One question I hear from site owners who were hit by a link update is: "If I clean up my links, when will my rankings come back?" Today the honest answer is "it depends, and often gradually." In 2012 the answer was much harsher: not until Google runs Penguin again.
That is why Penguin 1.1 matters. It was not a big update. Matt Cutts said it affected less than 0.1% of English searches. But it was the first proof that Penguin was not a one-time event. It was a filter Google would re-run from time to time, and each run was a chance for punished sites to recover, or for new sites to get caught.
In this post I will explain what happened on 25 May 2012, what a "data refresh" actually means, whether anyone recovered, and why this periodic model shaped link cleanup work for the next four years. For the full sequence of updates around it, see our Google algorithm updates timeline.
Penguin 1.1 at a glance
| Item | Details |
|---|---|
| Date | 25 May 2012 (US time), reported 26 May depending on time zone |
| Type | Data refresh of the Penguin webspam algorithm, with minor tweaks per Google's later report |
| Confirmed by Google? | Yes, by Matt Cutts on Twitter |
| Scale | Less than 0.1% of English searches |
| What it targeted | Same as Penguin: link schemes, keyword stuffing and other webspam |
| Who was affected | Some sites hit in April reportedly recovered; some new sites were hit |
| Other names | "Penguin 2" in Google's internal count; later renamed 1.1 by the industry |
A quick recap of the April 2012 Penguin launch
Penguin went live on 24 April 2012. In his Webmaster Central announcement, Matt Cutts said it noticeably affected about 3.1% of English queries, and about 5% of Polish queries. It targeted keyword stuffing, link schemes and sites with unusual linking patterns, including spun text with outgoing links unrelated to the content.
The launch landed weeks after Google had taken down paid blog networks such as BuildMyRank, which I cover in the 2012 blog network deindexing. Together they made spring 2012 the most painful period link builders had ever seen. My full write-up of the launch is in Google Penguin (2012): the update that changed link building.
What nobody knew in April was how Penguin would behave over time. Would it run constantly? Once? Every month like Panda? Penguin 1.1 answered that question.
Matt Cutts' "minor weather report"
On Friday 25 May 2012, Matt Cutts posted a short tweet that confirmed the first refresh. Search Engine Roundtable reported it the next morning:
“Minor weather report: We pushed 1st Penguin algo data refresh an hour ago. Affects <0.1% of English searches.”
— Matt Cutts, quoted by Search Engine Roundtable, Official: Google Penguin 1.1 Now Live
Three details in that tweet are worth noticing:
- "1st" told everyone more refreshes were planned.
- "data refresh" suggested the algorithm itself stayed the same, and only the data it worked on was updated.
- "<0.1%" was tiny compared with the 3.1% launch. Most sites saw nothing.
Barry Schwartz added that webmasters had already reported Penguin-like movements on 13, 15 and 24 May. He suspected those were live tests ahead of the refresh, but said openly he could be wrong. I would treat them the same way: plausible, unconfirmed.
What a Penguin "data refresh" actually meant
Let me explain the idea in simple terms, because it confuses people even now. Old Penguin worked like a batch job. Google would run its link spam analysis across the web, produce a list of sites that crossed a threshold, and apply a demotion to them. That list then stayed fixed until the next run.
A data refresh re-ran the same analysis on newer data. If you had removed bad links since April, the refresh could see that. If you had added more spammy links, it could see that too. An algorithm update, by contrast, changed the rules themselves.
| Term | What changes | Penguin example |
|---|---|---|
| Launch | New algorithm goes live | Penguin, 24 April 2012 |
| Data refresh | Same rules, re-run on new data | Penguin 1.1 (May 2012), 1.2 (October 2012), 2.1 (October 2013) |
| Algorithm update | Rules change ("next generation") | Penguin 2.0, 22 May 2013 |
| Real-time | Runs continuously as pages are recrawled | Penguin 4.0, September 2016 |
The line between a refresh and an update was blurrier than the tweet suggested. In June 2012, Search Engine Roundtable pointed to Google's monthly search quality report for May, which said:
“This month we rolled out a couple minor tweaks to improve signals and refresh the data used by the penguin algorithm.”
— Google search quality report, quoted by Search Engine Roundtable, Penguin 1.1 Was More Than A Refresh Says Google Report
So Penguin 1.1 was mostly a refresh, plus small signal tweaks. Schwartz also noted that Matt Cutts mentioned minor tweaks at SMX Advanced. A tweet just did not have room for that detail.
Did anyone recover in the first refresh?
Some did, according to reports at the time. Search Engine Roundtable said some webmasters noticed their sites either "recovered or newly hit" after the change. That two-way movement is exactly what a refresh would produce.
But I want to be honest about the limits. Only a month had passed since the launch. Most hit sites had barely started auditing links, and many of the links they wanted gone sat on sites whose owners never replied to removal requests. There was also no disavow tool yet. So the realistic picture is:
- Sites with a few bad links that were quickly removed had a chance to recover.
- Sites whose spammy links lived on networks that Google had already deindexed may have improved, because those links no longer existed in Google's data.
- Sites with years of paid links, directories and exact-match anchors mostly stayed down.
- A small number of new sites crossed the threshold and were hit for the first time.
Nobody published reliable recovery statistics for Penguin 1.1. Any figure you see quoted for it should be treated with suspicion.
Why the refresh cycle shaped link cleanup for four years
Penguin 1.1 set an expectation: if you were hit, you fixed your links and then waited. The problem was that the waits got longer. Here is the release history as Google confirmed it, using figures from Wikipedia's Penguin article, Search Engine Journal's Penguin history and Matt Cutts' own posts:
| Release | Date | Share of queries | Gap since previous |
|---|---|---|---|
| Penguin 1.0 | 24 April 2012 | ~3.1% (English) | — |
| Penguin 1.1 | 25 May 2012 | <0.1% (English) | ~1 month |
| Penguin 1.2 | 5 October 2012 | ~0.3% (English) | ~4 months |
| Penguin 2.0 | 22 May 2013 | ~2.3% (English-US) | ~7.5 months |
| Penguin 2.1 | 4 October 2013 | ~1% | ~4.5 months |
| Penguin 3.0 | 17 October 2014 | <1% (English) | ~1 year |
| Penguin 4.0 | 23 September 2016 | Real-time, part of core | ~2 years |
A one-month gap made Penguin look manageable. By 2014 the gap was a year. Here is a hypothetical example of what that meant for a real business. Imagine an online store hit in April 2012 that lost 40% of its organic sessions, say from 20,000 to 12,000 a month. If it finished its cleanup in August 2012, it could not see a result until the October refresh. If it missed that window, the next chance was May 2013. Eight months at 8,000 fewer sessions is 64,000 lost visits, even after the problem was fixed.
That waiting game is why Penguin recovery became such a specialist service. I explain the later chapters in Penguin 2.0 and Penguin 2.1.
The disavow tool arrives five months later
In October 2012 Google launched the disavow links tool. It let site owners upload a plain text file listing URLs or whole domains (using domain:) that Google should ignore. Google's announcement suggested the "Links to Your Site" report as "one great place to start looking for bad links," and warned "you don't want to disavow all your links!"
For Penguin victims, the tool solved half the problem: you no longer needed every webmaster to remove a link. But you still had to wait for the next refresh to see the effect. The tool arrived just after Penguin 1.2, so many sites that disavowed in late 2012 waited until May 2013 to find out whether it worked.
If you are dealing with bad links today, start with our beginner lesson on bad links, negative SEO and the disavow tool, then read our guide to toxic backlinks and the disavow tool before you upload anything.
Periodic versus real-time Penguin: what changed in 2016
The refresh model was eventually abandoned. In 2015, John Mueller said Penguin and Panda "currently are not updating the data regularly," as quoted on Wikipedia. In September 2016, Penguin 4.0 made the system real-time and part of Google's core algorithm. Wikipedia summarizes the key change: Penguin "does not demote a web site when it finds bad links" anymore. It discounts them instead, and can act on specific URLs rather than whole sites.
Google's ranking systems guide now lists Penguin as retired: "This was a system designed to combat link spam," folded into core systems in 2016. Here is how the two eras compare:
- Timing: periodic batch runs (2012–2014) versus continuous evaluation as pages are recrawled (2016 onward).
- Effect: site-wide demotion versus mostly ignoring bad links.
- Recovery: wait for the next refresh versus gradual improvement as Google recrawls.
- Disavow need: almost essential versus mainly useful for manual actions or clear paid-link history.
My post on Penguin 4.0 and real-time link devaluation explains the modern model in detail.
Questions about Penguin 1.1
Why is it called both Penguin 1.1 and Penguin 2?
Google counted launches in order, so the May 2012 refresh was its second Penguin release. The industry later preferred version numbers, calling it 1.1 and reserving 2.0 for the May 2013 algorithm update. Matt Cutts himself called 2.0 the fourth Penguin-related launch.
Was Penguin 1.1 just a data refresh?
Mostly. Matt Cutts called it a data refresh, but Google's May 2012 search quality report said it also included "a couple minor tweaks to improve signals."
How many sites recovered in Penguin 1.1?
Nobody knows. Reports at the time said some sites recovered and some were newly hit, but no reliable numbers were published.
Could I have filed a reconsideration request for Penguin?
Not usefully. Penguin was algorithmic, so reconsideration requests were for manual actions. Algorithmic recovery needed a refresh after the cleanup.
Does any of this apply in 2026?
The waiting part does not, because Penguin now runs continuously within core systems. The lesson does: links built to manipulate rankings get discounted, so cleanup only helps if you stop building them.
Turning the Penguin 1.1 lesson into a link plan today
Penguin 1.1 was small, but it taught the industry that link penalties have a rhythm and that recovery is possible when the bad links are truly gone. Today the rhythm is continuous, which is better news: fixes can show up as Google recrawls instead of months later.
If your rankings fell after a link update, or you have inherited a messy backlink profile, we can audit it with you and separate links worth keeping from links worth ignoring. Have a look at our Google penalty recovery service or get a quote and tell us what happened.
Google updates around the same time
Before this update:
- The Knowledge Graph Launch (May 2012): 'Things, Not Strings' (May 2012)
- Google Penguin (2012): The Update That Changed Link Building (Apr 2012)
After this update:
- The Pirate Update (2012): How Copyright Notices Affect Rankings (Aug 2012)
- The Exact Match Domain (EMD) Update: Why Keyword Domains Lost Their Edge (Sep 2012)
Or browse every Google update on one timeline, newest first.
Sources for this Penguin 1.1 refresh guide
- Search Engine Roundtable: Official: Google Penguin 1.1 Now Live (26 May 2012)
- Search Engine Roundtable: Penguin 1.1 Was More Than A Refresh Says Google Report (June 2012)
- Google Webmaster Central Blog: Another step to reward high-quality sites (April 2012)
- Google Webmaster Central Blog: A new tool to disavow links (October 2012)
- Matt Cutts: Penguin 2.0 rolled out today (May 2013)
- Search Engine Journal: Google Penguin algorithm update history
- Wikipedia: Google Penguin
- Wikipedia: Timeline of Google Search
- Google Search Central: A guide to Google Search ranking systems
