Hoping to pick the brains of some old pros!

should i change my user name to THE TEACHER :laugh: Excellent Rod i personally think your great !!

ok the joke may be on me but who cares. ( am i the most stupid person in this thread ? )

so a few questions, not just for rod for anyone

ancillary links

when i redone my site to 1.5 i switched off all side boxes / cross sell etc, i do have featured on front page. so for me i get no ancillary links. if my site is found on google it should be a direct link to the product or category as this is the only place the words will be found,

so are you better with loads of ancillary links and hopefully the proper link will be the first one they come to but at least they have found your site or stick with the no cross sell etc and direct hits to the product / category or is this a personal preference

content

ok big theme you read is content is king, ill be 100% here and say most of my product descriptions are from supplier feeds and you can upload 100s of products very easy with easy populate, so my content is not king its just the same as loads of other sites. once uploaded i dont change product description so page wont change at all

at the minute im adding a manufacturer description, so on product info page you will have the product description then a bit about the manufacturer. hoping that this means there is slightly more content on my page than some of the others sites, but again when google sees this new page it will then never change again so will this extra content make any difference

next along the content theme still, if i use cross sell and side boxes etc every time google would visit the page the content would change, so i know you get the ancillary links but does google see all this extra changing content as a good thing as your site is constantly changing

bn

RodG:

Well, it certainly has solved the mystery for me.

I think the jury has not come in quite yet. it seems that google has already dismissed the url due to duplicate content. Theoretically, the URI could eventually also be dismissed (due to duplicate content) and the site will disappear from WD searches on google. If that happens, then no new/useful info comes from noticing the anomaly (might just be index catching up to cache again).

But if Google persists in showing that home page, it seems that the only logical conclusion is that the keywords appear in uri pointing to the domain. Right?

Holy snap! That would be interesting indeed!

Wow. That presents a number of interesting possibilities, stuff I've never considered for SEO.

Too early to know if google will persist on displaying that page in the results. Might just be another non-starter.

Still, I'd be interested in Niccol's take on that. I sure I'm missing something, I'm stuck in a box on that one. What could cause what is happening, other than keywords loaded into a phantom URI?

niccol:

I am not sure how I can demonstate without saying what I think is a good approach. In my opinion:

I would set up two sub-domains

Easy done, but only 2? Is that a big enough sample base to be able to draw any valid conclusions? Wouldn't that be like trying to test the 4/8 cylinder car speeds with only one car of each type?

niccol:

I choose to name them as similarly as I can
I would password lock them so they can't be indexed
I would install zen on both sub-domains simultaneously ( yes I know that is impossible)

Are you serious? In order to get a valid result we need to do the impossible?

Besides, if we could do the impossible, would any of the test results be valid? After all, didn't you recently state that the age of the domains are a factor that we need to consider. Doesn't this mean we'll need to repeat all the tests with domains of different ages before we can make any valid conclusions?

niccol:

at this point I have pretty nearly identical installs which logically should perform identically in google.
Then I would change one single thing on one of the installs. So, if you are interested in 'keywords' in the URL just change that. Nothing else.

I'm confused. It wasn't very long ago that feznizzle serendipiously found himself in a position where he had a single store where the product content variable could have been eliminated, leaving the 'perfect' test scenario where **everything would have been 100% identical in every possible way, except the format of the URL. **
I was lambasted that this test wouldn't prove anything because of all the other variables. I fail to see the difference between your proposal and mine. Exactly what is it I'm missing here?

niccol:

I would not publish where I am doing this publicly because people clicking around can effect the results.

This wouldn't have been an issue with the feznizzle single site test, because any clicks would affect the site as a whole, with no bearing on the product/URL's.

niccol:

Sorry, by the way, I clicked around your sites without thinking about it!

No big deal, it wouldn't affect any of my tests. Also, out of curiosity, was it also you that initiated the 'SEO check' on our site from chme.com? If it wasn't, you may be interested to know we only scored 87% (they recommend 90% or better as being good result). All in all not a bad score considering we do nothing in regards to SEO... its all pure native Zencart. It would have done even better if I gave a little more consideration to the text formating of the Wuldbitzer product.

niccol:

Then remove the password lock and let the games begin.
This way I have isolated the single change that I have made. Logically if there is a difference in performance then it must be due to that single change I have made.

But wait up, this doesn't take all of the variables into account. What about the speed of the server? The speed of the network? What if google crawls one of the sites at a busy time of the day and the other during the quiet period? Surely these factors have to be accounted for in some way or another, otherwise the results will be invalid.

niccol:

Of course, some idjit like me will come along and say that just doing it on a single pair of sites then the results are dubious.

<grin> I said that before reading this. :)

niccol:

But can you really be bothered to set up a hundred pairs of subdomains?

We can't rely on any results unless we do this (or, we could make a reasonable assumption that the domain names really aren't that important after all, and simply recruit more volunteer sites, which is what I chose to do).
Even though it is reasonable to assume that the results from more sites with different domain names is going to tell us far more than any results from just two domains, this is/was apparenlty another flaw in my testing/logic.
What does that tell us about your logic?

niccol:

I might also say that any result you get in the first period is going to be noise rather than signal so that you need to wait a bit longer than you have.

1. I really have no idea what you mean by this.
2. Since 'time' is another of the variables that we need to consider, then in order for the result to have any meaning, don't we need to monitor and record the results over a period of time? You seem to have missed this in your proposed tests. On the other hand, my 'flawed' tests had this well in hand (look back at the speadsheet I'd created very early when I was 1st devising the tests).

niccol:

That is how you might, if you chose to, set up a slightly more rigorous 'test'. I genuinely do not mean this negatively. It is a positive suggestion about how you can proceed and improve your results.

I appreciate your efforts, but seriously, it has just as many 'flaws' as my tests. It doesn't take all the variables into consideration (by a long shot), not to mention that in your own words the test scenario is impossible to set up.

niccol:

Having said that I kind of appreciate the pissing competition approach because it does make people think about stuff and does raise questions that may not have been considered before. I agree that it can be kind of fun and can highlight interesting areas.

The bottom line though is that we both have flawed test methods and flawed logic, and the concensus is that no matter what we do, or try to do, any results or observations we make are totally and completly worthless because there are far to many variables that weve not (or are unable) to take into consideration.

As of this day I now know that I can't find out whether 4 cylinder cars are faster than 8 cylinder cars or not by racing them, unless I am sure none of them are being driven by a professional driver, oh, and of course, they all have to be the same colour.

This was a hard lesson for me to learn, and I have you to thank for teaching it to me.

niccol:

Listen, Rod, I genuinely apologise if at any point I have over-stepped the bounds of politeness.

Listen Nick, you should know by now that I don't give a flying duck about politeness. It serves no purpose in regards to solving technical puzzles, and this topic has been,and always will be a technical issue.

niccol:

Which I know I have. I am going to quietly leave this thread now.

You choose to leave now? I think you should clear up some of the mess and misinformation that you've been sprouting before walking out on us.

niccol:

I really wish you the best of luck with it and hope you get useful and interesting information.

But Nick.... You've convinced me, and no doubt many others that all of our testing is pointless and will never prove to be useful because we aren't taking all the variables into account.

niccol:

PS. I think that if we go back and look at the first post of this thread we are all going to have to admit that we are all very much 'off topic' :-)

On the other hand, if you look at the topic of this thread, it is very much on topic. It just isn't the direction any of us were heading.... and why? Because you and you mate DivaVocals decided you had nothing better to do than to try prove that arrogant ########## RodG wrong in any and every way possible, with no regard to any of the other participants in this forum.

You have constantly and repeatedly twisted almost every word I've written to make it appear I've stated things and made claims I've never made. Any useful information that could have been gleaned from the tests have been lost in the noise. In short, you have succesfully destroyed anything good to come from this thread, which is now a lost cause.

Well done. I hope you are really proud of yourself. You too Diva!

I'm out of here. As far as I'm concerned this thread is now closed.

Rod

bn17311:

( am i the most stupid person in this thread ? )

I don't think you are even a contender. I'd say pretty much everyone BUT you has been accused of that!

so a few questions, not just for rod for anyone...

so my content is not king its just the same as loads of other sites.

I have recently begun doing that. I usually preface by saying something like "Mfr Description:". There are tons of competitors in our business who use mfr descriptions. Come to think of it, wouldn't the mfr description be the best one?

once uploaded i dont change product description so page wont change at all

This is where rotating sideboxes, cross-sell plugin and my idea of making 50 text banners in the footer to rotate through the site come in to play. Other than that, it isn't practical to update each product daily...weekly...or even monthly, for us.

at the minute im adding a manufacturer description, so on product info page you will have the product description then a bit about the manufacturer. hoping that this means there is slightly more content on my page than some of the others sites, but again when google sees this new page it will then never change again so will this extra content make any difference

This is where "gibberish" can come in to play, if google really is okay with that. :wink:

feznizzle:

So if you set those two clouding issues aside, you are still doing quite well.

Yes. But if "pretty urls" were supposed to give an advantage perhaps they could have overcome those differences?

I wonder if you have altered anything also?

Yes, prommart.com's template just the other day, as well as fashion-mart.biz. Think template changes will affect google? If so, that could partially have been my downfall last year...constantly changeing templates in the quest for the right look.

But eventually I will begin to try things to move up in the listings. Perhaps see what works (cross-sell plugin? rotating text banners?) and what doesn't.

With regard to gibberish, I think if you use systemically (as SPH has considered, I think jokingly?), then you will have a problem.

No, I was not joking.

RodG:

I'm out of here. As far as I'm concerned this thread is now closed.

No! We're just getting started with the possibilities.

RodG:

Because you and you mate DivaVocals decided you had nothing better to do than to try prove that arrogant ########## RodG wrong in any and every way possible, with no regard to any of the other participants in this forum.

Rod

i have dealt with both these people on this forum and both are helpfull and give time to people like me who have no programming knowledge but want to run our own websites, i personally think you ruined this thread / test yourself

so here goes rod

so if we put all these cars everyone keeps talking about out for the drivers to pick, 10 drivers are professionals, 10 are just inexperienced drivers like me

so on google i am getting about 18 pages of links so 180 cars all different specs but only 2 colours, grey or red, different manufacturers different models different engine sizes, all variables

so the drivers are told to go out and select the cars they would like

now we are only interested in the first ten past the finish line (page 1 google),

so say 8 out of the top ten cars were red, it would be hard not to notice that red was a really popular colour, i would say with this information even batman and robin
wouldnt argue with this

excellent that cleared that up, simple no arguments Red is the Most Popular Colour

but some people arnt convinced and really want to know all the variables for their brains to accept this information as true ( they both know who they are )

what they would like to know is everything, loads of variables in this, say one of the variables was that all the 2 litre sports models were all red then we could say the colour red had bugger all to do with it and every body was picking red because of the engine size so they could win the race

so you can see where these people are coming from rod they want all the info not just some of the info, thats the way their brain works

they also dont want people like me who have no knowledge reading something on here which is not guaranteed to be true / helpful to their site

people gave their opinion, you didnt like it but we could have continued, at one stage i think you mentioned they sabotaged the test, HOW ?

did they list products to mess with the test ? i dont think so,

i didnt read most of this because every time i clicked on here you seemed to be ranting at someone,

im here to extend my little knowledge of zencart/php/mysql and i enjoy it, you made this thread unenjoyable

my biggest gripe is you secretly changing things on your site and not informing the minions who joined in the test, that really didnt show much respect for us believers

please note some names have been changed to protect the innocent :smile:

i dont want to rub anyone up the wrong way because no doubt ill need you before you need me but Rod you ruined this no one else

bn

Feznizzle:

I notice that your keyword url prod (prom mart) is #3. Above it is my native site (bi***zesitehost).

I don't quite see that. But remember (this is for "lurkers"), one has to go to the end of the google search results pages and click to show omitted results to see what I consider the actual rankings.

Just to tie up a loose end.

bn17311:

at one stage i think you mentioned they sabotaged the test, HOW ?

I said it would be easily possible to sabotage my little personal test if I revealed what I'd done. I didn't say it was sabotaged.

bn17311:

my biggest gripe is you secretly changing things on your site and not informing the minions who joined in the test, that really didnt show much respect for us believers

Firstly, I didn't make any changes to the original test product. I've stated that quite clearly several times.

I will not change anything in regards to the original test product listing, ever!. That is my 'control'. It is a known constant that I can and will use to make valid comparisons against any other change/test I choose to make. Others are still welcome to use our site as a control, with full confidence that the test product is exactly as specified when the tests began.

Secondly, what makes it so wrong for me to conduct my own tests, totally independent of anything anyone else is testing?
The change I made wouldn't have any effect or bearing on what anyone else was (or wasn't) doing.

Thirdly, the test was so simple it is almost too embarrassing to mention. :)

So, what did I do? I duplicated the WB product, and added the keywords to the description another 3 times (interspersed) with the original text. I added a paragraph of additional text that detailed what I was expecting to see from the results. (I've lost count on the number of times I've expressed the importance of doing this... A test is next to useless unless you know the expected results).

The theory, was that if "Content is King", which we all seem to agree is based on fact, then by having a single store, with two products identical in every way except the content, by 'improving' the content (by adding the keywords a mere 3 times) the only difference from googles' perspective would be the content. All other 'variables' being constant.
I was/am assuming the URL difference
http://shop.vcsweb.com/index.php?main_page=product_info&cPath=3&products_id=222 vs
http://shop.vcsweb.com/index.php?main_page=product_info&cPath=3&products_id=223
would be of little to no significance.

On this basis, the theory dictates that sooner or later, the newly added product would outrank the original product based on nothing more than the additional keywords and extra paragraph.... IOW, not only more content, but just as importantly, more relevant content.

If this were going to be the limit of my test, I would have had no reason to **not **let you all know what I'd done, and why.
I wasn't going to leave it at that though... because if the theory did prove correct, it was then a matter of waiting a little longer to see if it would eventually also outrank the other tests sites that still had the original product descriptions, as theory dictates it should. Had I made this plan known, I had every good reason that believe that someone with the original product description would/could also make the same/similar changes that would make this observation impossible.

As I say, such a simple test that it is almost too embarrassing to mention, especially if the results didn't match the theory. (Which it didn't).

At this stage I don't know whether the theory was disproven because I simply didn't make enough changes for google to place any significance in them, or, whether it is still to soon to make any conclusions, or, dare I say, Content isn't the king we've all thought it was.

Casual observation of the current top Google listings using the WB search query also seems to suggest that content isn't king either. Many of the top listings have far less references to the WB query than lower ranking sites with the more/better content.

Naturally, this goes against everything we all 'know' as being 'fact'. I've currently no explanation for it, but I'm not jumping to any conclusions either. The observations are indisputable though. We just need to find an explanation. More testing needed :)

Cheers
Rod.

NOW, I'm out of here.
Have fun guys, and once again, rest assured that the original WB test product in the vcsweb store (and indeed, the entire store itself) can be treated as a valid control for comparison purposes.. just be careful how you interpret the results. OK?

RodG:

Secondly, what makes it so wrong for me to conduct my own tests, totally independent of anything anyone else is testing?
The change I made wouldn't have any effect or bearing on what anyone else was (or wasn't) doing.

the rules make it wrong !!

RodG:

**
Rules/Instructions **

Sites going 'ZenCart Native' should simply add this product to their store, sit back and wait.

RodG:

The test product is to be placed in** ONE category only**.

test product not test products

im not here to fight with you rod and on technical knowledge youll win hands down, i have zilch knowledge but spent the time to join in

RodG:

Thirdly, the test was so simple it is almost too embarrassing to mention. :)

So, what did I do? I duplicated the WB product, and added the keywords to the description another 3 times (interspersed) with the original text. I added a paragraph of additional text that detailed what I was expecting to see from the results. (I've lost count on the number of times I've expressed the importance of doing this... A test is next to useless unless you know the expected results).

The theory, was that if "Content is King", which we all seem to agree is based on fact, then by having a single store, with two products identical in every way except the content, by 'improving' the content (by adding the keywords a mere 3 times) the only difference from googles' perspective would be the content. All other 'variables' being constant.
I was/am assuming the URL difference
http://shop.vcsweb.com/index.php?mai...roducts_id=222 vs
http://shop.vcsweb.com/index.php?mai...roducts_id=223
would be of little to no significance.

RodG:

The product name must be "Wurldbitzer Debibulator". The product description must be an unedited copy of the text above.

i thought our initial testing was about the url not about the content so this why you stressed the name and description, i know some people wernt happy with description and some people used a different description, so did this not give us 2 different descriptions to compare allready, so we now have a third description, if we were going to get anything from this test is it not being diluted by individual testing rather than us all sticking together.

bn

In the interest of keeping this thread clean for those who are interested in the results:

http://www.zen-cart.com/showthread.php?204835-Thought-Experiment

An observation that some may find of interest: 3pm Sth.Aust time:

"Wurldbitzer Debibulator" search query returns: "**About 203 **results" (with 7 pages @10 listings per page).

This means Google has suppressed/omitted 133 listings from these results. (203 - 70)

Pages 1-6 all show the "About 203 results" text. Page 7 changes this, and shows "Page 7 of 65 results". It is only on this page where we see the message "In order to show you the most relevant results, we have omitted some entries very similar to the** 65 **already displayed. If you like, you can repeat the search with the omitted results included".

Repeating the search *with the omitted results included *the query returns: "About 203 results" (with 20 pages @10 listings per page).
Pages 1-19 all show the "About 203 results" text. Page 20 changes this, and shows "Page 20 of 200 results". (I wonder where the other 3 went?)

Different people will take away different things from this. Some will have already noticed the same things I have (from prior observations/knowledge), some things will back up what we already 'know', and armed with 20/20 vision, some things are plainly obvious.

Here's what I got from this observation.

1. With or without the omitted results, the "About XXX" doesn't change. The number given here is the total number of listings that meet the search query.
2. There are FAR MORE pages being omitted than I thought there were (would be). Well over half of them not being shown with the 1st request.
3. We can calculate how many pages are being omitted at any time by noting the number of pages listed, multiplied by the number of listings per page and subtracting that from the **"XXX" **
4. The last page of any given set of results uses a different calculation than the other pages. I've no idea what this number is based on, but since the wording has changed from "about" to "of" I am assuming that this is a counter variable, rather than a calculated variable. It also has no bearing on the number of listings returned.
5. Many (most?) of the pages being omitted from the 1st query are actually of more relevance to the search term than those that aren't being omitted.

Comments:
These omitted results have always been a concern for many of us (haven't they?). It is a double edged sword, that I am currently seeing as not being good for anyone (merchants and customers alike).

It's not good for merchants (especially those that sell the same products as other merchants, and use the same, or very similar descriptions), because it is quite possible that it is your site that is being omitted from the initial listings.

It's not good for the customers (especially those that are looking for a particular product to purchase, and wish to compare prices from as many stores as possible) because many of the stores carrying the product won't be displayed in the initial listings for the exact reason that you are doing the search in the first place, namely, you/we are looking for stores with the same, or nearly the same product/descriptions.

Cheers
Rod

Although I think that the duplicate "omitted results", when clicked, shows how our test product has been ranking on the various entries, this omitted results thing is very troubling.

I just googled one arbitrary dress and and it showed over 21,000 results. A Designer (category page) over 500,000 results. I noticed one the other day over a million. Perhaps I'd be up there IF one were to wade clear to the end to show the omitted results. LOL So far it looks like one would have to do just that to see the omitted duplicate results...which, as you point out, really are likely to be the most relevant.

Undoubtedly one of our goals, in the real world, is to avoid being buried in the duplicate results. So:

1. How do the actual WB product pages that DO show in the initial listing manage to avoid being buried as duplicates? Can some study show how or why they are distinct enough to remain in the initial listings.

2. What is it about all of these ancillary links that make them unique enough to avoid being buried?

RodG:

[U]

It's not good for merchants (especially those that sell the same products as other merchants, and use the same, or very similar descriptions), because it is quite possible that it is your site that is being omitted from the initial listings.

i would put myself in this category using feeds from my supplier for my descriptions just like 100s of other shops from same supplier

my main link to the product has been omitted but one has popped up under the all products link.

so is my main product link omitted as there are no sideboxes / cross sell so this page has nothing to offer content wise than any other site listed

the link to all products has other products on this page so has some different content to the other pages in google

so should i add content to see does the page reappear ?

bn

My recent CEON entry just hit with the test product itself - original "gibberish" description. Go to show omitted duplicate results at google.

It is my site here:
http://www.prom-mart.com/debibulators/wurldbitzer-debibulator

Observations:

1. CEON jumped ahead of my Ultimate SEO and Simple SEO entries using the "gibberish" description. It also jumped ahead of my prommart.com (Ultimate SEO) using a new description.

2. But it did not have enough "umph" to jump ahead of Fez's "bizsite" (dynamic url - "gibberish" description) or the remnants of sabianstore (dynamic url - new description).

I believe that Fez has stated that his site had some extra help from his other sites? In any event, CEON may not be enough to overcome that...assuming there is any google benefit for keywords in the url.

bn17311:

so is my main product link omitted as there are no sideboxes / cross sell so this page has nothing to offer content wise than any other site listed

Could be.

so should i add content to see does the page reappear ?

I'd say now is the time to begin tweaking things, and I have already begun just last night. I added a "test" text footer banner on fashion-mart.biz and real text footer banners on prommart.com, to see if they add content and move things up.

These banners appearing on all pages. The thing with prommart.com is that I've made several, rotating through, hopefully "freshening-up" content. We'll see. Will make even more such banners.

10:35am on 1/17 my listings are still at the top of Google and my other site is still showing up, too, so it's been 72 hours. I had updated my sitemaps at the time of removing the listings and pinged the SEs so it obviously just takes time to drop off.

I guess one can learn something new everyday. Just "discovered" google's Advanced Search link at bottom of google results page. I selected to show exact test product name in title only, and got this:

https://www.google.com/search?hl=en&as_q=&as_epq=%22Wurldbitzer+Debibulator%22&as_oq=&as_eq=&as_nlo=&as_nhi=&lr=&cr=&as_qdr=all&as_sitesearch=&as_occt=title&safe=images&tbs=&as_filetype=&as_rights=#q=allintitle:++%22Wurldbitzer+Debibulator%22&hl=en&lr=&tbo=d&as_qdr=all&ei=FG34UKqyLITfqAHRg4CwAg&start=0&sa=N&filter=0&bav=on.2,or.r_gc.r_pw.&bvm=bv.41248874,d.aWM&fp=5cc0277e8548afcc&biw=1680&bih=945

Shows the entries that google has discovered the test product on and, I think, their order of ranking. Much shorter list.

what you seeing as no1, and where are you based

bn

The remnants of sabianstore are nos 1-3.

Fez's bizsite is no. 4

My prom/fashion sites are 5-8

Your cscycles is no. 9

Rod's #223 entry is no. 10

Rod's #222 entry is no. 11

Your onyourbike is no. 12

Again, results are for test product itself only. I have other entries not yet picked-up by google, as is your CEON entry not yet picked-up.

I'm in California, USA