Google Updates and SERP Changes - June 2011

[dataguy]

< continued from [webmasterworld.com...] >

FWIW, my early change indicators are ringing, starting about 5 am CST this morning, and it seems to be Panda related. Possibly reverse-Panda related.

[edited by: tedster at 3:31 pm (utc) on Jun 1, 2011]

[walkman]

I think it's as simple as them putting a site high up on page 1 to assess it.

Loads of people report seeing "junky" sites on the top, and loads of people are reporting sharp swings in SERPs, were stuff is suddenly bumped right up.

I think this is deliberate - they've been bumped up to see what the user reaction is.

That makes sense, but can Google do that within a day, a lot of data to calculate and move through DC's? And lately I have not gotten enough users for Google to judge the site/pages statistically. From that day to the next I lose only a bunch of pages that maybe got a visitor or two during the supposed testing period. The rest of the pages always seem to get the same referrals.

Ohno, I see a lot of foreign traffic, especially non-english speaking. Looks like the Pandalised sites are having the same experiences. Even if most suck, there has to be some that have improved enough, no?

[MelissaLB]

We've seen the same thing over here: On the 13th we had a good spike in traffic, Google traffic being up about 15-30% more than usual, even shopping results went on the upswing. On the 14th we saw the traffic slow back down to a more normal level but conversions went through the roof. Today we are at a slightly less than normal traffic level with normal conversion rates.

this is all after we had a major rebound in the serps on the 10th. we were hoping the traffic from the 13th to be our 'normal levels' but it appears to not be so.

[chrism]

I'm starting to wonder if Google has really managed to drill down into what specific users are looking for, much more so than before. Those of us seeing less traffic but normal conversions being the obvious example, where google have managed to route a set of non buying traffic elsewhere.

Perhaps the boost before the drought is the acid test to see if a larger than usual chunk of traffic believed to be well qualified actually supports a suspicion of a "low quality site" *for that user intention* before the axe falls. The result of the user behaviour (perhaps all people with G toolbars installed, chrome users, etc?) then dictates the proportion of traffic to change, either as a steep drop in traffic of a specific user intention, or a boost accordingly. It seems to me that a potential boost would be much smaller than the potential drop, as those dropping are likely to have their traffic spread across the remaining results on the SERP.

This is just a gut feeling, (and no doubt Tedster will shortly issue a reminder about editorial notes :) ) however since I first thought along these lines we seem to get more and more reports that fit in.

[outland88]

What is frustrating is I got a head start on Panda (to no avail) and along comes June 14th. I spent months increasing the quality of one site (absolutely not for Google) but to increase its sales. It was great work and the significant increase in revenue showed it. Ironically sites I never touched and with weak pages increased in the rankings on the 14th.

Bottom line I’m getting tired of Google promoting the idea of stickiness when many good sites are meant for commerce not to promote the idea they can write like Ernest Hemingway. I’m simply fed up with Google. The dominant search engine sees itself as an educational tool not as something to keep world economies spinning. Many might argue that the role of a search engine is not that but the last time I checked they weren’t accepting my “good will” or especially my good looks as payment.

[tedster]

seeing less traffic but normal conversions

That is becoming a rather common report. I'm still not sure how Google is achieving it, but this kind of better targeted traffic is becoming a major improvement, IMO. Not so good if your site depends on volume only (CPM ads - or even MFA) but awesome for e-commerce.

I've talked with a number of site owners whose analytics show this pattern, but since sales/conversions are level or even up, they don't care about lower total traffic at all.

I'm hoping to catch the mechanism for this change in the way the SERPs are working.

[AG4Life]

outland88

Bottom line I’m getting tired of Google promoting the idea of stickiness when many good sites are meant for commerce not to promote the idea they can write like Ernest Hemingway.

Although I don't run an eCommerce site, I totally get what you mean. Google is trying to guess intent behind queries, but that assumes people make their intent clear via queries, which just isn't true. Searching for "widget" could mean a myriad of things, and to assume "information" instead of "purchase" is very much jumping the gun, because the searcher could have neither of these intents. The problem is compounded by the fact that big sites and brands are given a leg up, especially for long tails, and it's done in a way that totally ignores user intent and sometimes the query itself.

And even when Google gets the intent correct (ie. "purchase widget"), it's trying to judge eCommerce websites without taking into account price, customer service, customer loyalty and a lot of signals they don't have.

Anyway, don't want to drag this thread in other directions, so back on topic.

14th was a major hit for us too (we're more of an information/news website), but it really started last week (around this time), recovered very well on Sunday/Monday, and now it's bad again. When traffic from Google is up or down suddenly, for our sites, it usually signals some change immediately afterwards. It's as if Google resets the SERPs to something old just before rolling out the new. And just before it started to get bad last week, there was a site wide, massive crawl of our site by Googlebot.

Also, I'm noting a few new title re-writes (even though the re-written page ranks 8th, not something that I would think is important enough for a re-write), and weird things like two entries for my website in the SERPS but not continuous (so I have results in 2nd and 5th place).

And despite the massive crawl, a lot of the pages we've marked for "noindex" are still in the index, even though it's been nearly a month since those changes were made. Maybe when these get processed, it will signal the current round of changes by Google are complete.

[walkman]

@outland88
many are reporting comebacks and drops only to reverse them a day or two later so have some hope.

[tedster]

At times like this, it might be useful to record the Google IP address that is responding to your search query.

[walkman]

At times like this, it might be useful to record the Google IP address that is responding to your search query.

To see if it's international ? Or spot the Dc?

[outland88]

I've talked with a number of site owners whose analytics show this pattern, but since sales/conversions are level or even up, they don't care about lower total traffic at all.

Even though I know what you're talking about it’s not a sustainable flow and it can backfire dramatically. Eventually less traffic equals less revenue. In fact it somewhat contradicts Google telling people to chase what’s hot but they in turn limit to you what they feel is converting for that site. Wouldn’t you still become nervous if they cut your traffic from 2000 hits a day to 200 but you still had the same cash flow? I like the ability to determine my fate and no matter how much sling-shotting is involved I’ll take that traffic. Google is basically determining you’re profits but there’s no guarantee you can increase it if the traffic drops even further.

I'm amazed more people haven't commented on this because it has really become rather sophisticated. That's why I asked where the traffic throttling people were a month ago.

[tedster]

To see if we can query a particular IP address rather than google.com - it may be possible to consistently see a new data-set that way and study other new SERPs there.

[walkman]

Thanks. I will wait to see what Google has in store, maybe they'll remove the scarlet letter from my site.

I agree with Outland88, low traffic should worry people. I was doing very good before Panda even though my traffic went down. Even though Mayday cut my traffic and $$ quite a bit I didn't react too much because I was still doing very good. Just needed 1 more year or less :)

Oh well. Now I'm doing what my competitors are doing and it isn't 'best practices' but they are ranking and it's still a light gray shade.

[outland88]

many are reporting comebacks and drops only to reverse them a day or two later so have some hope.

I was aware of that but thanks.

[ohno]

OK, one of our sites deffo had some kind of recovery from around 3pm yesterday, this continued into the evening. That site is still seeing lots of foreign traffic so today I expect things to be slow. The lack of one box shopping results is worrying-is Google trying to push the shopping tab on the left hand side for certain queries? It certainly looks that way from here, one term that used to be VERY popular(backed up by thousands of hits in GMT) is gone(for us & everyone), yet if I search for an obscure product that never appeared in the one box reults it's now there! Something is deffo being tested.

[danijelzi]

I've reported on June 3rd that about a month ago site:mysite.com started to show random pages of my site on #1. On June 3rd index page got again on #1, and today random pages are again on #1.

[zeus]

hmm can it bee that the rankings has even got worse then before, I dont hope this is the time where some sites get pulled back into rankings.

[foxfox]

Seems there is a great changes start from 15th June. Our sites dropped by 20-30% in search traffic.

[traffik daddy]

I've talked with a number of site owners whose analytics show this pattern, but since sales/conversions are level or even up, they don't care about lower total traffic at all.

Its funny you mention this because I was having this exact conversation with my supplier just last week. I was noticing two different phases of traffic and I was talking to him about conversions and how I didn't understand how one type of search traffic I was receiving converted better than the other.

Keep in mind I deliver only to the UK.

Phase 1 was mainly UK based visitors, long tail searches, so I expected a bigger conversion ratio. Hits are up around 10-15% on phase 2.

Phase 2 was a mix of UK and worldwide traffic, similar long tail searches, about 60% UK, 40% worldwide. Number of overall hits reduced by about 10-15% on phase 1.

Strangely enough, phase 2 converts better than phase 1

[dataguy]

I'm seeing heavy fluctuations, but I wouldn't dare speculate at this point if this will result in something positive or negative for my sites.

What I think is interesting is that I've read all morning about people saying they've had a surge of visitors from California in the past few days. I just logged into Analytics and I see the same thing for my sites using the "Intelligence" report.

Immediately this makes me think that Google has been crawling my sites more in the past few days (and they have) but the Googlebot doesn't normally show up as a 'visit' in Analytics.

I wonder if Google is using human evaluators with California IP's, or if maybe there's a javascript-enabled bot they are now using more than before.

According to Analytics, these California visitors have a near-100% bounce rate. They are viewing one page then leaving. Hmmm.

[Martin Ice Web]

I talked to my supplier, too. And he admitted that most of his webshop-customers with webshops are wipped out. Instead they are seeing are increase of customers who are seeling on amazon.
Nice job google.

[danijelzi]

Google Search today looks either completely broken or like under construction. A query for AMD widgets displays results with "widgets" and "and" (with bolded "widgets" and "and" words in results' descriptions). I'm sure there are a lot of pages on the net about AMD widgets. And these displayed results include both blogspot spam and high quality sites in a mixed order.

Also, for a small widgets query they serve on #4 a big brand's Mini widget model, with Mini and widget synonym bolded in the result description, outranking quality pages about small widgets. I don't want to read about the Mini only, there are hundreds of Mini-like widgets on the net. If I wanted that I would search for the Brand Mini widget.

However, for the same small widgets query, they no longer serve 6 first page results about widgets for small business, but 1. That's an improvement over the same search done a couple of weeks ago.

Furthermore, last night I decided to invest in ordering widgets to have detailed reviews of them on my site, like my competitor, but then I saw that all his top quality reviews are nowhere in the first 5 SERP pages. He's outranked by every type of site, from scrapers, no content sites, other quality sites...

Now, I'm quiting any changes on my pandalized sites. I'll continue to work like pre-panda, I don't care anymore.

[AlyssaS]

Regarding the split testing going on, someone on the WF posted up this hypothetical schedule:

6th June

Group A (site a,b,c) Back to Page 1,2
Group B (site x,y,z) Gone

10th June

Group A (site a,b,c) Gone
Group B (site x,y,z) Back to Page 1,2

14th June

Group A (site a,b,c) Back to Page 1,2
Group B (site x,y,z) Gone

18th June

Group A (site a,b,c) ?
Group B (site x,y,z) ?

Just to update on this - Group B seems to now be in play. For the keywords I watch, the sites that disappeared on the 14th are back today (17th).

I guess by the start of next week they'll have finished testing and we'll know the result...

[elearner]

I just checked my website index in google using "site:" and noticed that home page is not appearing in top 10, usually home page appear at no. 1.

secondly, my indexing is getting lower day by day, even it is giving same results using different parameters.

please share your experience about above points.

[ohno]

elearner, I see none of that I'm afraid. If anything Gbot is on the sites more........which is when I worry!

[elearner]

for point no. 1: I have checked some top websites including few of my competitors and found that majority of them has the same issue as mine

I also checked latest visits from cpanel and google crawler speed seems lower than average.

[ohno]

The only issue I am seeing right now is the lack of Google shopping results box in the SERP's. Images one box is right down the bottom the page now too.

[germanprisoner]

I see huge changes on 12 different portals. Traffic decrease between 20-40% in several non english speaking countries. Not sure if this is the international Panda or if Google is doing some other algorythm change these days? Did you notice anything?

[Shaddows]

Our referrals are all over the place, like nothing I've ever seen before.

Best selling products on a go-slow, stuff we cant get ranked for love nor money are seeing sales.

Our revenue is more-or-less on track, but this is going to get highly disruptive for some people.

[ohno]

Shaddows, interestingly the products I now see in the shopping results box are things that have been rare/never seen! The best sellers are gone from those results but still rank well in the SERPs. Fun times ahead!

[indyank]

I think the sites that see fluctuations are predominantly e-commerce and the current changes might not have have anything to do with Panda.

Those already affected by panda are either seeing normal traffic or a very small improvement in traffic, indicating what is happening right now might not be related to the Panda iterations.