Algorithm update of March 12 – Medic update?

For a few days now, SEO specialists have been discussing the most recent Google algorithm update. It’s worth emphasizing that this is a core update, which means a change in the core search engine algorithm. Google have a habit of notifying such changes in advance, and so they did this time as well.

Experts gave it the working title ‘Florida’. Many also claimed that it’s Medic update #2 (with the first one implemented in August 2018) for YMYL pages.

The update considerably affected Polish search results. In SERP Weather, we’ve been observing big changes.

It’s evident that the first shifts in search results began in February, and since March 13 we’ve been experiencing quite extensive changes (just to remind you, the change index.  is a metric showing how many websites underwent a change in visibility by at least 20% over a week. The index presents an extent of shifts in search results).

I’ll try to answer some questions in this article:

• Is it actually Medic update #2 (for YMYL pages)?
• Who loses the most?
• What were the reasons behind the drops?

With the last update to Senuto, we’ve been able to implement the classification of websites and keywords. All keywords and websites are classified into 403 categories (IAB standard), which allows us to analyze thematic categories of the websites* instead of the entire index.

*applies only to Polish market.

Was it the end of the Medic update?

To verify that, I analyzed the following:

• A number of websites in the category which had had 200–5000 keywords in TOP10 and later lost at least 15% of their keywords – I meant somewhat visible websites that weren’t big.
• A number of websites in the category which had had over 5000 keywords in TOP10 and later lost at least 15% of their keywords – I meant only those websites which noted dramatic drops. SEO experts usually use examples of big websites; I wanted to make a distinction here.

Changes among websites from TOP10 >200 and <5000

The table below presents an analysis of websites which had 200–5000 keywords in TOP10 before the update, but lost around 15% of their visibility after the update.

Let me explain what individual columns represent:

• Category: a category into which the websites are classified. There’re 403 categories in Senuto in overall, but not in this analysis because most of them wouldn’t meet the pre-defined criteria.
• Websites affected: a number of websites in the category that meet the pre-defined criteria.
• Websites in the category: a number of websites classified into the category, the visibility of which is between 200 and 5000 keywords in TOP10.
• Impact: a percentage revealing how many websites of them all have been affected by the update and lost at least 15% of their keywords.
• Total drops in TOP10: a total number of keywords in TOP10 by which the visibility of all the websites affected by the update decreased.
• Pre-update total keywords in TOP10:  a total number of TOP10 keywords of the websites affected by the update before its roll-out.
• Average drop %: an average % of visibility lost by the websites that noted a drop.
• Average drop in TOP10: an average number of keywords in TOP10 lost by the websites affected by a drop.

Table 1 – not found

Conclusions drawn from the analysis are pretty obvious. As far as small websites are concerned, it wasn’t a Medic update. Health categories weren’t the ones to note the most substantial drops. Let’s see how it looks like in the case of highly visible websites, meaning those which are supposed to be rich in content.

Changes among websites from TOP10 >5000

Here I analyzed websites with more than 5000 keywords in TOP10; they also noted a considerable 15% drop in visibility.

You may observe that a majority of big websites which noted a drop are classified into Health categories. When their visibility was on the decrease, they lost 23% of their keywords on average, which translated into an average number of over 18 thousand keywords. There’re also such categories as Pregnancy and Upbringing which were considerably impacted by the update as well – nearly 7% of the websites that met the pre-defined criteria.

Who’s capitalized on the situation?

When someone loses, there’s someone who gains. Let’s reverse the situation and analyze the same two scenarios, yet in reverse – websites that gained at least 15% keywords in TOP10.

Changes among websites from TOP10 >200 and <5000

Table 3 = not found

Here things get very interesting. We noticed that only big websites had noted a drop in Health category. In this case, you can observe that their place was taken by multiple small websites; they gained even 35% of keywords on average, which translated into an average increase of 222 keywords in TOP10 per website. In total, they gained 111 thousand keywords in TOP10, that is over 30% of the drop noted by the big websites from the same category. The remainder of the keywords was picked up by tiny websites visible for a few dozen keywords.

An example of such a website is czytelniamedyczna.pl, which increased its visibility post-update by 55%. What’s interesting, this website (like many others) had previously noted a visibility drop with the August 2018 update (Medic #1); I’m going to cover it later in this article.

Changes among websites from TOP10 >5000

The greatest shifts here were also noted in Health category. However, it needs to be mentioned that websites on the increase weren’t as visible as the ones on the decrease. If you divide the number of websites by the former total of keywords in TOP10, you’ll see that each of the websites that noted an increase was visible for 10 thousand keywords in TOP10 on average, whereas those websites that noted a drop were visible for 124 thousand keywords on average. This proves that small and medium (in this case) websites took place of big websites.

Winners and losers

We analyzed the last update by category, now let’s do it by website. Let me show you who’ve won the most and who’ve lost the most on a few specific examples.

Drops in visibility of big websites – Health

When it comes to big websites, Health category exhibited the most drastic drops. Let’s take a look at the most significant decreases.

This table shows additional data, explained below:

• TOP10 on July 30, 2018: Medic #1 update was implemented on August 1, 2018. I provided each website with its TOP10 figures before that update to check how its visibility changed from then.
• Change when compared to Medic #1: current visibility when compared to July 30, 2018.
• Change by Medic #1 before the drop: an increase in the number of keywords in TOP10 before the drop when compared to the number of keywords prior to Medic #1.

The following conclusions can be arrived at:

• In fact, when compared to July 30, 2018, all the websites became more visible (despite the present drop). Only few of them are less visible than before.
• The websites noted similar drops in visibility in terms of keyword loss percentage. An average drop is 19% and a standard deviation is merely 6%.
• If Google made only adjustments when compared to Medic #1, then Change when compared to Medic #1 should be 100% (the website went back to its pre-update condition). As you can see, it’s true only in some cases (e.g. poradnikzdrowie.pl, zdrowie.tvn.pl). However, there’s no correlation between those two – for instance, poradnikzdrowie.pl capitalized on Medic #1 and zdrowie.tvn.pl lost on it.

Drops among small and medium websites – Health

I analyzed small and medium websites separately. Below are listed 10 websites in the category that noted the biggest drops.

Only few websites among those that dropped the most went back to their condition before Medic #1. Most of them have been getting ranked lower for quite a long time now and the recent update only sped this process up.

Global drops

Here the hugest drops without any category filters are taken into account.

As you can observe, 5 out of first 6 websites that noted a drop are classified into Health category. Those were the most spectacular drops, which I’ve mentioned earlier. You’ll find here a couple of websites which regained their visibility level from before Medic #1.

There’re also websites which didn’t manage to restore their condition from before July 31, 2018, but they’ve been on the decrease for a long time and the update only accelerated it. What’s interesting, polki.pl, zdrowie.gazeta.pl and wylecz.to – in spite of considerable drops – are still more visible than prior to Medic #1 – they were growing rapidly over that period of time.

Increases among big websites – Health

Here it’s no secret that Google made adjustments after August 1, 2018. Virtually all the largest websites that gained on this update, lost on the previous one. On the other hand, only few websites were able to get back their full visibility. This means that Google made an adjustment, yet it didn’t concern all the keywords. It’s as if the formerly used algorithm was improved.

Increases among small and medium websites – Health

In this case, it’s even clearer that websites which capitalized on this update are those (all of them) affected by the update of August 1. In principle, only 2 out of 10 regained their full visibility after Medic #1. Therefore, it’s only a mere adjustment this time as well.

Global increases

This is also an interesting situation. The websites on the increase include several sites of Onet, which noted a drop in October 2018 and which have begun to regain some of its visibility now. There’s also rossmann.pl which wasn’t migrated well enough, and the update somehow sped up its return to search results. Basically, only two websites from this list were affected by Medic #1; increases were noted not only in Health category (only 4 out of 10).

A study of 2 cases

We analyzed the biggest winners and losers, now let’s have a look at two specific cases and try to find the cause of their drop.

Polki.pl

This is a website which was second to lose the most on this update, right after books.google.pl. When you look at the visibility chart, you can notice a drastic decrease in the number of keywords in TOP3 and TOP10. As far as a percentage loss of visibility is concerned, it doesn’t stand out when compared to others, but it definitely lost the most in terms of figures.

You may observe a drop of over 120 thousand keywords in TOP10 and over 70 thousand keywords in TOP3. I started with an analysis of the sections of polki.pl – the website which discusses health topics, but not only; I wanted to be sure whether it had been those topics precisely that had suffered post-update.

The screenshot below contains certain information:

• Keywords count: it shows a number of keywords for which a given section is visible.
• Visibility diff: it’s an estimated traffic drop in a given section.
• Avg pos & avg lost position: it’s an average position and an average position drop.

It’s evident that polki.pl suffered the most severe traffic loss in the /health section, which is the website’s largest section without a doubt. There’re also the /family and /diet-and-fitness sections that also can be classified into Health category.

Subsequently, I analyzed keywords lost by the website.

A few examples, translated into English for convenience of the reader:

• Herpes zoster – from rank 8 to rank 13
• Hashimoto – from rank 4 to rank 13
• Endometriosis – from rank 8 to rank 19

I noticed an extremely important thing: in numerous situations, when the website was on the decrease, the URL was changed as well, for instance:

Herpes zoster:

• Rank 8: polki.pl/zdrowie/choroby,polpasiec-jak-sie-nim-zarazamy-zdjecia-polpasca-powiklania-leczenie,10427904,artykul.html
• Rank 13: polki.pl/zdrowie/choroby,polpasiec-objawy-jak-rozpoznac-polpasiec-i-dlaczego-boli,10428143,artykul.html

Depression:

• Rank 6: polki.pl/zdrowie/psychologia,jak-rozpoznac-depresje-objawy-depresji,10427964,artykul.html
• Rank 40: polki.pl/zdrowie/choroby,sport-dobry-na-depresje-i-pamiec,10381048,artykul.html

Looking at those articles, one may observe that long and really informative articles were replaced by their shorter and worse versions, which naturally contributed to the drop.

There’re as many as 22 thousand similar cases! Therefore, the website fell prey to its own poor architecture of information. That’s the reason why so many great articles lost their visibility just like the ones described below.

There’re articles that lost around 5000 estimated visits.

Conclusions

• The core issue is the architecture of information. The good news is that if one does their homework and does it right, it’s possible to regain lost visibility.
• It wasn’t the website that suffered from Medic #1, so despite a huge drop it’s still more visible when compared to August 1, 2018.

Zdrowie.radiozet.pl

Now a positive example – zdrowie.radiozet.pl; it’s a website which lost its visibility because of Medic #1, but managed to get it back and even increase it by 5%.

Let’s analyze the following sections:

The website gained the most in the /diseases section – it noted a visibility increase of as many as 300 thousand visits.

Below are listed the most frequent common words within key phrases that noted an increase.

Typical diseases, as you can see. For example, there’re 45 key phrases that include the name the name “rwa kulszowa” (sciatica in English) gained by the website, which generated 18 thousand visits overall.

I also checked whether the increases were caused by changes in the URLs (like in the case of polki.pl). There were only few similar situations, so the increases weren’t caused by them.

For good measure, I analyzed if the website gained new keywords or if it was rather the old keywords that went back to their place. I tested a 200-keyword sample and got the following results:

• July 1, 2018 – average position: 11.32
• August 2, 2018 – average position: 16.40
• March 14, 2019 – average position: 5.98

Clearly, the old keywords returned, but got ranked 5 positions higher on the average. The website based its growth on them specifically.

Final conclusions

Whatever name we could give this update, there’s one thing certain – it’s made quite a commotion. One may arrive at a few unambiguous conclusions:

• It was a patch to the update of August 1, 2018 – especially among small and medium websites, those that were growing were the same that had been losing due to the previous update. However, few of them managed to restore their full visibility. This means that Google didn’t draw back the changes, but only made some adjustments.
• Drops across large websites were noted not only in Health category, but they were most drastic there. What’s more, they weren’t websites that profited on Medic #1, but rather those which grew steadily recently.
• Certain websites lost their visibility due to technical errors in the website (like in the case of polki.pl) related primarily to the architecture of information.
• This was one of the few recent updates to promote small and medium websites. It transparently follows from the data that small websites took the place of big ones. The reason might be the fact that medical websites are often run by doctors and undersigned by the author, thus having better E-A-T (Expertise – Authoritativeness – Trustworthiness), whereas medical content at large websites is created by anonymous journalists. I would pay special attention to that when it comes to extensive websites.

As evidenced by the data, the update is still rolling out and further shifts are probably yet to come. We’ll monitor the situation and get back to you in a week or two.

What are your thoughts? If you have something interesting to share about the recent update, leave a comment down below!