Does AI referral traffic convert better than Google search for law firms?

The conversion figures you’ll see quoted are real numbers, but none of them can tell you anything about your firm. AI search still matters, and there’s something true under the claim. I get to that further down.

What the studies say

Studies comparing AI-referred traffic with other traffic report very different conversion rates.

When reported conversion lift from AI-referred traffic ranges from 23x to no measurable difference, you can find a number to support almost any pitch. A single figure from that range tells you more about which study was chosen than what you should expect.

Only one study used a control

Amsive was the only study that compared AI-referred traffic with a proper control group. Across 54 sites and six months of GA4 data, organic traffic converted at 4.6% and AI referrals at 4.87%, which is effectively the same.

Every other study in the table came from a vendor or platform analyzing its own traffic without that kind of comparison.

Amsive did find a bigger difference on B2B sites, where AI referrals converted at 2.17% versus 1.16% for organic. That makes it worth watching for law firms, but it is still a long way from proving claims like 5x higher conversion.

The retail figure reversed in a year

Adobe reported AI traffic converting 38% worse than non-AI traffic in March 2025, and 42% better in March 2026. The publisher and the method were the same, and the conclusion flipped in twelve months.

Whatever that’s measuring, it isn’t a stable feature of AI traffic. A figure that swings eighty points in a year shouldn’t set anyone’s budget.

None of the studies measured a law firm

Every study above measured software signups or retail checkouts. Those happen in one session, on one device, with the whole path visible in analytics.

A personal injury case works differently. The conversion is a phone call, followed by an intake screen, a decision to sign, and often a wait of days. The signed case gets recorded in a case management system, where no browser analytics can see it. Applying a SaaS signup ratio to that means measuring something else entirely, so take these numbers with a grain of salt.

Why the multiples come out so high

In the Ahrefs figure, AI referrals were 0.5% of all sessions. So the study compares a very small, heavily filtered slice of visitors against all organic traffic.

Think about what someone arriving from an AI answer has already done. They asked a question and an AI platform answered it. The answer named a firm in a context that already matched their situation, and they still chose to click through. They’d been qualified several times before they reached the site.

Organic search includes every casual searcher, every person comparing four firms, every student, every competitor and every accidental visit. A tiny pre-filtered group will convert better than that. So AI visitors convert better mostly because they were already close to deciding when they clicked. That doesn’t make AI traffic worth five times as much as search traffic.

What’s really driving conversion

People arriving from an AI answer may convert better because the AI has already done some of the qualification. By the time they reach your site, they may know more about their options and be closer to choosing a firm. The evidence on how much that improves conversion is mixed, but the relationship matters.

That’s why chasing AI traffic because it supposedly converts 5x misses the bigger point. AI answers are moving qualification upstream, before the prospect reaches your site. If your firm is missing from those answers, you may never get the chance to convert them at all.

What a firm should measure instead

Measure these three, in this order.

1. Signed cases by source. Sessions, conversion rates and form fills don’t count here. If your intake and call tracking can’t trace a signed case back to where it came from, none of your channel comparisons are real, and that’s the first thing to fix.
2. Whether you appear at all, on each AI platform. Citation pools barely overlap between platforms. Roughly 91% of citations appear on only one, so a single blended “AI visibility” score hides where you’re missing. Check yes or no for each platform before you look at how often you’re cited.
3. Cost per signed case, by channel, over a long enough period. That’s the number that decides budget. It also holds up whatever the AI platforms do next, because you measure it in your own case management system.

If someone leads with a conversion multiple, ask them for the first item on that list for your firm. Without it, the multiple tells you nothing about your cases.