March 2026 Core Update Post-Mortem: Winners and Losers

Google’s March 2026 core update started rolling out on March 26 and finished on April 8. 13 days from start to finish. We waited one week after completion, then pulled Ahrefs organic traffic data across 30 domains to measure what actually moved.

Here’s what we found:

The biggest movers

Top losers by traffic change:

• legalzoom.com: −36.9%
• airbnb.com: −29.5%
• coursera.org: −15.1%
• redfin.com: −12.5%
• huffpost.com: −9.6%
• techradar.com: −7.1%
• bankrate.com: −4.6%
• ebay.com: −4.0%

Top winners by traffic change:

• forbes.com: +14.3%
• udemy.com: +9.4%
• medicalnewstoday.com: +9.0%
• amazon.com: +5.6%
• zillow.com: +5.3%
• buzzfeed.com: +4.6%
• everydayhealth.com: +4.5%
• cnet.com: +4.1%

LegalZoom’s 37% drop is the headline figure, but the more useful story is in the patterns that emerge when you look at vertical by vertical.

Sector breakdown

Health: across-the-board gains

Every health site we tracked gained traffic.

Healthline added 1.3 million visits (+2.5%). Medical News Today added 1.9 million (+9.0%). Everyday Health added 99,000 (+4.5%).

Health publishers have had a complicated few years with core updates, with many hit hard in 2023 and 2024. This update reversed that trend, at least for these three. The common thread is editorial depth: all three publish condition-specific, symptom-specific content written and reviewed by medical professionals or credentialed contributors.

Jobs: across-the-board losses

The jobs sector went the opposite direction. Every site we tracked lost traffic.

Indeed dropped 1.95 million visits (−1.7%). LinkedIn dropped 1.8 million (−1.3%). ZipRecruiter dropped 277,000 (−2.6%).

The LinkedIn number is probably most notable. It’s one of the highest-authority domains on the internet by any measure. It has a massive referring domain count, strong brand signals, and high engagement. Yet it still lost traffic in this window.

The fact that three structurally different companies in the same vertical all moved in the same direction suggests the update treated job-related content as a category, not individual sites.

One possible explanation: job listing pages are low-differentiation content. A listing for a software engineer role at a mid-sized company looks more or less the same on Indeed, LinkedIn, and ZipRecruiter. Google may be tightening how it treats this format.

Travel: uniformly down

Booking.com lost 1.96 million visits (−3.2%). TripAdvisor lost 1.08 million (−1.7%). Airbnb lost 3.9 million (−29.5%).

Airbnb’s drop is the most dramatic in absolute terms across the entire dataset. The scale is partly a function of how much traffic it had going in, but 29.5% in two weeks is a significant correction regardless of the starting point.

All three travel sites are fundamentally aggregator or marketplace models. They organize and surface inventory from third parties. The update appears to have assigned less weight to this format in travel, similar to what it did in jobs.

Ecommerce: Amazon up, marketplaces down

Amazon gained 22.97 million visits (+5.6%). Etsy dropped 1.25 million (−2.9%). eBay dropped 2.13 million (−4.0%).

This split is one of the more counterintuitive results. Amazon is arguably the largest ecommerce marketplace in the world. It aggregates third-party sellers as much as it sells its own inventory. Yet it gained while Etsy and eBay lost.

The likely distinction: Amazon has developed a level of brand authority and direct navigational traffic that Google treats differently from niche marketplaces. A search for almost any product category now resolves to Amazon as an assumed destination. Etsy and eBay don’t have that same category-level dominance.

Finance: editorial wins, aggregators lose

Bankrate dropped 494,000 visits (−4.6%). NerdWallet was essentially flat. Forbes.com gained 2.5 million (+14.3%).

Bankrate and NerdWallet are both rate-comparison and personal finance aggregators. Forbes.com is editorial-first, with financial content mixed across reviews, news, and analysis.

The pattern holds with what happened in travel and jobs: sites that primarily surface and compare third-party data lost ground relative to sites that publish original editorial content.

*Note: the Forbes.com figure reflects the full domain, not just the /advisor section.

Legal: information sites win, the service site loses

LegalZoom dropped 209,000 visits (−36.9%), the largest percentage drop in our dataset. Justia gained 115,000 (+3.5%). FindLaw dropped 49,000 (−3.2%).

LegalZoom is primarily a legal services company: it sells document preparation, formation services, and subscriptions. Its organic presence is built largely on commercial-intent pages. Justia, by contrast, is a legal reference database; case law, statutes, attorney directories, and free legal information.

This is one of the clearest commercial vs. informational splits in the data.

Tech: mixed, with a surprising winner

TechRadar dropped 274,000 (−7.1%). PCMag dropped 199,000 (−2.7%). CNET gained 420,000 (+4.1%).

CNET’s gain is notable because it contradicts the narrative from recent years that CNET was struggling with trust issues following controversy over AI-generated content.

The traffic data doesn’t show that, at least not in this window. It’s possible that CNET’s editorial cleanup and renewed focus on original product testing are being reflected here.

TechRadar and PCMag both publish large volumes of reviews. The distinction between them and CNET that explains the divergence isn’t immediately obvious from the data alone.

Real estate: a clean split

Redfin dropped 797,000 (−12.5%). Realtor.com gained 174,000 (+1.1%). Zillow gained 1.9 million (+5.3%).

Redfin has built a significant content operation around home listings and neighborhood guides, but a core part of its model is its own brokerage and transaction services.

Zillow and Realtor.com are more purely informational in their organic content strategy, aka they surface listings and data without being transaction-first in the same way.

Education: the Coursera/Udemy split

Coursera dropped 2.27 million visits (−15.1%). Udemy gained 665,000 (+9.4%). Khan Academy was essentially flat (−0.2%).

Coursera’s drop is one of the largest in absolute terms in our dataset. Both Coursera and Udemy are course marketplaces, which makes the divergence interesting.

One possible distinction: Coursera has moved heavily toward corporate and institutional partnerships, with much of its content gated behind subscriptions. Its organic pages often lead to registration walls.

Udemy’s course pages are more openly browsable. Khan Academy’s flatness is its own story. The update didn’t reward it, but it also didn’t penalize it.

News: mixed results

The Guardian was essentially flat (−0.2%). HuffPost dropped 119,000 (−9.6%). BuzzFeed gained 339,000 (+4.6%).

The Guardian’s flatness is worth noting. It’s the highest-traffic news site in our dataset, and it didn’t meaningfully move in either direction. HuffPost’s drop follows a longer trend of that property losing search visibility over multiple updates.

What this update appears to be doing

A few threads run through the data.

Aggregators vs. publishers. Sites that primarily organize and surface content from third parties — job listings, travel inventory, rate tables, course catalogs — underperformed relative to sites that produce original editorial content. This was consistent across finance, travel, jobs, legal, and, to some extent, ecommerce.

Sector-level patterns, not just site-level ones. In several verticals, every site we tracked moved in the same direction. Every health site gained. Every travel site and every job site lost. These movements suggest Google is adjusting how it weights entire content categories, not just individual domains.

Brand authority has a ceiling. LinkedIn is one of the most authoritative sites on the internet by most measures, and it still lost 1.3% of its traffic. That suggests this update wasn’t primarily about link authority or domain-level trust signals. Something about the content format or vertical itself was being adjusted.

Within-vertical differentiation. In finance, legal, tech, and ecommerce, sites in the same space had opposite outcomes. Bankrate and Forbes.com are both finance sites. LegalZoom and Justia are both legal sites. The separating factor in most cases was whether the site’s primary value is original editorial content or facilitated transactions and aggregated data.

Methodology

• Data source: Ahrefs organic traffic estimates
• Baseline date: March 26, 2026 (day the update started rolling out)
• Measurement date: April 14, 2026 (one week post-completion)
• Domains tracked: 30 across 10 verticals (health, finance, jobs, travel, real estate, ecommerce, legal, tech, education, news)
• Traffic change calculated as: (Apr 14 traffic − Mar 26 traffic) / Mar 26 traffic × 100
• Note: Forbes.com data reflects the full domain, not the /advisor subdirectory

Want to know how your site fared? Jump on a free strategy call with us, the Loganix team.

*These are Ahrefs estimates, not verified server-side analytics. They reflect estimated organic search traffic based on keyword rankings and click-through modeling. As with all third-party traffic tools, treat the figures as directional rather than exact.