The latest Google updates are not as powerful, with data showing less than optimal performance since the updates were introduced. This is a sign that companies need to be cautious when investing in online advertising– and also shows how quickly things can change on the internet.
The “google algorithm update 2022” is a question that has been asked before. Google’s updates are becoming less powerful as the company releases more core updates.
Google’s fundamental upgrades, it goes without saying, may make or destroy a website. But what if the core updates were to become more stable? What does this imply for websites and the SEOs that maintain them?
The present “core update period” has resulted in some of the most significant algorithmic adjustments in the history of the SEO business. It’s possible to argue that the improvements have altered the entire basis of the industry’s discussion of page and site quality (look no further than the renewed focus on E-A-T). The SEO industry nibbles its nails in eager anticipation with each new roll-out, since previous modifications have thrown rankings into a tizzy.
But, at least to the same degree, is it still true?
Are today’s Google core upgrades as powerful as previous juggernauts, such as The Medical News in 2018?
I’ve been following these adjustments since March of 2018, and my sense is that the changes Google has made lately are less significant.
I looked at the amounts of rank volatility in every core update since the notorious Medic Update to find out.
This is what I discovered…
Core Update Analysis: What Was Analyzed & What Was Found
Despite what you may believe, this is the most crucial portion of the research. It’s impossible to have enough context to appropriately apply anything without first knowing what was evaluated and how it was analyzed.
It’s one of my pet peeves when SEO studies don’t adequately communicate their limitations. Understanding a study’s boundaries and conceptual “borders” is critical to making appropriate use of it.
With that out of the way, let’s have a look at the data I evaluated and some of the top-level results.
(Of course, you may skip to the following H2 if you want to jump straight in — which I don’t encourage.)
Concerning the Analyzed Data
To begin with, this research only extends back as far as the “current” core revisions. Core updates have existed for a long time, long before they were dubbed “things.” Google announced broad core algorithm modifications multiple times a year long before it started disclosing them in March of 2018.
To that end, I started my investigation with the August 2018 Medic Update, rather than the March 2018 Core Update. The Medical News is virtually emblematic of the present age of SEO, therefore I did it. Many industry thought leaders saw new trends surrounding how Google profiles site and page quality after the upgrade. Your Money Your Life (YMYL) and E-A-T were brought to light by The Medical News. So, despite the fact that Google’s Danny Sullivan certified the March 2018 update as the first, I decided to start with The Medical News.
The entire list of core updates examined in this investigation is as follows:
- The Medical News
- The Core Update for March 2019
- The Core Update for September 2019
- The Core Update for January 2020
- The Core Update for May 2020
- The Core Update for December 2020
- The Core Update for June 2021
- The Core Update for July 2021
We documented the volatility levels as indicated in the Webinomy Sensor for each of these updates. The issue is that the Sensor assigns a score to the degree of volatility witnessed each day, which looks like this:
The score you see here is based on the volatility of the previous period. Meaning, even a little shift might be considered “volatility” if you had an extended period of algorithmic calm.
In other words, you cannot just say, “Well, The Medical News scored a 9.9 on the Sensor, and The Core Update for March 2019 scored a 9.7; so therefore The Medical News was ‘bigger’ (for the record those are not the real Sensor numbers for either update).”
To get around this, we calculated the percentage difference between the time of stability before each update and the period of volatility indicated by each update. In this example, the number (i.e., a volatility level of 9/10) is less important than the amount of volatility that rose compared to the time of stability that preceding it. So, if the levels increased from 2/10 to 4/10 during one update then from 5/10 to 10/10 during the next, both would have a 100 percent increase in volatility.
We estimated the increases by comparing the three days of relative quiet before the update to the first three days that reflected the upgrade itself, in order to better portray the general stability of each update. This is due to the fact that most updates take more than one day to go out.
Not every update is the same these days. Some updates are quite volatile for 3–4 days, while others are volatile on the first day but not so much on the second, and yet others might be really volatile for a long time. This analysis of how updates may be distributed was not done on a per-update basis. Across the board, the same process was used.
Finally, the total study of the updates includes all of the industries that make up the data in the Sensor. For each update, around 25K keywords were examined.
There were a few times when I looked at the trends within each specialty business. I didn’t employ all industries; instead, I concentrated on a small number of specializations. As I go through my analysis, I’ll show you when and why this happened.
Findings from the Highest Levels of Data
Now that the context of the data has been properly accounted for, here are some of the top-level data highlights:
- The average increase in rank volatility noticed with a core upgrade is 59.77 percent.
- Only two of the core upgrades examined (The Medic and September 2019 Core Updates) resulted in 100% or more increases in volatility.
- The average level of volatility seen during an update has decreased by 51.7% since The Core Update for January 2020.
- Volatility isn’t ubiquitous yet. Rank volatility in some of the most well-known businesses may still be quite unpredictable, despite what seems to be a reduction in overall rank swings.
Naturally, all of this information must be explored, clarified, and qualified, which is precisely what I want to do!
The Evidence Suggests a Decrease in Core Update Potency
The most logical place to start is, of course, with total volatility over time. For this, we pulled the rank volatility increases for each of the core updates that Google released, starting with The Medical News and ending with the most recent update: The Core Update for July 2021.
The rank volatility increases seen during each core update from The Medical News and on
As you can see, there is a clear downward trend in the rank volatility increases displayed by the core updates over time. Even excluding the only two updates to show rank volatility increases of 100% or more (i.e., The Medical News and The Core Update for September 2019), the trend remains true. That is, starting with The Core Update for December 2020, there has been a trend where the volatility increases fall significantly below 50%.
However, there is a bit of a caveat to the trend since Google split The Core Update for June 2021 into two parts (the first being reflected as The Core Update for June 2021 and the second being The Core Update for July 2021). The question is what would have happened had Google released both the June and July updates together as was initially intended?
Still, there is no reason to question the 40% volatility increase seen during The Core Update for December 2020. This update still fell well under the 59.77% volatility increase average seen across all of the updates and is clearly a divergence from the preexisting volatility trend
Even so, there are a few options for parsing this.
On the one hand, you could look to make a cut-off with The Core Update for September 2019 as it was the last time rank volatility increased by 100% or more. Conversely, you could argue that the real shift in the data trends begins with The Core Update for December 2020.
Personally, I think The Core Update for September 2019 represents a good line of demarcation as it would seem to represent the end of the “monstrously large” core updates. Though I do see the other side of the coin as well.
To that end, here’s how the statistics play out when parsed in different ways:
Based on different groups of core updates, the rank volatility rises.
The volatility patterns tend to suggest to the updates being less and less powerful no matter how you slice it, whether you group the updates by January 2020 and on or by December 2020 and on.
Is There a Reduction in the Power of Core Algorithm Updates?
Is this the start of a new trend, or is everything just coincidence?
Because I don’t work for Google, I can’t tell with sure if the data patterns are intentional or not. However, it’s worth noting that it’s been over two years since we’ve seen a core upgrade with a 100 percent rise in volatility. That is a considerable length of time.
As far as the decrease in volatility seen since The Core Update for December 2020, my intuition — based on studying the updates for years and in seeing this data — is that the updates have become a bit less powerful over time.
Later in this essay, I’ll explain why I believe so, in order to provide further evidence for my theory.
Let’s keep digging through the facts for the time being.
The level of Rank Volatility in the Industry Remains High
Aside from the general rank volatility tendencies shown by the core updates, one of the things I was curious in was how the different specialist sectors had performed over time. Is it true that certain sectors are more volatile than others, or vice versa? Are there any new patterns at the industry level that correspond to the overall data trends?
I looked at the following industries to do so:
- Literature and Books
- Purchasing Real Estate
- Garden and Home
- Food & Beverage
There are various reasons why I chose these niches over others. Fundamentally, it came down to two criteria. First, if the industry was YMYL (Health, Finance, and technically Shopping as well as Garden and Home — though retail keywords are not the same in terms of their YMYL status as far as the core updates are concerned). Secondly, the nature of the keywords found within the industry. For example, Literature and Books include a lot of entity-based keywords, whereas Science contains many informational keywords.
One thing I found when delving through the industry-level statistics was that several of the niches I decided to deal with had higher-than-average levels of volatility. That is, their volatility spikes were often much greater than the industry average. (Of course, there have been periods and sectors that have performed below average.)
As a result, I estimated the percentage increase/decrease in each industry in comparison to the baseline.
Across all core updates, here’s how much greater/lesser the rank volatility for the industries was relative to the baseline:
As is evident above, the Heath, Purchasing Real Estate, and Travel niches are incredibly more volatile than the average. Overall, all of the niches analyzed have been 57.34% more volatile than the overall average across all core updates. (For the record, I would like to dive into the Purchasing Real Estate niche before I draw any conclusions about it, as I wonder if listing changes are behind a substantial amount of volatility here.)
Is it true that the more volatile sectors are getting closer to the general average volatility as time goes on?
Based on what we’ve previously learned about the Medic and September 2019 updates being especially volatile, we should anticipate industry-wide volatility to be higher than typical when looking at these updates.
While these industries are normally more volatile than the average, the first three core updates evaluated showed that they were around 14 percent more volatile than “ordinary.” Given how turbulent things were in August 2018 and September 2019, this is quite understandable.
When I looked at the remaining updates (all core updates from January 2020 onwards), it seemed that the industry had stabilized, with all niches performing closer to the general average.
When compared to the statistics indicated by the first three core updates evaluated, this data would indicate an 18% improvement in stability relative to the baseline. Furthermore, when compared to the whole set of core updates, this would result in an 8% gain in stability compared to the overall average.
However, if we look at just the updates from The Core Update for December 2020 and on, the numbers tell a bit of a different story:
When looking at the updates from December 2020 and on, the individual industries shown above were not closer to average volatility (as would be expected)
When you effectively remove The Core Update for January 2020 from the mix, the industries represented are 59.18% more volatile than the dataset overall, which is a tad higher than the average seen across all updates (i.e., 57.34%).
To put it another way, this graph shows how much more volatile these niches were in comparison to the total data, based on various groups of the updates.
When looking at updates from December 2020 and on, the individual sectors investigated did not show volatility increases that were closer to the overall average, contrary to the general trend.
The general tendency toward less and less volatility did not remain true, as can be shown. If it had, the niches studied would not have been 59.18 percent more volatile than the general average from December 2020 onwards – a statistic that is larger than when January 2020 is taken into account.
The niches aren’t always in line with the broader trend of lower volatility. In terms of general volatility, there are still niches that are both very volatile and quite stable. It is not the case that the niches’ rank movement has gotten more subdued in comparison to the general average. The drop in rank volatility isn’t consistent.
This becomes clearer when we look at the trends within some of the specialty sectors.
Patterns in Niche Industries Across the Core Updates
Data is only as good as the information it contains. Instead of merely discussing industry-level patterns, I’d want to take a look at a couple of them to see how their core update volatility has changed over time.
To accomplish so, I’ll compare the levels of volatility seen within a certain industry throughout the course of the core updates to the overall average volatility observed.
Trends in the Health Care Industry
There’s no better place to start our per industry analysis than with the Health industry. It’s the only industry to have a core update named after it (i.e., The Medical News). tI’s the most “YMYL” of all the YMYL niches (as bad advice here can actually kill people), and it sits at the center of the E-A-T conversation.
Simply by looking at the statistics, it’s clear that the general volatility patterns do not apply at the industry level.
The health niche was just unbelievably volatile through the first three core updates I analyzed. For both the Medic and The Core Update for September 2019s, the Health niche displayed significantly more than double the levels of volatility recorded overall.
For our purposes and from a trends perspective, the Health industry was 214% more volatile than the average industry during The Core Update for December 2020. Moreover, it displayed over 100% increases in volatility as well. Remember, the December Core Update was the first update to record volatility levels lower than 50%. Yet, the Health industry was still incredibly shaken up by the update.
Parenthetically, the Finance industry (which I might consider to be the Health niche’s YMYL sibling) showed a very different data trend. Surprisingly, the Finance niche never produced levels of volatility that exceeded the overall average by more than 100%. The highest divergence from the overall average came with The Core Update for September 2019, where the niche was 60% more volatile than average.
That isn’t to imply that there wasn’t a lot of volatility. However, the Finance niche has 28.46 percent greater volatility than the average across all updates, which is nothing compared to the Health niche’s 114.64 percent.
Trends in the Travel Industry
Another notable example of volatility not being constant or following the general trend of reduced rank change is the travel sector. During five of the eight updates I looked at, the Travel industry’s rank volatility was higher than the general average.
However, two of the times that it has not reached such levels of volatility came during the June and July 2021 Core Updates. The third instance was during The Core Update for January 2020, where the niche was just 41% more volatile than average.
Still, during The Core Update for July 2021, the industry was far more volatile than average (87% more to be specific).
The same inconsistency can be found within the Literature and Books industry, which again I chose because it correlates very heavily with entities. There are massive swings between the industry displaying above average as well as significantly below-average volatility.
The argument is that the broad trends indicating lower volatility do not apply uniformly. Despite the broad patterns, it’s completely feasible that an update may reveal significantly greater volatility inside a certain specialty. If there is less volatility, it may not apply to all markets.
Consistency in the Niche Industry Across the Core Updates
Specific industries act differently, as shown by the statistics. The modifications to the core have no common pattern. That isn’t to suggest that there hasn’t been consistency across industries throughout time.
For example, to the exclusion of The Core Update for July 2021, the Garden and Home industry has always displayed less rank volatility than the average niche.
The Science business (which I included since it accurately represents informative content) has always shown “a little” more volatility than the norm. In reality, with the exception of two cases, it’s between 30% and 40% higher on average.
The same is true for the Shopping niche, which outside of The Core Update for June 2021, is generally less volatile than the average and generally in the 20% to 30% range. As an aside, this is clear testimony to the fact that despite retail Shopping being YMYL, the algorithm treats it differently than something like health content. This makes total sense as Shopping sites are YMYL, not because of the content per se, but because checkout requires a credit card.
The idea is that discrepancies among industries are unique to each business since each follows its own path.
Why Could Core Updates Be Less Effective Than Before?
Something had shifted. The algorithm has changed dramatically in the contemporary age of core updates. That change is undoubtedly a hot topic of discussion. In my opinion, it’s been about Google growing much better at qualitatively comprehending the content on a page and site, which corresponds to significant advances in natural language processing and other areas.
There’s been a distinct change in what Google can accomplish when it comes to comprehending a site, according to myself and many others in the business. A lot of what I’ve observed has to do with Google’s ability to profile a site’s identity and, as a result, comprehend what comes within the scope of the site and what doesn’t (and if the site lives up to its purpose). I’ve seen revisions select out text that communicates the incorrect “tone,” as well as examples where relevance became significantly more complex (resulting in the extinction of several “ultimate manuals”).
However you want to understand it, clearly something has changed since circa The Medical News.
It’s then quite logical that the first applications of this change be the most impactful. (Which, Looking at the data, they have been in the form of The Medical News and The Core Update for September 2019.) With the first application of a “new ability,” there is just simply more to fix. As time goes by and Google has applied its new abilities to more sites and pages in all-new ways, the overall impact should, in theory, decrease exponentially. (This is not a new pattern and has been seen with things like RankBrain in the past.)
Simply said, since Google performs the “repair” on a more frequent basis, the necessary alterations to the ranks are less dramatic.
That isn’t to imply that Google isn’t working to improve its capabilities. Google, of course, is.
What I’m saying is that 2018 marks a fundamental change in what Google can and cannot accomplish. This fundamental shift may and has been adjusted throughout time, but the effect of these changes will settle over time until another fundamental shift occurs (more or less).
While further interactions between fresh core updates would assist establish if the present trend is a true pattern, there is already a lot of rationale behind what we’ve seen so far.
Is it the Quiet Before the Storm?
I believe there is a widespread underemphasis on understanding the present SERP ecology. If the data confirms what we’ve seen so far, it might suggest we’re in the midst of a rare moment of SEO quiet. (I’d suggest to appreciate the peace, but that’s hilarious in many respects since “stability” is very subjective and meaningless to a site affected by an upgrade.)
However, given the current state of affairs, any relative quiet is unlikely to stay long. MUM has already been tested in the real world by Google, and it has the potential to drastically alter the SERP. Who knows what Google has in its sleeve as it continues to experiment with massively complicated learning algorithms.
The “google algorithm update september 2021” is a question that has been asked by many. The answer to the question is no, but it seems that Google’s core updates are becoming less powerful.
Frequently Asked Questions
When was the last Google core update?
A: The last Google core update was on February 8th, 2019.
How often does Google do a core update?
A: Google does not update their core software frequently. When a new version comes out, you need to go into your settings and download the latest updates manually.
Did Google change their algorithm again?
A: Google has not changed their algorithm.
- latest google algorithm update 2021
- google algorithm update october 2021
- google algorithm update november 2021
- google core update
- why has google changed today