May 2026 Core Update Complete: What Changed and What to Check

If your traffic shifted in the last two weeks of May, you now have an official reason to look into it. Google has marked the May 2026 core update complete: it started on 21 May and finished on 2 June 2026. The short answer to “what changed?” is this. Google reassessed how it ranks content across all kinds of sites, and it gave no new guidance. The real work is finding out what changed for you.

This guide is about that work. I cover when to compare, how to tell May’s effects apart from the March core update just eight weeks earlier, how to check pages and queries, and how to plan a recovery that doesn’t waste effort.

May 2026 Core Update Complete: The Confirmed Timeline

Google’s Search Status Dashboard entry for the May 2026 core update is the only source you need for the dates. All times are US Pacific.

• Started: 21 May 2026 at 08:43 PDT. Google’s first note said the rollout “may take up to 2 weeks to complete.”
• Completed: 2 June 2026. Google updated the dashboard at 05:43 PDT to say the rollout was complete as of that day.
• Duration: 11 days and 21 hours. That’s a little faster than the March 2026 core update (12 days and 4 hours) and well inside the two-week estimate.

The dashboard showed no other confirmed ranking updates between the end of the March core update on 8 April and the start of this one. That matters. It gives you a fairly clean stretch of data to use as a baseline (your “normal” before the change). If you missed the launch coverage, our piece on Google’s May 2026 core update rolling out has the background.

What Google Said, and What It Didn’t

Google’s message was the same one it used for March. It called this a regular update designed to better show relevant, satisfying content for searchers from all types of sites. It also said again that there is nothing new or special creators need to do.

No blog post explained what was targeted, and Google published no new documentation. The completion notice confirmed the date and nothing more.

In practice, I take three things from that:

• Your reference point hasn’t changed. Google’s guidance on core updates and your website and its advice on helpful, reliable, people-first content are still the documents to work from.
• A drop is not a penalty. A core update won’t put anything in the Manual actions report. If something does show there, that’s a separate problem with a separate fix.
• Anyone who claims to know the exact target is guessing. Google doesn’t publish the signals a core update changes. So claims like “Google now rewards X” can’t be checked.

The update also launched in the same week as Google’s I/O announcements. Those included several changes to AI features in Search. Some SEOs linked the timing. Google didn’t link the two, so I treat any connection as a guess. Our roundup of every search change announced at Google I/O 2026 lists what was actually announced. That helps when you’re ruling out other causes.

What People Saw During the Rollout

By 3 June, nobody has a clean week of data after the update. That’s the honest starting point for anything you read about this update, including this article.

Here is what people could fairly say, based on third-party rank tracking tools and SEO reports during the rollout:

• It started showing fast. Volatility trackers (tools that measure how much rankings move) picked up big movement by the first weekend, around 23 and 24 May.
• It came in waves. A second burst of movement came around 30 and 31 May. Another spike came the day before Google said the rollout was complete.
• It felt bigger than March. Several experienced SEOs said May was more noticeable than the March update, which many found fairly quiet. May looked more like a typical broad core update.
• It was broad. People reported impact across many industries and countries. That fits Google’s description of a global update for all types of sites.

What no one has shown is a checked pattern of which industries, site types or content formats won or lost. If you see percentages or sector lists published before mid-June, treat them as early guesses. Your own Search Console data will tell you far more.

The wave pattern has one practical result. A site that moved on 23 May and a site that moved on 2 June may have been hit by different parts of the rollout. I always write down the exact date pages moved. It helps later when you group losses.

When to Compare, and Which Dates to Use

Google recommends waiting at least a full week after a core update finishes before you analyse. With completion on 2 June, your first clean week after the update is 3 to 9 June. You can’t check it properly until around 11 June, because Search Console data runs a couple of days behind.

Google also recommends comparing that week with the week before the rollout began. For May 2026, I’d use these windows:

1. Pre-update baseline: 14–20 May 2026. The seven days before the rollout started on 21 May.
2. Post-update week: 4–10 June 2026. This matches the baseline day for day (Thursday to Wednesday), so weekday and weekend patterns line up. If you can’t wait, 3–9 June is fine, but you’ll be comparing slightly different days of the week.

Two calendar quirks are worth knowing:

• Bank holiday on 25 May. Memorial Day in the US and the spring bank holiday in the UK both fell on 25 May, in the middle of the rollout. Many sites see odd traffic on public holidays. Neither comparison window includes that date. That’s one more reason not to judge the update on rollout data.
• Seasonal businesses. Late May into June is a seasonal turning point for travel, gardening, education and events. If that’s you, add a year-on-year check (June 2026 against June 2025). Then you won’t mistake normal seasonal change for an algorithm change.

How to Separate May’s Effects From the March Core Update

This is where May 2026 is different. Two broad core updates landed about eight weeks apart, so one simple “before and after” check can mislead you. A page that dropped in March and recovered in May looks unchanged if you only compare around May. A page still falling after both updates needs a very different response from one hit for the first time.

If you checked the March update using our March 2026 core update completion guide, you already have half the data. If not, pull three periods for every important page:

1. Before March: 17–23 March 2026. This is the week before both the March spam update and the March core update.
2. Between updates: 30 April–20 May 2026. This is three weeks after the March rollout had settled and before May began, with no confirmed updates in between.
3. After May: 4–10 June 2026, once it’s available.

Use average daily clicks for each period, not totals, because the windows are different lengths. Then sort your pages into one of these patterns. This is the table I work from:

The “down in March, down again in May” group needs the most attention. One drop can be noise or competition. Two drops across back-to-back broad reassessments usually means Google’s systems keep judging those pages as weaker than the other options.

Be careful with the “up in March, down in May” group. If a page simply went back to its pre-March level, you haven’t really lost anything. You’ve lost a short-term gain.

Page-Level and Query-Level Analysis

Once you have the week after the update, work from the biggest changes down. Site-wide totals hide too much.

Start With Pages

In the Search Console Performance report, set the date comparison to 14–20 May against 4–10 June. Turn on clicks, impressions, CTR (click-through rate) and average position. Open the Pages tab and sort by click difference. Export the top 50 losers and gainers.

Then group them by template or purpose: product pages, category pages, guides, blog posts, location pages, tools. A loss focused on one page type tells you much more than a scattered one. For example, if most of your losing URLs are short blog posts while your in-depth guides held steady, you know where to look.

Then Open the Queries Behind Each Page

For each big losing page, click into it and switch to the Queries tab. You’re looking at how big each position change is, not just whether it fell.

• Small slips, such as position 3 to 5, rarely need action. Google clearly advises against changing content that already does well.
• Big, lasting drops, such as position 5 to 25, point to a real reassessment of that page.
• Impressions down but position steady often means demand fell or the results page changed. It doesn’t mean your ranking did.
• Position steady but CTR down usually means something above you changed. For example, more features or AI-generated answers may be taking attention.

Check the Context Around the Drop

Use the filters before you draw conclusions:

• Search type: check Web, Image, Video and News on their own. Core updates can also affect Discover, so check the Discover report if you get traffic there. Discover also had its own Discover core update in February.
• Country and device: a drop in only one market, or only on mobile, is a narrower problem.
• Intent mismatch: search the lost queries yourself. If the results now favour a different kind of page, such as forums, videos or product listings instead of articles, the issue may be fit, not quality.

If you need to repeat this after every update, a reusable report saves hours. I’d set one up now. Our guide to building a free SEO dashboard in Looker Studio shows how to connect Search Console and GA4, so the comparisons are a few clicks away.

Rule Out the Boring Causes

Before you blame the algorithm, check what else changed between 21 May and 2 June. In my experience, this is where a lot of “update drops” really come from. Site migrations, template changes, a new consent banner, robots.txt edits, broken canonicals or pages set to noindex by mistake can all cause drops that happen at the same time as an update. Check the Page indexing report and the Manual actions and Security issues reports too.

Planning Recovery Without Wasting Effort

Recovery planning is mostly about choosing what not to do. A rushed, site-wide rewrite can damage pages that were never the problem.

Prioritise by Impact and Pattern

Rank your affected pages by lost clicks. Then weight them by the March and May pattern above. A page that fell in both updates and used to bring real traffic or revenue goes to the top. A page that dropped a little once goes to the bottom, or off the list.

Assess Honestly Against What Ranks Now

For each priority page, compare it with the pages now ranking above it for its main queries. Ask practical questions:

• Does it answer the question faster or more fully than the others?
• Does it show real first-hand experience or expertise? Or does it sum up what others have already said?
• Is it clear who wrote it and why you should trust them?
• Would you bookmark it or recommend it to a friend?

Google’s self-assessment questions cover this in more depth. Our guide to helpful content in 2026 explains how to use them without guesswork. Google also suggests asking someone not linked to your site to review it, because owners tend to be kind to their own content.

Make Lasting Changes, Not Quick Fixes

Google advises against cosmetic changes made because you heard something “hurts SEO.” Focus on changes that help readers over time: rewriting thin sections, making the structure clearer, adding original information, updating old advice and merging pages that overlap. Google calls deleting content a last resort, for pages that can’t be saved.

Set Realistic Timelines

Google says some changes can work within days. But it can take several months for its systems to see that a site keeps producing helpful content. Sometimes the improvement only shows at the next broad core update. Google may also make smaller core changes without announcing them, so gains can come between big updates.

The rhythm I like is simple. Track your improved pages as a group in Search Console, review every four weeks, and keep a dated log of what you changed. When the next core update arrives, that log turns guesswork into evidence.

Common Mistakes to Avoid This Time

• Judging the update on rollout data. Rankings were still moving on 2 June. The week after the rollout is the one that counts.
• Ignoring March. With two core updates in eight weeks, one comparison can hide both losses and recoveries.
• Undoing April changes in a panic. If you improved pages after March and they then dropped in May, check the page data before you undo the work. The improvements may not be the cause. I see this panic often, and it usually throws away good work.
• Chasing other sites’ theories. Comments from others are useful for context, but they can’t tell you why your pages moved.
• Treating every drop as a quality problem. Changes in demand, search intent and the results page layout can all cut clicks without any reassessment of your content.

The Bottom Line

Google has said nothing beyond its standard description. Wait for the 4–10 June data and compare it with 14–20 May. Then look across both 2026 core updates to see which pages are recovering, which are newly hit and which are in steady decline. Put your effort into the last group first.

With the May 2026 core update complete, here is your next step. Once the week after the update is in Search Console, export your top losing pages and sort them into the March and May patterns above. If you set up a reusable Looker Studio dashboard now, that job will take minutes, not an afternoon.