2: Google MAYDAY Update - SERP Changes May 2010

[cangoou]

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

Puh, this is really strange: I still see 2 resultsets - and after 7 days I can say that SERPs #1 are shown from right at 0am to about 5am local time and SERPs #2 the rest of the day. I thought this was some kind of dancing around again but it seems to be quite stable now for 1 week.

Funny thing is: SERPs #1 are better then before, SERPs #2 are worse (in rankings of main keywords).

Google is no longer following my robots.txt file. It has indexed hundreds of pages that I have disallowed.

Same here, got a complete set of new pages with robots=noindex in the index.

Generally it's best to make any changes when SERPs are stable, never within an update due to panic!

In general I would agree, but I haven't seen something like "stable SERPs" since January... This makes it quite hard not to panic ;-)

[edited by: tedster at 7:43 pm (utc) on May 13, 2010]

[freejung]

Given MC said "no caffeine yet"

When did he say that?

Edit: just saw your edit, Shaddows. That's what I thought -- so we really don't know where the progress of Caffeine is, but I suspect it's a lot farther along than when he said that.

[anand84]

Interesting arguments Freejung. Really intelligent thoughts. Anyway, I just hope that alone would not affect websites to a great deal since we all rank on page 1-2-3 for a variety of keywords.

[MadeWillis]

What can we do to bring back the traffic to our sites? Dealer sites are primarily full of car listings so unique content on each page is a bit hard.

What did you do to get the good rankings in the first place?


How many people seeing these drops have bought links in the past? Just out of curiosity.

[freejung]

I just hope that alone would not affect websites to a great deal since we all rank on page 1-2-3 for a variety of keywords.

Yeah, we do, but I bet a lot of us here are well-established players in our niches. We benefit from stagnation of results, since we're already well-positioned. This sort of change would hurt us to the benefit of newer, less well-established players.

Having said that, I haven't lost any traffic (yet, knock on wood). But my niche is still pretty stangnant, and tends to be more so than others. I am seeing some fluctuation in ranking for particular keywords, but so far overall traffic is pretty steady.

Edit: and thanks for the compliment! I'm just throwing out ideas here, making an intuitive jump based on scant evidence. I think it's at least worth investigating. Maybe we should declare a temporary moratorium on the no-keywords rule so we can find out more? Or maybe discuss this elsewhere?

[edited by: freejung at 5:34 pm (utc) on May 19, 2010]

[ohno]

Guys that video linked to is the same one i linked to-it is NOT current-look at the comments dates, nearly a month old. MC said Europe was next for a test data centre, me thinks that happened last week! In any case we are deffo seeing niche products(ones that only we sell & make) selling again. Generic products that we used to rank well for are not selling, from what i can see the common products now have lots of new/spammy sites on the front page.

[walkman]

One more thing: I looked at about 600 goog referrals, I only have two that have identical words in the referral string (they are sorted /threaded by that) so Google is certainly jumping. I find this very odd.

Edit: Google would love to announce that caff is live, if it was. A 6 months delay doesn't look good for them

[dusky]

From the video interview [live.webpronews.com ] which is few weeks old, nonetheless it's the last what we heard from MC.

Q to MC: Caffeine.....We're still sitting on the same Data centre?

A from MC: That's right......have to be cautious.....when it's a very large rewrite of your Indexing Infrastructures.....nothing new to report...

As long as there is no new blog from MC, I take it, there is nothing new to report as he said and the wholesale re-indexing and rewriting is still going on.

We can all agree now what I always been saying seems to be happening, re-spidering and re-indexing from scratch which explains the rewrite, when that's is being done they somehow have to leave ranking and trustrank to the end when all data is gathered about sites, some will improve and some will deteriorate no doubt, BUT now many large sites are stripped from a lot of their trustrank while being re-indexed and re-spidered from scratch.

He may have also meant very large rewrite of your Index AND Infrastructures the answer was too quick, but I think was the above!

[rowtc2]

Rewrite is different from major update index.

Is possible to rewrite live their database? Could be a chaos in results.Or maybe to rewrite on offline datacenter until job is complete.

If ther is just an accident, i don't think drop just 15% of traffic.It looks like an algo update.

Maybe 950 or 350 penalty is slightly modified and caught some sites?

If someone could syntetize common simptoms of affected sites in a list,will be a good point.

[rocco]

It is still going on... I have rising stars, but the sinking dogs are more interesting.

One of my sites is loosing every day a bit of google traffic and also loosing sites in the index with the "site:" command. The site was very heavy spidered for the last 5 weeks and now is in usual spider-mode which is 5 times less googlebot hits. Still, the site is getting massive traffic, so there is no "penalty" of a kind I know of.

[whitenight]

FULL STOP!


Because it pains me to see people so utterly lost in the woods... a brief re-appearance (with full sarcasm in effect.)

it's the last what we heard from MC.

Good lord. Why is ANYONE even listening to him?!

He also SAID LAST JULY (and again in Dec) that Caff had been tested for several months BEFORE that and was ready to roll.

When exactly am I or anyone else to supposed to take MC at his word?!

Here's a hint...never.

very large rewrite of your Index AND Infrastructures

Oh, so NOW that MC has "officially announced" that it's BOTH an infrastructure change AND an algo update everyone suddenly believes?!

Goodness, no one here has been beating this point across everyone's head for over 9 months.. <rollseyes>

-----------------

Now again for "advice" (feel free to ignore, instead of arguing)

Caffiene is LIVE and LAUNCHED. Has been for over a month now.

It is buggy?
Yes.
The ALGO is ALWAYS buggy. Use it to one's advantage or don't.

When can one expect for it to be "fixed" to one's liking?
Not anytime soon.
For kicks and giggles, let's say 6 months minimum

And MOST IMPORTANTLY, what Gorg finds 'buggy' and what everyone here THINKS should be "fixed"
are two completely different expectations

I would HIGHLY SUGGEST that everyone affected by Caff (yes, again, it's live if not LIKED by all) figure what has changed in the
NEW ALGO CHANGE
cause there's a high probability that it will remain that way for a LONG TIME, if not, until the next "update."


For those desperately seeking reassurance from Gorg that this will all be "fixed" in short time...

Ask yourself HOW LONG has it taken Gorg to roll out Caff (even if you assume that what everyone is seeing NOW is NOT Caff)...
We're going on 11 months now and you expect them to fix the bugs in less time than that?!

Good luck with that logic.

Get to work people!

_{CainIV, remember my prediction about the after Caff results? :)}

[pontifex]

Get to work people!

well, never stopped here... BUT:

1st - i am seeing some recovery since last monday and a formerly 20% drop reduced itself to now 14% drop in traffic in a 2 day average compared to the same time prior to MayDay!

2nd - MC is not an evil smoke bomb, he is just not as informed as he thinks he is - and I think this is totally normal in a company with 5k+ employees. Last week I talked to the CEO of a very large German company in a SEO session and he was totally surprised by all the stuff the IT team did about 4 months ago (despite the fact that he ordered it about 6 months ago). So, you can listen to MC, but I suggest you take everything as rough guesses :->

3rd: the site: query is counting now upwards for me, starting a few minutes ago!

P!

[dusky]

whitenight, It was not just about what MC had to say, it was the proof in the pudding / eating, we are seeing and experiencing what I said which is exactly in line with what he's saying. Caffeine IMO is not rolled out, experimental on some DCs yes, but not rolled out IMO, the day it will roll out, well see 100s of posts here alone in one day, bookmark my words.

The official roll out will hit the Internet like never before, it will be like a tsunami, Florida will feel and sound like breaking wind in the bath tub in comparison. The proof of that pudding is the waves we are seeing now before the big one, and any DCs that are serving caffeine data are experimental. Some of us will sky-rocket, some stay the same and some will plunge to oblivion. One thing about searching G* is it's heading towards AI and user behavior based results (not webmaster behavior / SEO) as many of us are thinking!

IMO, We have two things happening, one is the on-going experiments from various caffeine DCs, the other is a total recall of the entire Internet links to all web pages and reassessing them from scratch, hence a gradual Algo update, yes Algo update and I strongly believe it is the biggest Algo update ever seen once it's done.

We can speculate, suggest and give our POVs, however, each one of us is basing that on observation, research, trial and error etc.

Good lord. Why is ANYONE even listening to him
Well, that's because he represents G* to webmasters all over the world at least, and believe it or not, behind the cool dude, laid back guy he wishes to portray himself to be, there lays a corporate PR savvy professional, for every giggle of laughter he deliberately pushes out of his mouth, there lays behind it a calculated, number crunched like answer and he does it well!

[backdraft7]

Dusky, you're right on the money. We will soon bow to Google for the right, excuse me, PRIVILEGE to receive top listings. I think MC may be a Google manufactured PR android.

[walkman]

I agree with pontifex, backdraft and dusky,
data is being regenerated and while my total is down, I am getting many great sporadic money term referrals so that data is somewhere.

[internetheaven]

I think MC may be a Google manufactured PR android.

I would say more like a "shiny object". PR is all about misdirection. MC's blogs are all about misinformation.

whiteknight's post was usually long but unusually lacking in information. Normally it's 90% rant, 10% great observation. I think that's a sign of the times. In fact, despite having loads of work and an SEO direction(s) to go for myself, I don't think I can offer any advice on the current Google state either.

Whilst many are crying the END OF SEO AS WE KNOW IT - I think it is more like the end of SEO tips and discussions as we know it. These days, there is no one-size-fits-all SEO pattern to follow. With more than 100 sites to manage I'm finding that I'm having to do different things to boost rankings for each depending on industry, site size, target audience, product buzz, monetizing method etc. etc.

Normally I'm quite free with advice, but these days I find myself refusing to even comment on title tag structure with other webmasters.

[Reno]

Some of us will sky-rocket, some stay the same and some will plunge to oblivion.

If you are correct about this Dusky, then here's my question:

Is there anything that you see happening now that gives us any indication as to where we'll fall vis-a-vis your quote above? In other words, if we've taken a hit on this end, that does portend disaster ahead? Or, is the "final" rollout going to turn the world upside down, and nothing that is happening now should be taken as a clue one way or the other?

Just wonderin'.....

...................

[backdraft7]

It sounds to me like the roll out is more like a tidal wave...some are recovering but I am currently in the middle of the quagmire (giggity) and whatever traffic I'm getting is either bots, un-targeted or misdirected to my site. That's the only explanation I can muster for the sudden drop in sales conversions.

[freejung]

END OF SEO AS WE KNOW IT

People always say that. And here we still are...

Even if dusky is right about AI, all it means is that SEO becomes slightly more about the ability to manipulate user behavior (usability/UI/marketing) and slightly less about manipulating search algos. That's a shift in priorities, but not a total paradigm shift IMO.

[aristotle]

IMO, We have two things happening, one is the on-going experiments from various caffeine DCs, the other is a total recall of the entire Internet links to all web pages and reassessing them from scratch, hence a gradual Algo update, yes Algo update and I strongly believe it is the biggest Algo update ever seen once it's done.

I don't understand this idea that Google is reassessing everything from scratch. In the past they have always revised their database gradually bit by bit as new data is collected by crawling. Why would they just suddenly throw everything out the window and start over? I don't understand it.

[walkman]

Aristotle,
maybe they have many more links now that they have the caff infrastructure.

[backdraft7]

This grand experiment, or rollout, whatever you prefer, is surely costing many business owners their immediate livelihoods. I for one am not sure if this trend of zero sales will go on for another day, week, month or year. I've been a good boy and have always adhered to the GWMGL's, I spend a good $500/mo on Adwords, I'm a BBB OLR member and my site is monitored by Alertsite for my customer's safety. I'd hate for all the work of the past 10 years to go down the drain now. Where's my Kleenex? - AHA! a sale just came in, SERIOUSLY! The curse may be broken! lol

[aristotle]

maybe they have many more links now that they have the caff infrastructure.

If you mean that Caffeine is enabling Google to greatly expand the size of the undelying database, then I agree that's a good possibility. I also agree that this could cause changes in the SERPs as new data is incorporated. What I don't understand is this idea that they are reassessing everything.

[BillyS]

I agree with pontifex, backdraft,dusky, and walkman because this makes a lot of sense to me. If Google had to reindex the entire web, and we know that ranking is an iterative process, then we're going to have to wait things out while the whole shake up settles down.

If we're looking at a buggy roll out, then my experience tells me it will take about 3 months or so to correct all of the significant bugs and 90% of all bugs in less than 6 months. This could happen faster, Google has a lot of money as stake and a lot of experienced coders.

[backdraft7]

Slow roll out has a pleasing after effect for Google because as your organic results are being shuffled about, and traffic lost, you will need to increase your Adsense budget and Google's bottom line. I should buy a few more shares of GOOG.

[dickbaker]

I've never even suspected that Google would ruin their search engine just to boost Adwords revenues. A poor search engine doesn't get visitors.

I've been looking at sites coming and going, and trying to figure out why they're where they are. A site that should be on page 90 makes it to page 1 for a few hours or a day, and then disappears. A site that's been on page 1 disappears completely, comes back to page 1, then a few days later disappears completely, not even moving to page 2. Meanwhile there's MFA and all sorts of other sites of questionable quality getting very good rankings, and many of them are stubbornly sticking in the top results.

If there's a pattern here, I sure don't see it. Whitenight, I'm trying to do what you've been doing, but I can't imagine how you could see a pattern to this.

[dusky]

Here is what I think they are doing, a common man's analysis.

Infrastructure switch:

Water Reserve A (old index) : 10,000 Litres capacity, cloudy, murky and saturated water

The Lagoon (the web), 10 trillion Litres of water, also inhabited by (amongst others) busy beavers, aquatic preys and hunters...

Water Reserve B (new Infrastructure): 9 trillion Litres capacity

G* wants to switch to using Water Reserve B to serve us what we need, Water Reserve A is too small and is no longer good enough for its business model as a water supplier.

G* is not throwing its old Water Reserve A along with what's in it, it is slowly pouring it in Water Reserve B Using a water filter, filtering out what it decided should be left out and only served in a case of drought, all the while, Reserve B is also being filled through a different new high-tech water pipe which includes its own filter AND we are mostly still being served from Reserve A, little by little that water is getting scarcer and scarcer. The good clear water has already been poured in Reserve B (new IS) that's why we either can't get it (find it in search results) in Reserve A or get it but at the bottom of Reserve B (still being filled and water ways still being channeled).

New Algo:
Those busy beavers (SEOs, Webmasters and site builders...), Spammers (hunters) and the fish and preys (the browsing population) in that Lagoon are now seeing the Lagoon being emptied slowly and that water is going in Reserve B. All three sections of those three concerned parties are now monitored separately. Beavers can get in Reserve B where underwater aquatic CCTV cameras are there watching, persistent spammers are left out in the fringes of the 10% full Lagoon (once all Reserve B is filled), preys and fish are monitored in Reserve B for their consumption and fed whatever G* things each and everyone of them is likely to want, hence behavior based service and AI comes into play.

To sum it all, G* is moving the entire old Internet index from one place to another and adding more than 10000% to it, living out what's dangerous and monitoring what's being moved, when all is moved, it will be monitored even stricter.

Conclusion:
Good, unique and fresh content that appeal to to the many will prevail over the good, unique and fresh content appealing to the few. Whether in a market niche or a large segment, all the same. SEO will be about understanding the market than about understanding how websites are preferred / ranked, though that was heavilly linked before, less now (when done), it's purely about looking at eye balls, search orientation, needs and wants, differentiation, market positioning....the classic textbook marketing strategies which we all somehow neglected online and looked for meta marketing strategies instead.

It's wilde out there, not just a wild life but wild wild web!

[supercyberbob]

Hi, I'm new to the forum. This is one crazy thread.

freejung's theory on what I would call dynamic results or rotating serps is very very interesting.

If it isn't already in the works, and freejung doesn't work at google, you just gave them some very good ideas.

I'm sure we've all seen the same URL stuck at #1 for many keywords. It would make a lot of sense to rotate even #1 and #2, and them collect (additional) data (CTR etc etc) and do some algo magic with that.

And yeah, throw everything you think you know about SEO, in terms of conventions. 10 results on a page could be 20 tomorrow.

Real interesting changes going on here.

Great forum btw.

[ohno]

@ whitenight....

"For those desperately seeking reassurance from Gorg that this will all be "fixed" in short time...

Ask yourself HOW LONG has it taken Gorg to roll out Caff (even if you assume that what everyone is seeing NOW is NOT Caff)...
We're going on 11 months now and you expect them to fix the bugs in less time than that?!

Good luck with that logic.

Get to work people! "

I'm afraid your logic does not stack up with what we are seeing, the graveyard week was LAST week for us(others are now reporting the same for them THIS WEEK), this week has been MUCH better and we have done NOTHING! Things seem to be changing by the day.

[wizardmansa]

Wow, that is the most reading I've done about an SEO topic in my entire existance.

Is anyone noticing a BIG drop in META relevance.
We based quite a large portion of our SEO on (well structured) Meta TITLE tags & Meta DESCRIPTION tags as this would normally give us the edge over competitors.

NOW, this seems to have dropped to almost NO relevenace. Where searching for description tags in Google doesn't even bring my site in the results. I've tested this on over 10 sites as well?

[dusky]

So, wizardmansa, you are following me on WebmasterWorld then, as I don't believe in quitter :)
Yes, soon that old faithful of meta descriptions and title tags being the centre of our world, will sadly be laid to rest, for G* anyway!

They will still I think be useful to G* as a starting point of reference about a site, but its decision on rank will lay on user voting if that description merits what it says, and in any case as I see it G* is taking snippets from the content and ignoring that description all together for many, many sites. However, the conventional SEO should still go on for Teoma / Ask, Y!, B*ng and other old and emerging new SEs, I won't abandon using meta SEO because G* is not relying on them anymore as a metric.

I have seen the importance of the meta description entirely disappear for the majority of the sites I manage and dozens of sites I watch, you're not the only one, and this is a small window to their new Algo.

There are of course newer ways to get eye balls and bookmarks organically, we just have to adapt a little.