Many site owners I talk to still believe that Panda was a one-time event from 2011 that only punished obvious content farms. That is not quite right. Panda kept coming back for years, and one of its biggest returns was Panda 4.0 in May 2014. It hit some very large, very well-known brands, and it changed how a lot of people thought about "quality."
In this post, let me explain what Panda 4.0 was, what Google actually said, who lost (and who gained), and what was only guessed at the time. I will also show you why the lessons still matter in 2026, even though Panda no longer exists as a separate system.
If you want to see where this update sits among all the others, our timeline of every Google algorithm update puts it in context. For the full Panda story from 2011 onward, read our complete guide to the Panda update. Here I will stay focused on the May 2014 version.
Panda 4.0 at a glance
| Detail | What we know |
|---|---|
| Start date | May 20, 2014 (announced by Matt Cutts on Twitter that day) |
| Type | Named algorithm update, widely seen as a new version of Panda, not just a data refresh |
| Confirmed by Google? | Yes |
| Reported scale | About 7.5% of English-language queries |
| Main target | Low-quality, thin, duplicate and aggregated content |
| Notable losers | eBay, Ask.com, RetailMeNot, celebrity gossip sites, press release and aggregator sites (per third-party data) |
| Overlapping update | Payday Loan 2.0, confirmed for the weekend before (around May 16–18, 2014) |
| Status today | Retired as a named system; Panda became part of Google's core ranking systems in 2015 |
How Matt Cutts announced Panda 4.0
Panda 4.0 was announced in the simplest possible way: a tweet. Matt Cutts, who led Google's webspam team at the time, posted it on May 20, 2014. The Register reported his words exactly:
“Google is rolling out our Panda 4.0 update starting today.”
— Matt Cutts, as quoted by The Register, Panda-monium as Google updates key search algorithm
That was almost all Google said in public. There was no long blog post and no list of new signals. The same Register article noted that Google had not explained what had changed inside Panda.
Why does the name matter? By 2014, Panda had been running as quiet monthly refreshes for some time. Google had stopped announcing most of them. So when Cutts gave this one a full version number, 4.0, people read it as a signal that something bigger had changed. Search Engine Journal's Panda history says most of the industry believed it was a real algorithm update rather than a simple data refresh.
The "softer Panda" promise from March 2014
There was a lead-up to this. In March 2014, at the SMX West conference, Cutts said Google was working on a "next generation" Panda that would "soften" the algorithm. According to Search Engine Roundtable's report from March 14, 2014, the aim was to help small businesses that might have been caught by Panda.
Many people assumed Panda 4.0 was that softer version. Google never said so directly. So I treat this as a reasonable link, not a confirmed fact.
What 7.5% of queries actually means
The figure most people remember is 7.5% of English queries. Search Engine Land reported it, and later sources repeated it. Wikipedia's Timeline of Google Search also lists Panda 4.0 at about 7.5%.
That number is easy to misread. It does not mean 7.5% of websites lost traffic. It means that the results for roughly 7.5 out of every 100 English searches changed in a way a normal user might notice.
To compare it with its neighbors:
| Update | Date | Reported share of queries |
|---|---|---|
| Original Panda | February 2011 | Almost 12% (as reported at the time) |
| Panda 4.0 | May 20, 2014 | About 7.5% of English queries |
| Panda 4.1 | September 2014 | 3–5%, depending on locale |
| Panda 4.2 | July 2015 | 2–3% of English queries |
So Panda 4.0 was the largest Panda change since the first one. You can read about the two later versions in my posts on Panda 4.1 in September 2014 and the slow-rolling Panda 4.2.
Who lost: eBay, Ask.com and the aggregators
Google did not name any sites. Everything we know about winners and losers comes from third-party visibility tools. The most widely shared data came from Searchmetrics, which ran a quick preliminary analysis. Search Engine Journal summarized it on May 21, 2014. Here are some of the estimated losses:
- ask.com: about −50%
- ebay.com: about −33%
- retailmenot.com: about −33%
- biography.com and history.com: about −33% each
- examiner.com and starpulse.com: about −50% each
- aceshowbiz.com: about −75%
- yellowpages.com: about −20%
Please read these as rough estimates. Searchmetrics measured "visibility" across the keywords it tracked, not real traffic. It also said the update might not have finished rolling out.
Even so, the pattern was clear enough. Many losers published syndicated, duplicate or aggregated content. Gossip sites, coupon sites, directories and press release sites were common on the lists. Wikipedia's Panda page, citing Forbes, sums up the targets as content aggregators, rumor and gossip news sites, and price comparison sites.
Why eBay became the headline
eBay is not a small, spammy site. That is exactly why its drop got so much attention. Search Engine Land ran a piece asking whether eBay was a big loser in Panda 4.0. The usual theory was that eBay had huge numbers of category and search-style pages, many with little unique text on them, and that Panda judged those pages weak.
Google never explained eBay's drop. So the honest summary is: eBay lost a lot of visibility at the same time as Panda 4.0, and thin category pages were the most popular explanation. Nobody outside Google can prove more than that.
Who gained, and what that tells us
The same Searchmetrics list showed some big gainers. A few examples from the SEJ summary:
- glassdoor.com: about +100%
- emedicinehealth.com and medterms.com: about +500% each
- myrecipes.com: about +250%
- consumeraffairs.com: about +100%
- buzzfeed.com: about +25%
Interestingly, some aggregators and coupon sites were on the winning side too. Couponcabin.com gained while RetailMeNot lost. SEJ noted that this made a firm conclusion difficult.
That is an important lesson. Panda never punished a business model. It judged the quality of the actual pages. Two coupon sites could go in opposite directions because one had more useful, original pages than the other.
Was Panda 4.0 gentler on small sites?
This is where reporting and reality get mixed up. Because of the "softer Panda" comment in March, many people expected small sites to do better. Some did. Glenn Gabe, an SEO consultant who tracked Panda closely, wrote that he heard from previously hit sites seeing gains of 200% or more after May 20, 2014.
But small sites were also hit hard. In his case study published on August 13, 2014, Gabe described a small affiliate site with fewer than 60 indexed pages that lost 86% of its Google organic traffic overnight on May 20. The problems he listed included:
- A large amount of thin affiliate content, with many links to Amazon
- Ads styled to look like content
- A basic design with weak trust signals and some broken HTML showing
- An author profile and photo that looked fake
So "softer" did not mean "safe." My reading is that Panda 4.0 got better at telling good small sites from thin small sites. The good ones had a better chance. The thin ones were still in trouble.
The Payday Loan 2.0 overlap that confused everyone
Panda 4.0 did not arrive alone. A few days before, Google released Payday Loan 2.0, an update aimed at very spammy queries such as payday loans and similar niches. Search Engine Land's SearchCap from May 21, 2014 quoted a Google spokesperson saying, "Over the weekend we began rolling out a new algorithmic update."
The two updates had different targets:
| Panda 4.0 | Payday Loan 2.0 | |
|---|---|---|
| Date | May 20, 2014 | Weekend of about May 16–18, 2014 |
| Focus | Content quality across all kinds of sites | Spammy queries and spam tactics in specific niches |
| Typical victims | Thin, duplicate and aggregated pages | Spam sites in loans, adult and similar queries |
| Fix | Improve or remove low-quality pages | Stop spam tactics; clean up links |
If your traffic dropped that week, it was hard to know which update caused it. The usual approach was to look at the exact day and at which keywords fell. A drop on May 17 in a payday-style niche pointed one way. A drop on May 20 across many informational pages pointed the other. I explain this kind of detective work in my guide to untangling overlapping updates, and the Payday story itself is in our post on the Payday Loan updates of 2013–2014.
How sites recovered from Panda 4.0
Recovery from Panda was slow, because the system only re-scored sites when Google ran it again. Here is the process that worked for most sites in 2014, and that I would still follow for a quality problem today:
- Confirm the timing. Compare Google organic traffic before and after May 20, 2014. A sharp drop on that day, not a slow slide, pointed to Panda.
- Find the weak sections. Panda judged the site as a whole, but a few sections usually caused most of the problem: tag pages, thin category pages, syndicated articles, or auto-generated pages.
- Decide page by page. For each weak page, choose to improve it, merge it into a stronger page, noindex it, or remove it.
- Fix the user experience. Ads that looked like content, aggressive popups and broken layouts were common on hit sites.
- Wait for a re-run. Nothing changed until Panda refreshed again. Gabe's case study showed a temporary bump in July 2014 that vanished five days later, because the site had not changed anything.
A worked example (hypothetical)
Let me use a simple example. Imagine a recipe site with 4,000 indexed URLs in 2014:
- 1,200 full recipe pages with original photos and steps
- 1,800 tag pages such as "easy chicken dinner" that just list 3–5 recipes
- 1,000 syndicated articles copied from a partner site
Only 30% of the indexed pages are original, useful content. If this site lost 40% of its traffic on May 20, I would noindex most tag pages, remove or canonicalize the syndicated articles, and put the effort into the 1,200 real recipes. After the cleanup, the useful share goes from 30% to almost 100%. That is the kind of shift that gave sites a chance at the next Panda run. It did not guarantee anything.
For the hands-on version of this audit, see my guide on how to identify and fix thin content.
What Panda 4.0 still teaches us in 2026
Panda is gone as a separate system. Google's guide to its ranking systems lists it under retired systems and says it "evolved and became part of our core ranking systems in 2015." You can read more in our post on Google's retired ranking systems.
But the ideas behind it live on in core updates. Here is what I take from Panda 4.0:
- Big brands are not immune. eBay and Ask.com proved that size does not protect weak pages.
- Your weakest pages count. Thin sections can drag down a whole site.
- Copying is not adding. Syndicated and aggregated pages need something extra to deserve a ranking.
- Short-term bumps can fool you. Real recovery needs real change.
If you were hit by a recent core update, the playbook is similar. Our guide to running a content quality audit after a core update walks through it step by step. And if you are new to how updates work in general, the Learn SEO lesson on algorithm updates is a good primer.
Questions people still ask about Panda 4.0
When exactly did Panda 4.0 roll out?
Matt Cutts announced it on May 20, 2014, saying it was rolling out "starting today." Some trackers list May 19 or May 20, depending on time zone and when they noticed changes.
How many searches did Panda 4.0 affect?
About 7.5% of English-language queries, according to figures reported at the time. That made it the biggest Panda change since the original 2011 launch.
Was eBay really penalized by Panda 4.0?
eBay lost about a third of its visibility in Searchmetrics data that week. Google never confirmed the reason, so whether it was Panda alone or something else is still unproven.
Is Panda 4.0 the same as Payday Loan 2.0?
No. Payday Loan 2.0 rolled out a few days earlier and targeted spammy queries. Panda 4.0 targeted low-quality content on all kinds of sites.
Can a site still be "hit by Panda" today?
Not by a separate Panda update. Panda's ideas are now part of Google's core ranking systems, so quality problems show up during core updates instead.
Where to start if quality is holding your site back
In brief, Panda 4.0 was Google's loud reminder that thin, copied and aggregated pages are a liability, whatever the size of your brand. The update itself is history. The question it asked, "Is this page actually worth showing?", is still the question every core update asks.
If you are not sure which of your pages pass that test, we can help. Our team runs content quality audits and recovery plans as part of our SEO services. Get a quote and tell us what changed in your traffic, and we will show you where I would start.
Google updates around the same time
Before this update:
- Penguin 2.1 (October 2013): The Refresh That Hit 1% of Searches (Oct 2013)
- Google In-Depth Articles (August 2013): A SERP Block for Long-Form Content (Aug 2013)
After this update:
- Google’s Pigeon Update (Jul 2014)
- HTTPS as a Ranking Signal: What Google's 2014 Announcement Really Meant (Aug 2014)
Or browse every Google update on one timeline, newest first.
Sources for this Panda 4.0 guide
- The Register: Panda-monium as Google updates key search algorithm (May 21, 2014)
- Search Engine Journal: Who was hit hardest by Panda 4.0? (May 21, 2014)
- Search Engine Journal: Google Panda update history
- Search Engine Land: SearchCap – Panda 4.0 live, Payday Loan algo released (May 21, 2014)
- Search Engine Roundtable: Matt Cutts on a new softer Panda (March 14, 2014)
- Glenn Gabe (GSQi): Panda 4.0 temporary recovery case study (August 13, 2014)
- Wikipedia: Google Panda
- Wikipedia: Timeline of Google Search
- Google Search Central: A guide to Google Search ranking systems
