful Paint), how quickly the page responds to a tap (Interaction to Next Paint, which replaced First Input Delay in March 2024), and how much the layout shifts while loading (Cumulative Layout Shift). Search Console reports all three.

On defense sites, the usual culprits are oversized hero images, autoplay video backgrounds, chat widgets and a pile of tracking scripts. Layout shift deserves special attention: if the call button jumps down the page as an image loads, a visitor can tap the wrong thing. Fix the slowest charge pages first, note the date, and compare call and form numbers for the weeks before and after.

How to run a fair test

A few habits separate a real result from a lucky one:

• Change one thing at a time on one page, and write down exactly what changed and when
• Measure in whole weeks, since weekend arrest patterns differ from weekdays
• Compare against the same length of time before the change, and against the same weeks last year if you have the data
• Decide in advance how long the test will run, and don’t stop early because the first few days look good
• Keep a simple test log in a spreadsheet so you can look back months later

Outside events can swamp a result. A new competitor in the map pack, a Google update or a change in your ad budget can move call volume with no help from your page. If one lands mid-test, note it and consider running the test again.

Read results by cases, not clicks

The final measure of a defense site is signed cases of the kind you want. A change that doubles form submissions from people outside your service area hasn’t helped. A change that brings fewer but better calls may have.

Connect your intake records to the source of each lead where you can. Even a simple column in your intake sheet noting which page or number the caller came from will tell you, over a few months, which pages produce real clients. That’s the data that should decide your next experiment.

Questions about testing a defense firm website

How much traffic do I need to run a real A/B test?

There’s no single number, because it depends on how many conversions each version gets and how big a difference you’re trying to detect. As a rough rule, if a page gets only a few calls a month, a randomized test will take too long to be useful, and a careful before-and-after comparison is the better choice.

What should I test first?

Start with the first screen on mobile of your busiest charge page. Make sure it names the charge and the area, shows the phone number, and says when someone answers. That one page usually has more room for improvement than anything else on the site.

Do heatmaps and session recordings help?

They can show where visitors tap, how far they scroll and where they give up. Treat them as a source of ideas for tests, not as proof. Check that any tool you add doesn’t slow the page, and review your privacy policy before recording visitor sessions.

How long should a before-and-after test run?

Long enough to cover several full weeks and a meaningful number of calls. Many small firms find a period of one to three months workable. Set the length before you start and stick to it.

---

This guide is part of SBD Marketing’s series on criminal defense marketing. SBD Marketing works only with criminal defense law firms, building websites and running SEO aimed at calls from the cases a firm wants.

SBD Marketing · info@sbdmarketing.agency · Book a call: sbdmarketing.agency/scheduling · Serving criminal defense law firms across the USA · sbdmarketing.agency