| 4:14 am on Apr 1, 2011 (gmt 0)|
|The name of the product is made up, but G treats it like a spelling error for some other word. When searching 'brand product' you get the 'did you mean' option. I know of more than 20 big media outlets that reported on this new product. |
I reported the other day that I was trying to search for something using the word "plain", and Google was auto-correcting the query (without even giving me an option) and was returning results about "plans". Among these were business plans, and my query didn't even include the word "business". I copied my query over to Bing and the result I wanted was #1.
| 9:16 am on Apr 1, 2011 (gmt 0)|
This is strange. Bing shows the same behaviour as g does
| 12:29 pm on Apr 1, 2011 (gmt 0)|
One change I have noted in one of my niches is the demotion of pages with "too many" popular keywords in the title. In this niche, "fuzzy widgets" and "blue widgets" are popular two-keyword searches (with more than 100k monthly G searches). In olden times it was often desirable to have a page which included both "fuzzy" and "blue" in the title. But now, pages in this niche that have both anywhere in the title are way down the search results for the two-keyword combos (though video titles seem less affected).
| 1:03 pm on Apr 1, 2011 (gmt 0)|
|Here's another fail: |
One of the major manufacturers in my field launched a new product today that generated quite a media frenzy. The name of the product is made up, but G treats it like a spelling error for some other word. When searching 'brand product' you get the 'did you mean' option. I know of more than 20 big media outlets that reported on this new product. Unfortunately, g has only 3 and a half pages of results on it, half of which are not related. All the other reports about this product, all legitimate, are nowhere to be found in the SERPs.
Big fail. I guess their machines are down at the pub, instead of learning...
? If the product has only been launched TODAY, with a brand new word/name, then remember that G has no history of this word in it's database that it can use to decide what it is.
Even if "20 media outlets" reported on the product within a few hours of launch - G still has to a) spider those media sites, b)work out what they are talking about c) then spider the site they are linking to d) then work out the relationship between the media outlet stories and the new site and decide what it is about, classify it and decide how to rank it.
Not easy, as "new made up words" mentioned in the media could mean anything, a new teen slang, new abbreviation, anything. It won't be obvious what this new word is till they get links from related product sites mentioning this new brand, and then they'll be able to work out what it is.
The idea that they could do all this in a couple of hours with just info from "20 media outlets" who have gushed out a press-release...
Don't take this the wrong way, but the reason webmasters are getting into trouble is that they genuinely think G is telepathic and knows what their sites are about without any meaningful signals. It's a BOT that has to work through a sequence of tasks.
If you genuinely want your product with brand new name to be found in G within minutes of launch, you need to have done the online work a whole month or two before, complete with reviews from actual related product sites and so on. A few mentions in media outlets doesn't cut it. And your new site with the new brand name needs to have some decent backlinks too. "Natural" links from scrapers don't count.
| 2:15 pm on Apr 1, 2011 (gmt 0)|
|One of the major manufacturers in my field launched a new product today that generated quite a media frenzy. The name of the product is made up, but G treats it like a spelling error for some other word. When searching 'brand product' you get the 'did you mean' option. |
Welcome to MY world.
|If you genuinely want your product with brand new name to be found in G within minutes of launch, you need to have done the online work a whole month or two before, complete with reviews from actual related product sites and so on. A few mentions in media outlets doesn't cut it. And your new site with the new brand name needs to have some decent backlinks too. "Natural" links from scrapers don't count. |
Also, this is spot on. This one is not the search engine's fail.
| 11:58 pm on Apr 1, 2011 (gmt 0)|
Would like to ask: anyone filing DMCA requests? I am, but am having trouble with the form now! It's not taking a signature and keeps refusing my entry because it complains about the digital signature. Just wondering if anyone is having problems submitting DMCA requests to the search group.
| 12:06 am on Apr 2, 2011 (gmt 0)|
|If the product has only been launched TODAY, with a brand new word/name, then remember that G has no history of this word in it's database that it can use to decide what it is. |
I was thinking the same thing.
It reminds me of when I tried to update the sitelinks in Webmaster Tools. In a couple days all of my sitelinks disappeared. LOL They came back, but it was not what I wanted.
My guess is that it takes a few days (or weeks or longer) for Google to make the corrections. Also, the "did you mean" that Google puts out may still be there - I bet that will depend on how many people query the misspelling and how many query the brand name.
Those are my thoughts on this.
| 7:54 am on Apr 2, 2011 (gmt 0)|
Initially I thought that as soon as G would spider the sites and redo their monthly tallying we'd be back (assuming we fixed the problems, I had way too may very thin tag pages for example). Now I'm leaning towards a fixed time penalty, probably 3 months, for whoever upset Panda. I know my competitors, and I know what SERPS I'm getting so content is not exactly king.
It's very unlikely that no one fixed their content issues post-Panda and while they may keep secrets, in previous updates people have come out screaming "We're back!"
| 12:54 pm on Apr 2, 2011 (gmt 0)|
Anyone seeing a major shake-up in the SERPs today?
Here's what I'm seeing: MFAs seem to have dropped out of the SERPs for product terms.
On informational searches, for one keyword I watch, ehow has been downgraded to #3 from #1.
But on the downside, one of my competitors, "duplicate guy" is back - and he had been gone from the SERPs since Feb 24th (perhaps G has lifted the filter that made him disappear? The duplicate stuff is still on his site).
Anyone seeing similar stuff?
| 2:04 pm on Apr 2, 2011 (gmt 0)|
it usually takes G to learn a new word in my niche, especially when there's such an important new product, about 2-3 hours. It's been 48 hours since and it still hasn't learned it. incredible. major fail!
Here are few websites that reported it immediately: MSN, NY times, USA Today, AOL any many others of the same caliber. they all reported it within the first hour of the launch. If these huge outlets all reported it in 1 hour, imagine the number of small and medium websites that took the story too, not to mention the scrapers.
I sometimes see G get overwhelmed especially when there's so much reporting of a single subject in very little time (1-2 hours).
|And your new site with the new brand name needs to have some decent backlinks too |
this is not about a new site, and certainly not about my site. Just to give you an idea about how major this product is, it's like Microsoft launching a new Windoez version, or Intel a new processor, only that I'm not refering to the IT field in my case.
| 2:32 pm on Apr 2, 2011 (gmt 0)|
|it usually takes G to learn a new word in my niche, especially when there's such an important new product, about 2-3 hours. It's been 48 hours since and it still hasn't learned it. incredible. major fail! |
I think you'll find that the cases where you believe G "learned the new word" in 2-3 hours, actually put the word online a good month before the launch. You might have discovered the launch, and then Googled in two-three hours and found it there, but the webmasters of the site in question would have done the ground work to make that happen weeks before.
Think about the logistics involved from G's point of view. Bot 1 is spidering news sites, and discovers new links to a new product. Bot 1 makes a note of these links for Bot 2, and then continues spidering it's list (if it simply hared off on a tangent, the stuff on the original list wouldn't get crawled). The next day, Bot 2 crawls the new links (it couldn't do it the day before because it had to go through it's other work first). Then the day after that bot 3 classifies the site found based on the data harvested during the crawl. And then another bot probably decides relationships and how to rank it.
It's probably more complicated that that, but you get the picture.
That's why it takes a good 7 to 10 days to see changes in the SERPs after you have made changes in your site. G has to go through these sequences. It certainly doesn't happen in 2-3 hours - not with all the billions of stuff being poured online every day.
G is aware of that, that's why when they are struggling to fill in a search, they put twitter and news sites at the top. It's a lesson they learned from Michael Jackson's death.
You yourself observed that Bing was having the same difficulties as G with this new made-up product word. That suggests to me that the "Big Fail" lies with the webmasters, who simply didn't provide the search engines with the correct information and signals in advance.
You say that the site concerned is established - that may be so. But if the product is new and has a "made-up" name, then they need links to that specific product page from related product sites citing that new product name. Otherwise G won't know what it's about. They are not telepathic, and there is simply not enough computing power in their machine to go through all the sequences that fast. If you start from the launch date, then it will take about a week. That's the nature of the machine. It's unrealistic to think otherwise. If you want the stuff up at the time of launch, you need to help them out and do the ground work.
| 3:53 pm on Apr 2, 2011 (gmt 0)|
AlyssaS, for a good keyword I know, I am actually seeing a page from Digg on page 1 now. It had never been around before.The digg page is a very old one and not a recent submission.
For people who are so obsessed with thin content being as an important factor, I would like to say that I know of so many thin pages on top sites ranking on top for important keywords.It could be from digg or twitter or any other popular site.
There are so many examples where I could see thin pages from lifehacker or engadget moving on to the front page.Many of them are just two or three sentence pages.All they had done is write two lines about it and linked to a lesser known site as the source.
Panda is more about popular sites being pushed to the top, if they have a page for the keyword. G seem to be perfectly fine with thin or almost no content on popular sites.
All they seem to be trying to do is bring the the popular ones to the front, as it would then appear to people that G has unearthed QUALITY. They probably believe that they could beat the notion of spam in google SERPS as these are well known ones and not many would be complaining for thin or no content.
| 4:11 pm on Apr 2, 2011 (gmt 0)|
|Panda is more about popular sites being pushed to the top, if they have a page for the keyword. G seem to be perfectly fine with thin or almost no content on popular sites. |
To clarify, do you have any leaning as to what google might define as "popular"?
Do you mean they get a lot of search traffic?
Or do you think they get a lot of social media approval (e.g., lots of facebook likes)?
Or do they have lots of inbound links?
What would you say that google is defining as poularity?
| 5:44 pm on Apr 2, 2011 (gmt 0)|
They have them all but they weren't on top for those thin pages earlier.It just looks like google has turned on authority more for them while pulling it down for several others.
| 6:05 pm on Apr 2, 2011 (gmt 0)|
Isn't it possible that thin and low-quality pages can rank near the top simply because there isn't anything better?
I've done lots of searches where I couldn't find any high quality information. For many subjects, nobody has ever created a high-quality web resource. So thin and low-quality gets to the top of the rankings by default, because there isn't anything better.
| 6:36 pm on Apr 2, 2011 (gmt 0)|
>>>Isn't it possible that thin and low-quality pages can rank near the top simply because there isn't anything better?
That isn't what's happening. Here's a very specific example I saw.
I wrote a very detailed, long, authoratative article on a very specific subject a few days ago. More than 1000 words. There are 4 or 5 other long, detailed articles on the subject online as well. If I now search for the keyword which that article very specifically pertains to, of the 10 results on the front page 6 of them are basically blank pages with ads on them and a bunch of keyword stuffing to make them appear high in the SERPs.
One of them is someone who has scraped my article.
The other 3 don't relate to the keyword at all, but belong to big brands.
My article and the other 4 which are thick content and actually pertain to the keyword, don't begin to appear until page 2.
On Bing, the same search returns all 5 articles on the first page.
| 6:41 pm on Apr 2, 2011 (gmt 0)|
First of all, I think I have to clarify a few things. This is not about a website, it's a real life, everyday use, product, that I think it doesn't even have its own page yet. So it's not related to any website, or any kind of SEO effort. In fact, the company that makes it (as all the players in this field), goes to extreme lenghts to ensure nobody gets any info on the new product (by this I mean its name) until its launch.
I see this happenig at least once a week in my field. Someone launches smth new. It gets massive, and I mean massive media coverage, and it takes just a few hours for G to learn about the new word.
In this case it failed to work. completly.
I'm sorry i can't be more specific about this.
The thing that got this situation even more interesting is why is Bing having the same problems as G does.
| 3:51 am on Apr 3, 2011 (gmt 0)|
|Isn't it possible that thin and low-quality pages can rank near the top simply because there isn't anything better? |
There are good pages and some of them were found earlier on the front page.But their names wouldn't sound as familiar as Yahoo answers or any other popular site. POPULAR seem to be the key and they have found a method to push them up from third or fourth page to the front.
When people see such names on the front page, they may find it difficult to make a spam claim as these are reputed sites.Not many including me would agree that lifehacker or engadget or gizmodo or any such popular site is a spam.They have some excellent stuff.
But the fact is they also have a large db of thin pages too.It is these one or two liner pages that have moved to the top because of the site's overall reputation.I could correlate this to Matt's usage of "reward" in explaining panda.
You could win in this panda with thin or weak pages, if you could manage to get into the same boat that these sites find a place.But it is not easy.
| 1:17 pm on Apr 3, 2011 (gmt 0)|
same as old pagerank. Sites with lots of it would publish all the crap they wanted and they'd be #1. Looks like Google has given up ranking and it going with the safe option: if it's from Yahoo it must be good so put it #1 and they are enough 'yahoos' to drive everyone else out of business. By the time Google figures it out, it's too late for many small and medium sites.
Associated Content articles on "How to pour milk in a mug" are ranking #1 on some searches. At least Google should shut the #$%*^ up with "improve your content" nonsense.
| 1:50 pm on Apr 3, 2011 (gmt 0)|
Google recently changed all of their DMCA filings to a single page, fill out-the-blanks, and the electronic signature is just your typed name. Only thing I can imagine going wrong is if your signature doesn't match the name of the person filing the complaint.
| 2:56 pm on Apr 3, 2011 (gmt 0)|
Surely everyone realizes that most existing webpages are low quality. Less than 1% are what I would call high quality. Thus 99%+ are low or mediocre quality. This means that for millions of search terms there simply isn't any high quality information available on the web. That's the basic reason why you see so many low quality pages at the top of the SERPs.
The web in its present form is only about 20 years old. There are still big gaps where nobody has created any high quality information. It needs more time to develop.
| 2:34 pm on Apr 4, 2011 (gmt 0)|
|aristotle wrote: |
Less than 1% are what I would call high quality. Thus 99%+ are low or mediocre quality. This means that for millions of search terms there simply isn't any high quality information available on the web. That's the basic reason why you see so many low quality pages at the top of the SERPs.
I don't disagree, but I trust that those people complaining about low quality results are doing searches for which they *know* high quality content exists.
If 99% of the potential results for your search are crap, that 1% that isn't crap should be on top if everything is working properly.
| 5:13 pm on Apr 4, 2011 (gmt 0)|
Except for the big volume queries, I don't remember ever seeing ten solid results for most searches. That is the ideal, certainly, assuming that ten good results exist. The reality is that machine intelligence is not yet that good, and there's a lot of dreck out there muddying the water.
A lot of my searches are either for pages that I know exist or technical questions. Both of those query types had been getting mighty rough in recent months, and both of them seem to have improved for me in March.
Finding pages that I remember from months ago is still a bit challenged because the "freshness" dial seems to be cranked up too high for my taste, but it has at least become possible again. For a few months it often felt hopeless.
| 1:35 am on Apr 5, 2011 (gmt 0)|
Wow, the press is starting again with Google. They wanted eHow to go down, Google gave them a raise. [freakonomics.com...]
| 1:51 am on Apr 5, 2011 (gmt 0)|
Good to know at least someone is paying attention. The press has been very... silent on the whole Panda thing except from that first flurry of articles the first week telling Google how great it was.
| 11:29 am on Apr 6, 2011 (gmt 0)|
Google algorithm updates, always make some changes but gradually it remove bugs from it.
| 4:26 pm on Apr 7, 2011 (gmt 0)|
We've got a new thread where two members are reporting a big rankings shuffle yesterday (April 6) [webmasterworld.com]. I wanted to note those reports in this thread, too, in case it becomes important for future reference.
Has anyone else noticed a significant change on April 6? I can't say I see anything dramatic in the areas I watch.
| 4:53 pm on Apr 7, 2011 (gmt 0)|
Yesterday I received 11% increase in visits and the number of keywords that receive traffic more than doubled compared to the previous wednesday. The improvements in traffic mainly came to my blog and not deeper more informative content.
| 5:14 pm on Apr 8, 2011 (gmt 0)|
< moved from another locations >
I'm experiencing big traffic drop today (April 8) across the world. Is anybody experiencing the same ?
I don't know if this has something to do with Panda or not. If I'm right it didn't roll out yet over the world (at least according to my stats)
[edited by: tedster at 5:52 pm (utc) on Apr 8, 2011]
| This 146 message thread spans 5 pages: 146 (  2 3 4 5 ) > > |