March 2026 Google Volatility: Why Your Rankings Won't Stabilize

If your rankings spent late March bouncing around like a dropped Slinky, you weren't cursed and your tracking tool wasn't broken. The March 2026 Google core update genuinely was one of the rougher ones, and the ranking volatility it produced was measurably worse than the update before it.

We run SEO campaigns for a living, so we watched this one land across a few dozen client sites in real time. Some gained, some got clipped, a couple had a genuinely bad fortnight. Here's what actually happened, what the data says, and what we'd tell you to do if your traffic graph currently looks like a cliff edge.

What was the March 2026 Google core update?

The March 2026 core update was a broad change to Google's ranking systems that started rolling out on 27 March 2026 and finished on 8 April 2026, running for 12 days and 4 hours according to Google's own Search Status Dashboard. That's the first thing worth saying out loud: this was confirmed by Google, not an "unconfirmed update" that SEOs invented to explain a bad week.

Google described it, per its confirmation reported by Search Engine Journal on 8 April, as "a regular update designed to better surface relevant, satisfying content for searchers from all types of sites." Standard core-update language. The reality on the ground was anything but routine.

What made it spicier was the timing. A short, sharp spam update ran first, from 24 to 25 March, lasting under 20 hours, and then the core update launched two days later. So a lot of sites copped two punches in roughly a week and couldn't tell which one moved them. If you want the spam-update side of the story, we broke that down in our piece on what to do after the March 2026 spam and core updates.

How volatile was it, really?

Volatile enough that we stopped trusting our own eyes and went to the numbers.

Search Engine Land's analysis, published on 15 April 2026, laid out just how much the SERPs moved compared with the December 2025 core update. The headline figures are genuinely eye-watering:

• Top 3 results: 79.5% of URLs changed position in March, up from 66.8% in December. Only one in five pages held their exact spot.
• Top 10 results: 90.7% of URLs shifted, versus 83.1% in December. Stability in the top 10 dropped to 9.3%.
• Falling off a cliff: roughly 24.1% of pages that had been ranking in the top 10 fell out of the top 100 entirely, compared with 14.7% after December.

Read that last one again. Nearly a quarter of top-10 pages didn't just slip a few spots, they vanished past position 100. In our experience that's not "your title tag was a bit weak". That's Google deciding a whole category of result no longer deserves the visibility it had.

For context on speed, the March update wrapped in 12 days. The December 2025 update took 17 days, so March was faster and angrier, which is an uncomfortable combination if you were on the wrong side of it. Search Engine Land confirmed the rollout completed on 8 April at 06:12 PDT, slightly ahead of Google's own "up to two weeks" estimate.

Why did this update feel so brutal?

Because three things stacked on top of each other in five weeks: a spam update, a core update, and the ongoing creep of AI Overviews eating the top of the page.

Our take, based on what we saw across client accounts, is that the core update kept pushing the same direction Google has been heading for two years. It rewards sites that look like a primary source and squeezes sites that sit in the middle. Established brands, official and institutional pages, and genuinely specialist publishers tended to hold or gain. Thin aggregators, directories, and "we summarised what ten other sites said" content got thinned out. Search Engine Land's data showed exactly that pattern of movement towards official sources and away from intermediaries.

Layer AI Overviews on top and the maths gets worse even for winners. You can climb from position 4 to position 2 and still lose clicks, because an AI summary now sits above you answering the question outright. We dug into how steep that drop can be in our breakdown of the click-through rate decline from AI Overviews, and the short version is that "ranking well" and "getting traffic" are quietly becoming two different jobs.

So when people say March felt worse than the raw volatility scores suggested, we think they're right. The ranking churn was high, and the clicks behind each ranking were already softer than they were a year ago.

Who got hit and who got rewarded

Patterns are more useful than panic, so here's what we consistently saw.

Sites that took damage

Thin affiliate and "best X for Y" round-up pages with no first-hand testing got hammered. So did directory-style pages that exist purely to rank, and YMYL content without credible authorship. We've argued for a while that Google holds health, finance and legal content to a higher bar, and March reinforced it hard. If you run that kind of site, our guide to why Google judges YMYL sites more strictly is worth a read before you touch anything.

Sites that gained

Brands with a real reputation, original data, and content people actively search for by name did well. Specialist publishers who genuinely know their niche climbed. The common thread wasn't "more content", it was "content that clearly comes from someone who did the work". We pulled apart the specific winners and losers in our companion analysis of the March 2026 core update winners and losers, and the gap between the two camps was wider than usual.

One honest admission: a couple of sites we'd have bet on held flat, and one thin-ish page we expected to drop actually nudged up. Core updates are broad, not surgical, and anyone telling you they can predict every individual movement is selling something.

What does Google actually tell you to do?

This is the part most "recovery" articles skip, so let's quote the source directly.

Google's official documentation on core updates is clear that "core updates are broad in nature, and don't target specific sites or individual web pages." In other words, dropping doesn't mean you did something wrong or earned a penalty. It often means other pages got judged as more useful for the query than yours.

Google also recommends waiting "at least a full week after a core update completes before analysing your site in Search Console". Given March finished on 8 April, that means the earliest sensible time to draw conclusions was mid-April, comparing a clean post-update week against a week before the update started.

And on the fixing itself, Google's guidance is blunt: "avoid doing 'quick fix' changes" and instead "focus on making changes that make sense for your users and are sustainable in the long term." Recovery can take until a future core update or broad system change, sometimes months, which is the bit nobody wants to hear but everybody needs to.

Our short translation of all that: don't gut your site in week one, don't chase a magic checklist, and don't expect a same-week rebound. Diagnose calmly, fix properly, wait.

What we'd actually do if you lost rankings

Here's the order we work through with clients when a core update bites. It's deliberately boring, because boring is what works.

1. Wait at least a week, ideally two. Google says a week minimum, and we've watched plenty of "permanent" drops partially reverse in the back half of a rollout. Don't make irreversible decisions during the churn.
2. Pull the real data before you theorise. Open Search Console, compare a post-update week with a pre-update week, and find the exact queries and URLs that lost. "Traffic is down" is not a diagnosis. "These nine pages lost these twelve queries" is.
3. Separate intent loss from quality loss. Sometimes you didn't get worse, the SERP changed shape. If an AI Overview, video pack or shopping unit now owns your old position, that's a different problem from being out-competed by a better page.
4. Audit authorship and trust signals on the pages that fell. Who wrote this, why should anyone believe them, and does the page prove it? For YMYL especially, vague bylines are a liability now.
5. Cut or merge the dead weight. If a chunk of your site is thin, duplicative or last updated in 2022, it can drag the whole domain. Our content pruning playbook walks through how to decide what to fix, merge or delete without nuking pages that still earn their keep.
6. Check the technical floor. Google has no shortage of good content to choose from, so a slow, unstable, awkward-on-mobile site gives it an easy reason to pick someone else. Our notes on technical SEO that actually moves rankings cover the checks worth doing first.

Notice what's not on that list: rewriting every page in a panic, firing your content team, or buying a "core update recovery" tool. None of those fixed anything in March, and they won't next time either.

Is permanent volatility the new normal?

Mostly, yes, and we'd rather you hear that from us than learn it the hard way.

Google runs smaller, unannounced changes constantly between the big confirmed core updates. Its own documentation acknowledges these ongoing tweaks happen all the time, which is partly why your rankings wobble on a random Tuesday with no headline update to blame. The confirmed core updates are just the visible peaks. The background hum never stops.

Add AI Overviews and AI Mode into the mix, and the system is getting more dynamic, not less. So the goal of "find the perfect setup and lock it in" is gone. There is no setup to lock in. The sites that sail through volatility aren't the ones with the cleverest tricks, they're the ones Google keeps choosing because they're genuinely the best answer.

That's also why we keep telling clients to stop renting their entire business from Google. Build an email list. Get known by name. Earn direct traffic and a brand people search for on purpose. A core update can move your rankings, but it can't touch the audience that already trusts you.

The honest takeaway

The March 2026 core update was confirmed, it was 12 days long, and it was measurably more volatile than December, with nearly 80% of top-3 URLs moving and almost a quarter of top-10 pages dropping out of the top 100. Those aren't scare numbers, they're verified ones.

If you got hit, the fix isn't dramatic, it's disciplined. Wait, diagnose with real Search Console data, improve the things that genuinely make your pages more useful, and stop depending on a single traffic source. Do that consistently and you stop fearing updates, because each one becomes a chance to gain rather than a thing to survive.

If you'd rather not pick through it alone, that's literally our job. Take a look at how we approach SEO as a system we diagnose and fix, or tell us what your rankings did in March and we'll tell you straight whether it's the update, your content, or something technical underneath. No fortune-telling, just the data and a plan.