What percent of law firm websites survived the March 2026 core update?

Roughly 45% of legal sites tracked across the Taqtics panel held steady or grew through the March 2026 rollout, which closed on April 8. The remaining 55% saw declines of 20-35% in organic traffic. The split was sharper than any update since the 2023 Helpful Content shift.

What do law firm sites that survived the March 2026 update have in common?

Three patterns repeat across the survivor set. Original first-party data inside the content. Named attorney bylines with full Person schema and bar admissions. Clean technical fundamentals tied to a consistent brand entity across the web. Sites that hit all three usually grew. Sites that hit one or two mostly held.

Which practice areas survived best in March 2026?

Niche verticals outperformed generic personal injury. Nursing home abuse, NEC mass tort, mesothelioma, and specific occupational injury sites held better than generic car accident pages. Topical focus and depth beat keyword volume. The thinner the content moat, the worse the drop.

Can a law firm rebuild after losing traffic in March 2026?

Yes, but recovery from a core update runs 3-6 months minimum and isn't guaranteed. The faster move is channel diversification. CTV at $12-22 CPM and household-level attribution generates branded search demand the next algorithm update can't suppress. Rebuild SEO in parallel. Don't bet the firm on it.

45% of Law Firm Sites Survived March 2026. Three Patterns Separate the Winners.

Forty-five percent of law firm sites came through the March 2026 core update without meaningful damage. The rollout closed on April 8. The story everyone keeps writing is the 55% that lost traffic. The story worth writing is the 45% that didn’t.

Across the Taqtics panel of 3,720 legal advertisers and 210 US DMAs, the winners aren’t a random scatter. They share patterns. Three of them, mostly. A fourth, sometimes. The patterns are boring, slow, and impossible to fake in 30 days. Which is exactly why they hold up when the algorithm shifts.

This is the analysis nobody else can run, because nobody else watches every legal advertiser in every market month after month.

What “Survived” Actually Means

Survival in core-update terms is narrow. A site held its organic traffic within roughly 5% of its prior 30-day baseline. Some grew. A handful grew sharply. The rest of the 55% dropped 20-35%, with thin-content sites at the steeper end of that range.

The split isn’t gentle. Core updates have always created winners and losers, but March 2026 sharpened the gap. The middle ground is shrinking. Sites either had the structural signals Google’s classifier rewards or they didn’t.

Most agencies are framing this as a content quality problem. That’s part of it. The bigger story is that Google is now rewarding entity strength on top of content strength. The two compound. Sites that nailed both pulled away. Sites that nailed neither got buried.

Pattern One: Original Data Inside the Content

Every survivor we mapped publishes something a competitor doesn’t.

Sometimes it’s settlement breakdowns by injury type. Sometimes it’s local accident statistics pulled from public records. Sometimes it’s anonymized case timelines or recovery rates by claim category. The format varies. The principle doesn’t. The page contains a number, a chart, or a finding that doesn’t appear on 50 other firm sites.

The losers ran the same template the rest of the market ran. “Car accident lawyer in [city].” Five paragraphs of generic copy. A practice area list. A contact form. Replace the city name and the page works for any firm anywhere. Google’s classifier figured that out years ago. March 2026 finished the job.

What “original data” means in practice is narrower than most firms think. It doesn’t require a research team. A PI firm can publish its own quarterly settlement summary, its own breakdown of case duration by injury severity, or its own analysis of local court filings. Real numbers from real cases, anonymized properly. That’s it.

The firms publishing this material are also the firms getting cited inside AI Overviews on legal queries. The classifier and the generative model want the same thing. Primary sources. Real data. Stuff that exists nowhere else on the web.

Pattern Two: Named Author Bylines and Real E-E-A-T

Every survivor has a name on every article.

Not “Posted by admin.” Not “The Firm Team.” A specific attorney with a bio page, full Person schema, bar admissions listed by state, education credentials, and ideally some external proof of expertise. Speaking engagements. Published commentary. Bar association leadership. Anything that confirms the byline is a real human with real authority.

The losers have either no byline or a placeholder byline. Their attorney bio pages, when they exist, are 200 words of marketing copy without schema. Google can’t confirm the writer is a lawyer. The classifier downgrades the page accordingly.

E-E-A-T isn’t a ranking factor in the literal sense. It’s a system of signals Google’s quality raters use to evaluate output. The March 2026 update appears to have tightened how those signals translate into ranking. Sites with weak author signals dropped. Sites with strong ones held.

The fix is mechanical. Add Person schema to every attorney bio. List bar admissions, credentials, education, and external authority signals. Add author bylines to every article with the schema linking to the bio page. Make sure each attorney’s name appears consistently across the firm site, Google Business Profile, legal directories, and any third-party citations.

This work isn’t optional anymore. It’s table stakes for surviving the next update.

Pattern Three: Technical Fundamentals + Brand Entity Consistency

The third pattern is the one most firms underestimate.

Survivors had clean Core Web Vitals. Pages loaded under 2.5 seconds. Cumulative Layout Shift stayed below 0.1. Largest Contentful Paint cleared the green threshold on mobile. They had complete, validated schema markup across LegalService, Article, FAQ, and Person types. They had unique meta descriptions on every page, a single H1 per page, and clean internal linking.

They also had brand entity consistency. The firm’s name, address, and phone number matched across the firm site, Google Business Profile, every legal directory listing, and every third-party citation. No variant spellings. No abbreviated office names. No outdated suite numbers. The model could resolve the entity to one canonical version.

Losers had the opposite. Duplicate meta descriptions across dozens of pages. Missing H1s. Schema gaps or invalid markup. Page weight north of 4MB. Three different versions of the firm name across the web. NAP inconsistencies that broke the entity graph.

None of this is glamorous. It’s the work nobody wants to do. Which is why it separates winners from losers when the algorithm tightens.

The firms that paid SEO agencies $5K to $15K per month for years and got nothing but content output are looking at the gap right now. The agency wrote 200 blog posts. Nobody fixed the schema. Nobody cleaned up NAP. Nobody audited Core Web Vitals. The investment produced volume without infrastructure. March 2026 punished that exact tradeoff.

Pattern Four: Topical Focus Beat Keyword Volume

A bonus pattern, smaller but real.

Niche practice areas held better than generic personal injury. Nursing home abuse sites. NEC mass tort pages. Mesothelioma firms with deep historical content. Sites focused on a specific occupational injury type or a specific defective product. Topical depth in a narrow vertical beat shallow breadth across every PI category.

The reason is structural. A site that publishes 80 pages on nursing home abuse, written by attorneys with that specific practice area in their bio, with original case data and external citations from elder care advocacy groups, looks like a topical authority. The model trusts it on that subject.

A site that publishes one page each on car accidents, truck accidents, motorcycle accidents, slip and fall, dog bites, premises liability, and 14 other case types doesn’t look like an authority on any of them. It looks like a marketing site. Google now treats it that way.

Generic PI pages dropped harder than any other category in March. Niche vertical pages mostly held. The lesson generalizes. Pick the practice area that drives the most signed cases for the firm. Publish ten times as much depth on that area as on everything else combined. Let the rest of the practice areas live as supporting pages, not equal pillars.

What the Survivors Did Months Ago

None of these patterns are new. Every one of them was visible in core update analyses going back to 2022. The firms that survived March 2026 weren’t lucky. They were paying attention.

They invested in original research instead of templated content. They added Person schema to every attorney bio two years ago, not last week. They cleaned up their NAP across the web before Google penalized them for it. They picked a niche and went deep instead of spreading thin across every PI vertical in their state.

The ones who waited got the bill in April.

The Math on Recovery

The winners now have a cost-per-lead advantage their competitors can’t close in 90 days.

A survivor generating 200 organic leads per month at a stable cost per signed case just watched competitors lose 30% of their organic flow. Those competitors are now bidding more aggressively on Google Ads to compensate. CPCs in competitive PI keywords have ticked up across the panel since the rollout closed. The survivor pays the same SEO cost. The losers pay more for paid traffic to refill the gap.

Recovery from a core update takes 3-6 months minimum, if the diagnosis is correct and the fixes are right. Some sites never fully recover. The firms that recover fastest are the ones that already had two of the three winning patterns and are missing the third.

The firms missing all three are looking at a longer rebuild than they want to acknowledge. The honest path forward isn’t a panicked content sprint. It’s a structural rebuild on entity, schema, and original data. Same work the survivors did two years ago.

The Channel That Doesn’t Care About Updates

Here’s the part that makes the survivor advantage compound.

CTV ad campaigns generate branded search demand within weeks. A 30-second spot in a firm’s primary DMA produces measurable lift in branded queries, direct site visits, and calls to the firm’s tracked numbers. None of that traffic flows through the algorithm Google updates every quarter. None of it can get cut by the next core update.

CTV inventory across 210 markets is still wildly under-bought by legal. The Taqtics panel shows roughly 14% of legal ad spend on streaming nationally. Atlanta hits 48%. Washington DC sits at 3%. The firms moving early into their local CTV market are getting floor-level CPMs. CPMs run $12-22 in most markets. Completion rates clear 90%.

Pair an SEO survivor profile with a CTV brand layer and the result is durable demand generation. Branded search rises. Direct calls rise. The next core update doesn’t matter the same way. The firm has channels Google can’t shut off.

The losers are running the opposite playbook. SEO took the hit. Paid search is more expensive to refill the gap. Nobody’s running CTV. Nobody’s building branded demand. Every algorithm update lands harder than the last one.

What to Do This Quarter

The work splits along the patterns.

Audit the entity first. Pull every web mention of the firm. Standardize the name, the address, the phone, and the credentials across every property. Fix the schema next. Most legal sites have nothing structured beyond basic Article markup. Add Person schema to every attorney bio. Add LegalService schema to the firm. Add FAQ schema to question-driven content.

Publish original data on a recurring schedule. Quarterly settlement summaries. Annual local market reports. Monthly commentary on legal news in the firm’s practice areas. Real numbers from real cases. The kind of content the model can cite as a primary source.

Pick the practice area that drives the most signed cases. Build topical depth there. Let the rest of the site exist as supporting material.

While that compounds, build a brand channel that doesn’t go through Google. CTV in the firm’s primary DMA. Targeted broadcast in mid-size markets. Direct response that produces tracked calls. Anything that generates demand outside the algorithm.

The firms that survive the next update will look like the firms that survived this one. The patterns won’t change. The cost of ignoring them keeps rising.

The 45% who came through March 2026 didn’t get lucky. They built the right structure two years ago and ran it patiently. The window to do the same work, before the next update, is open right now.

References

Roughly 45% of law firm sites came through the March 2026 core update without meaningful damage. Taqtics' panel of 3,720 legal advertisers across 210 DMAs surfaces three patterns the survivors share: original data inside their content, named author bylines with real E-E-A-T signals, and clean technical fundamentals tied to a consistent brand entity. The firms with all three didn't just survive. They picked up share. The ones that paired survival with a CTV brand layer compounded it.