Google Updates and SERP Changes - March 2013

[gyppo]

< continued from [webmasterworld.com...]

I'm seeing daily switching between 2-3 datasets across around 600 keywords. Seeing roughly 10-20 keywords swapping in & out of positions.

[edited by: tedster at 2:47 pm (utc) on Mar 1, 2013]

[Saffron]

Mine is Joomla.

[epmaniac]

@Saffron

Is your website new?

[claaarky]

Tend to disagree with TOI regarding the effect of Panda on rankings. In my experience some pages can still rank very well on page 1 even with a very heavy Panda demotion.

It's Panda if it happens on or around a Panda update/refresh date, that's the way to rule that in/out.

[ScubaAddict]

I have to agree with claaarky - I had several (hundreds) of #1 rankings that dropped to 5-10 in the first panda and lost 60-70% of my traffic overnight overall (so about 85% of organic google traffic)- most of the traffic percentage from those top rankings.

A drop from #1 to #6/7 is devastating when it is across the board. In fact, I would bet that a drop from #1 to #5 is much worse (in percentage of traffic lost) then a drop from #11 to page 3 or even further.

On another note - is it possible that traffic losses this past week could be due to storms in the eastern U.S.?

[Wilburforce]

I would bet that a drop from #1 to #6 is much worse (in percentage of traffic lost) then a drop from #11 to page 3 or even further.

I think a lot depends on what is happening to the specific search terms(s) in that sector. If there are still several good relevant results in the top few places then I would agree, but that has not always been the case recently.

[epmaniac]

storm in eastern US will cause less searches, but not decline or vanishing of sites which invented microniches,.. or are you suggesting that it caused some malfunctioning of systems at google, they had to shut few down, so thats why we are seeing this unusual spam sites on top...

[Lorel]

I noticed a considerable drop in traffic during hurricane Sandy when it hit the very heavily populated east coast. I assume it was partly due to power being down for so many plus others watching the news and not surfing. I know I was stuck on the news for awhile and I'm on the west coast.

[netmeg]

I noticed a common element: all the sites that have suffered are wordpress based.

I wouldn't put too much weight on that; probably most new sites that are published are WordPress based. It's free, it's easy, and it caught on like wildfire.

epmaniac - Google doesn't know (or care) that you "invented a microniche" I think focusing on that might distract you from figuring out what is going on.

I'm pretty sure I read somewhere (but unfortunately can't remember where) that Google is okay with ranking scraped content higher than original content if the scraper has more trust and authority, and maybe if it's a brand. Which means, for example, if Amazon suddenly decides to swipe my content, until I issue a DMCA, I'm SOL.

So if someone is scraping your content, you can't wait for Google to sort it out - you need to issue DMCAs everywhere that it's appropriate, and while you're at it, you should study those scrapers sites and see if you can figure out WHY Google thinks they are more authoritative than you are. And then you need to make appropriate changes to your site to boost YOUR authority and trust.

[epmaniac]

i am more trust worthy, more authoritative site.. the scrapper sites are made just by scrapping my titles, they have zero content, i tried to file dmca on their pages... google said they cant take it down as they dont have any content (just my titles)

[randle]

So if someone is scraping your content, you can't wait for Google to sort it out - you need to issue DMCAs everywhere that it's appropriate, and while you're at it, you should study those scrapers sites and see if you can figure out WHY Google thinks they are more authoritative than you are. And then you need to make appropriate changes to your site to boost YOUR authority and trust.

Ya gotta admit its all a little ironic -

Google preaches original, good content is the key to rankings salvation in their search engine, then they rank the scraped version higher because the thieves that stole the content have greater "Trust".

And now the answer for the poor shlep who wrote it in the first place, is to run around filing DMCA's and spending his time trying figure out how to make his site more like the one that stole his content.

It's not Google's fault they have gained the market share they have, its to their credit - but this is the kind of bizzarro world that gets created when one entity pretty much runs the show when it comes to determining which sites get found by the searching public, and which ones don't.

If an entity doesn't come along and apply some sort of competitive pressure on Google, eventually were all going to be living in the search engine Twilight Zone.

[TheOptimizationIdiot]

Tend to disagree with TOI regarding the effect of Panda on rankings. In my experience some pages can still rank very well on page 1 even with a very heavy Panda demotion.

I have to agree with claaarky - I had several (hundreds) of #1 rankings that dropped to 5-10 in the first panda and lost 60-70% of my traffic overnight overall (so about 85% of organic google traffic)- most of the traffic percentage from those top rankings.

Fair enough.

I guess I've been thinking of "getting hit by Panda" as the "negative side of the re-rankings" not some other site (or 4) scoring higher on the positive side of the re-rankings and ranking higher, cause in the first case something is "wrong" on the site causing a demotion but in the second something is "better" on the other sites causing them to be promoted.

Part of the reason I've thought of it that way is it gives me better idea of do I need to "find something to fix" or "find something to improve" and for me they're really two totally different things to look at. Or in other words a 4 or 5 position drop would say to me something needs to be improved (as in adding more of something to what's there) but a "big hit" would say something is wrong with what's there already and it needs to be fixed.

And a 4 or 5 position change in ranking calculation could legitimately happen from any of the 500+ updates a year, so I haven't even bothered thinking "got hit by Panda" in those situations, but think more along the lines of "what's missing" or "needs to be improved" rather than "oh, there was a ranking change it must be Panda what's wrong and needs to be fixed".

There's really two different starting points I get out of the different perspective on what constitutes "getting hit" by Panda since it's a re-ranking piece of the algo not simply a penalty machine and the other algo adjustments change ranking positions too.

I'll try to adjust my thinking though.

[TheOptimizationIdiot]

For "evidence" of the 500+ algo changes a year possibly having a "small position change" in rankings the dates saffron and epmaniac are talking about aren't Panda run dates as far as we know and there are other "major drops in traffic" reported here fairly frequently for non-Panda dates so blaming Panda as the cause of small ranking position changes that may have a large impact on traffic (1 to 7 isn't a huge drop in rankings but certainly could be in traffic) seems to be a bit misleading to me and could possibly send people in the wrong direction when trying to decide what to do the improve back to where they were.

---

Oh, and btw it's a +10% (actually closer to 15% between last night and this am) difference in indexed page count day for me today with 2 new pages finally indexed.

[backdraft7]

everyone is describing the same symptoms, deleted search phrases for their sites across all niches & markets...so my question is: "where did those all key phrases go and why?"

Got to your adsense account and price the phrases you lost. You won't be surprised to see the higher the price, the less your chances of getting the organic listing.

[TheOptimizationIdiot]

Do you think that's because Google does it on purpose or because as the price of ads increase the ROI from hiring a high-level SEO increases or turns positive so there's more "top notch professionals" involved with sites in the high-dollar ad areas than there are in the lower cost areas and their involvement makes it tougher to rank for those terms?

[ScubaAddict]

@TIO -

Typically, the daily Google algo changes effect pages of a website independently. Some can go up and some down on the same day, some changes are small and some are large.

You can differentiate Panda from the "daily grind" by date of the change (was it a day panda occured?), the amount of traffic lost (not necessarily # of ranking positions lost), and that loss is usually a large portion of a sites pages - if not the entire website.

I don't mean to mislead you in any way, in fact I just butted in so you didn't mislead anyone else with:

"You didn't get hit by Panda if you're in the top 10." & "The main point I'm trying to make is you're not seeing a huge drop going from 1 to 10 in any way."

Because that statement is false and can mislead others. I do agree, however, that many people throw around the term Panda as a generalization of any loss to traffic or ranking - which will only mislead themselves.

[turbocharged]

I'm pretty sure I read somewhere (but unfortunately can't remember where) that Google is okay with ranking scraped content higher than original content if the scraper has more trust and authority, and maybe if it's a brand. Which means, for example, if Amazon suddenly decides to swipe my content, until I issue a DMCA, I'm SOL.

This is another disadvantage of being a small business in today's environment. We find ourselves spending precious time submitting DMCA notices instead of focusing on our core responsibilities. Surely domain authority has to count for something, but it should never trump the authority given to the creator of original content.

[TheOptimizationIdiot]

"The main point I'm trying to make is you're not seeing a huge drop going from 1 to 10 in any way."

We'll have to agree to disagree on this one.

That's not a huge drop in ranking position, cause as I pointed out previously, it's a difference in 0.09% of "score" awarded to your page (site) relative to the 1000 result "cap" on the index.

It's not a "huge scoring change" even though it may be a "huge traffic change", so making "huge changes to a site" is likely not the answer since the "scoring difference" is very small.

---

As far as the rest of what I said and why I look at it that way, I think people who read my previous TL;DR post will get a better idea of where I was coming from when I made that one, but like I said, fair enough and I'll try to adjust the way I look at it (for posting here anyway).

---

One thing I think quite a few people have forgotten or missed over time since the introduction of Panda is it's not simply a negative or "penalizing function" of the algo.

It's actually a re-ranking mechanism which means it "awards" positives, negatives and likely in some cases "neutral" or "near neutral" scores so you might "score even" or "close to even" in Panda (no significant positive or negative) and be passed by 3 or 4 sites that were close in the rankings before but "score positive" in Panda, which doesn't say "you got hit" to me, but rather says "you need to improve something" to "score positive" or "increase your Panda positive score" relative to the other sites in the niche.

There's a big difference to me anyway when you remember Panda is not simply a "negative scoring" or "penalty assigning" mechanism but is rather a re-ranking algo that can award positives and negatives.

---

Addition: One of the reasons I don't like referring to "small ranking position changes" as "getting hit" by Panda is "getting hit" has such a "negative scoring" almost "penalization" type connotation it seems to totally remove the possibility of other sites scoring "positive" or "more positive" than the site in question did from the equation and that eliminates half of the Panda scoring process from consideration as to the cause of a drop in rankings or traffic in my opinion.

[ScubaAddict]

TOI -

I guess we will disagree - I have my own results that prove that a 1 to 6+ drop is huge. In addition, many people have done studies and shown that:
position #1 will get 35%-45% of all clicks and
positions 6+ will get less then 10% of all traffic.
.

[edited by: Robert_Charlton at 8:56 am (utc) on Mar 10, 2013]
[edit reason] removed specifics, per Charter [/edit]

[TheOptimizationIdiot]

You're not reading what I'm saying correctly.

A small change in scoring can cause a huge change in traffic.

But a huge change in traffic does not mean you need to make a huge change on your site to get the traffic back.

If you drop from position 1 to position 3 on average the average score awarded to your pages changed 0.03% relative to the other pages you're competing with. (Yes, the traffic will change more than that % in almost every case, but that doesn't mean you need to make huge changes to your site to get the traffic back.)

Increasing your average score by 0.03% relative to other pages you're competing with does not usually require a huge change on your site.

Changing your average score by say 3% or 10% (going from 1 to 30 or 1 to 100) usually requires a larger change.

In one case (0.03% difference) you're really close to ranking #1 and the traffic that comes along with it. In the others (3%, 10%, some other larger % difference relative to the whole of the pages in the index) you're "out of the top spot" (and the traffic that comes with it) by more than "just a little adjustment" in most cases.

---

In other words: Traffic change % is not necessarily equal to scoring change % or necessary "size of changes" to regain rankings.

In some cases large traffic changes (in both directions) can come from very small changes on a site and know whether you should make small or large changes to increase your rankings and therefore traffic is not traffic level dependent, but rather ranking position dependent.

[edited by: TheOptimizationIdiot at 8:08 pm (utc) on Mar 9, 2013]

[HuskyPup]

If you drop from 1 to 3 on average the average score awarded to your pages dropped 0.03%.

The problem I'm having with this type of mathematics is that it is not a true reflection of what is actually happening in many niches. I am not referring to insurance, finance, holidays etc, I am referring to niches whereby there may actually only be 50 - 100 companies competing for specific terms with nothing running into the thousands let alone millions or billions.

There is a huge difference between ranking #1 or #6, in fact it is well-known that it is better to be #10 or #11, first on the second page, when using the standard 10 results layout.

[TheOptimizationIdiot]

So you would recommend making large, drastic changes to a site to increase average rankings from position 6 to position 1?

I think I'm starting to see why it seems so many have difficulty "fine tuning" rankings to get the traffic they want and are used to if the "traffic level change" is used by most to indicate the "level of change necessary" to increase rankings and therefore traffic on a site.

[ScubaAddict]

Okay - I think I now understand what you are saying - I still tend to disagree with your premise.

Speaking in generalities, I feel that if I had a page that ranked on page 5, I could get that to page 2 (say 30 positions) without too much effort. Whereas I feel it would be much more difficult to go from a rank of #4 to #1 (3 positions).

If I understand you correctly, it should be 10 times easier (to go from #4 to #1) based solely on the number of ranking positions increased?

Or I guess to be a bit more simplistic, you are asserting that you need to put forth 2X more effort to go from a ranking of 60 to 50 then you do for a 5 to 1?

[edited by: ScubaAddict at 8:20 pm (utc) on Mar 9, 2013]

[jay5r]

Something definitely happened and I can say with certainty that whatever it was got rolled out between 6 and 10am on the 6th (Eastern time). I have a bunch of sites - mostly Wordpress blogs, but there's also a vBulletin-based forum in the mix.

One of the blogs is written in Spanish (on a .es domain) and it has a long history of being loved and hated by Google. It's up one minute with great traffic and then almost nothing the next. I think some of the trouble Google has with the blog is that many of the inbound links are from English language sites. Anyway, Google traffic to that blog went from about 3200-3500 new visits a day down to about 160-170 new visits a day - so it lost 95% of it's traffic. And when I look hourly on the 6th - it's a straight line down from 6am to 10am.

The other blogs (and the forum) didn't see any noticeable change to their Google organic traffic.

[TheOptimizationIdiot]

If I understand you correctly, it should be 10 times easier (to go from #4 to #1) based solely on the number of ranking positions increased?

Well, not necessarily easier, because it's usually not but "less change" required.

So, using your example and what I've done in the past:

To go from page 5 to page 2 I would make a "larger change" like say rewording a few sentences through out the paragraphs on a page or even the paragraphs on a page to be more inclusive or concise or something along those lines.

To go from position 4 to position 1 I would change one word at a time.

It's not always [almost ever] as easy to find "the word" to change and what to change it to as it is to reword the text to be more concise or inclusive, so I would definitely not say it's "easier" to go from position 4 to position 1, what I would say is it's much more detailed.

[ScubaAddict]

Interesting take.

[HuskyPup]

To go from position 4 to 1 I would change one word at a time.

And herein lies the precise problem, Google has absolutely no idea of what a quality site is.

One can have the most authoratative page with the most amazing images yet Google will rank a page at #1 with one keyword in the titlebar together with a very poor keyword image.

I can be scraped and bombed out of the SERPs totally, nowhere to be found yet other scrapers rank right up behind the #1 scraper.

Because all of this is done behind closed doors with an algorithm absolutely no one at Google knows whether or not they are getting it right.

Maybe their insurance and holiday results "may look right to them" however they haven't a freakin' clue about my industry of what is good or bad, right or wrong, if they did do the SERPs would be in far better shape...like Bing's!

It's getting towards 9 pm...beer time:-)

[Wilburforce]

It's getting towards 9 pm...beer time:-)

Good suggestion. That'll do me!

[totalodds]

I can confirm there was definitely an unannounced algorithm update on March 6th;

I've asked around, tweeted Matt Cutts etc but had no response.

It could be a dry run before Penguin or Panda. Both are overdue.

I'm not sure why Google are being so quiet, when the preach transparency.

[Lenny2]

Ehh i havent seen anything change yet this month... Overdue i say...

[TheOptimizationIdiot]

It's getting towards 9 pm...beer time:-)

Good suggestion. That'll do me!

I'm on my way too, no joke! lol