If this is your first visit, check our documentation. To post, register or login.
Forums / Other Business Topics / Hoping to pick the brains of some old pros!

Hoping to pick the brains of some old pros!

Started by feznizzle on 21 Dec 2012, 12:30 PM · Views: 6,356

Results 21 to 40 of 624
26 Dec 2012, 9:21 PM
#21
drbyte avatar

drbyte

Sensei

Join Date:
Jan 2004
Posts:
63,512
Plugin Contributions:
176

Hoping to pick the brains of some old pros!

Feznizzle:

I notice that someone has already moved the conversation. Odd that it has become 'Private'… can only be read after logging in. What is that all about?
If it were left public, then your Wurlitzer Debobulator keywords would have become heavily indexed in a matter of hours, thus defeating the purity of your test.

.
Zen Cart - putting the dream of business ownership within reach of anyone!
Donate to: DrByte directly or to the Zen Cart team as a whole

Remember: Any code suggestions you see here are merely suggestions. You assume full responsibility for your use of any such suggestions, including any impact ANY alterations you make to your site may have on your PCI compliance.
Furthermore, any advice you see here about PCI matters is merely an opinion, and should not be relied upon as "official". Official PCI information should be obtained from the PCI Security Council directly or from one of their authorized Assessors.

28 Dec 2012, 7:53 AM
#22
rodg avatar

rodg

Deceased

Join Date:
Jan 2007
Location:
Australia
Posts:
6,263
Plugin Contributions:
4

Re: Hoping to pick the brains of some old pros!

SPH:

I've done an informal url test of sorts some time ago.

Alas, it is the 'some time ago' aspect that is coming into question here... The thing I am hoping to show/prove is that although URL's were significant 'some time ago' (5-10years), this is no longer the case.

SPH:

I think the only real test would have the exact same template, category and product names, and content. (Which my informal test may have had.) Then, as I understand it, the balance of the content on the sites would have to be the same. In short, the only difference being the url.

Although this would be the ideal test, It isn't going to be practical. Also, although we are aiming for a 'controlled' test, I believe that just the product and description should be enough to prove/disprove the claims, after all, our search query is going to be simply the product name, and page layout shouldn't have any effect on the results.
Besides, we are also going to be looking for significant benefits in using URL redirects/rewrites. If these aren't to be found then I don't see much point in using this technique considering the multitude of problems it causes for many people.

SPH:

But then there's a possibility that one, or both, may be slammed anyway for duplicate content?

Seriously, this is something I would LIKE to see happen. Why? Because it will counteract the SEO claims that URL rewriting 'does no harm'. This will be proof that it can and does do harm... in a big way.....

I'm actually a wee bit surprised you even thought of and mentioned this possibility, because whether it is factual or not, it tells me that even though you appear to promote these techniques you still fear there may be negative consequences in using them.
You are in fact backing up my opinion/comments about this potential negative effect.

SPH:

Then pop-up a Prestashop and Opencart using their default templates and the same product with their re-writing enabled. For the definitive test. ;)

Perhaps this will come a little further down the track. For the time being though it is best if we all use ZenCart because it gives a more level playing field for the URL testing.

SPH:

*Having come from the commercial Magic SEO URL, which did work. In fact, my best seasons/years were when using this on the Ic-Atheme template.

I won't argue your findings. I will argue that those years have long gone though, and the technique no longer works, has potential downfalls, and should no longer be promoted as a 'good thing to do'.

BTW, Will you be joining us in this experiment?

FWIW, I still have no idea whether these experiments will actually prove anything or not. Logic dictates that since we will be using a product that hasn't existed until now, then by searching for this product the only results returned will be for the few stores that are part of the experiment, as such we won't be able to easily observe the relative rankings between them. We may need to expand the search query to include a more commonly used word (that can be found on the test sites, and the internet at large) as this'll return a large range of results, with our test sites somewhat intermingled. Maybe.

Cheers
Rod

NOTE: We are sorry that Rod is no longer with us.
We are grateful for all his contributions to the Zen Cart community.

Ozpost - The Ultimate Shipping module for Australian Merchants. Click these links for its Homepage, or Download from Zen-Cart.com.

28 Dec 2012, 8:06 AM
#23
rodg avatar

rodg

Deceased

Join Date:
Jan 2007
Location:
Australia
Posts:
6,263
Plugin Contributions:
4

Re: Hoping to pick the brains of some old pros!

DrByte:

If it were left public, then your Wurlitzer Debobulator keywords would have become heavily indexed in a matter of hours, thus defeating the purity of your test.

This thought had crossed my mind too (I was even going to suggest a last minute change of the product name to help mitigate the issue) .... but then I decided it probably wouldn't hurt as it'll just be another 'known' site to carry this new product, and that the forum will end up trumping the stores anyway and we could just ignore those results..

OTOH, the fact that some of us have links to our stores in the forum means that this would give us an unfair advantage anyway simply because the SE's will follow the links and give those stores a slightly higher score for this fact alone.

Ohhh. So many variables...... I hope I'm not wrong in thinking we may actually prove something here......

Cheers
Rod

NOTE: We are sorry that Rod is no longer with us.
We are grateful for all his contributions to the Zen Cart community.

Ozpost - The Ultimate Shipping module for Australian Merchants. Click these links for its Homepage, or Download from Zen-Cart.com.

28 Dec 2012, 8:17 PM
#24
feznizzle avatar

feznizzle

Totally Zenned

Join Date:
Apr 2010
Posts:
900
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

Hi SPH,

Thanks for joining the conversation. I'm with you, it would be very nice if ZC would handle key word loading of URLs internally as a core function. Fingers crossed… maybe it will be in the next big version?

Thanks for the tip about EasyPopulate 4. Apparently it is not list in the ZC Addons directory for some reason. So I poked around and found it via the support link:
http://www.zen-cart.com/showthread.php?190417-EasyPopulate-4-0-Support-Thread

I will read up on it and give it a go, thanks!

Thanks,
Mike

28 Dec 2012, 8:47 PM
#25
feznizzle avatar

feznizzle

Totally Zenned

Join Date:
Apr 2010
Posts:
900
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

Ok Rod, let me take a run at convincing you once and for all that keyword loading of URLs is THE correct way to go. But first, let me address 2 points.

++++
POINT #1
I notice that you do not disagree that pretty much every search you run, there are at least one or two listings on page 1 with keyword loaded URLs.

What percentage of sites are dynamic/db driven versus static? My guess is that the VAST majority of sites are still static. Yet when you do random google searches, at least one or two listings with keyword loaded url's show up. But VERY often, the majority of listings have keyword loaded URLs. You can repeat it over and over, randomly. Try these:
Your argument is invalid
things that make you go hmmm
random search
pretty girls
SEO strategy
facts about boogers

The point of this exercise isn't that some loaded URLs are showing up, it's that tons of them ARE showing up everywhere.

This is probably because the God of all SE's is commanding webmasters to so (this is a pdf from GOOGLE):
http://tinyurl.com/google-doc-SEO

I draw your attention to Commandment #8 (page 8): Thou Shalt Load Thine URLs with Key Words

Yes. There is a caveat at the end saying don't be stupid about how you do it. But it appears to me, via actual experience, that Ceon URI does the job well. If you can show me a problem with Ceon URI methodology, I will stay away. Otherwise, seems like it does a good job doing what google asks me to do (they specifically request that I add keywords to URLs).

By the way, Google clearly states that they expect dynamic redirects. This fact seems to undo at least half of what you say here, completely invalidating your conclusion:

HOWEVER: If the website software doesn't generate URL's with keywords (for example, ZenCart's native behaviour) the only way to get a keyword packed URL is via a redirect or a rewrite. Both of these have a gross negative effect on google (and the server itself). Firstly for the simple fact the page is being redirected (an automatic loss in rankings. The target site will be the one with the higher ranking, rather than the keyword packed URL which really is nothing more than a pointer to the real page), and the second reason (one that I'm NOT convinced of, but most SEO experts seem to agree is factual) is that both pages will suffer a loss in rankings because it amounts to duplicate content.

Either way, assuming both of the negative affects are true, then clearly in this case keyword loading *is *bad.
++++

POINT #2
One of the deciding factors, many, many moons ago when I used to use Adwords heavily, was the fact that the amount bid on a particular keyword was not the only factor that governed ad placement. In other words, I could bid 10¢ per click and still receive more traffic than somebody bidding 50¢.

The reason given by Google was that they ranked ads the same way they ranked organic listings. Ads that generated more traffic (clicks) was deemed more relevant and pushed to the top.

This is a basic extension to PageRank theory. First allow other sites to vote via links (PageRank), creating a suggested group of results for a search. Then allow traffic generated to act as a vote, thus altering the original ranking via popularity of specific listings as they correlate to specific searches.

This is still fundamental, my earlier question was to see how much that has changed (if it has).

++++

Ok, here is my 'big point' that I was going to post the other night (forgive the tone, this was written before you softened the edges of your original opinion).

Very simple. Just two little images of one search for the words: "Your argument is invalid"

Click on these guys to enlarge...
Attachment 11667

Attachment 11668

The fact that keywords show up in URLs is not an accident, is not a coincidence, is not a conspiracy against guys who've been around since the web was delivered over strings attached to tin cans. This is very simply the correct way to do things in the world of SEO. Many intelligent programmers use the strategy. Just because everyone jumps off the bridge doesn't mean you should. But come on! You certainly cannot claim to be the keeper of all things Google, shake your fist at the sea, and simply dismiss everyone as stupid because they disagree with you. Can you?

I don't think that is your position. Rather, I think you would prefer this be built into ZC as a core function. I'm with you on that.

Now, before you ask the obvious, understand that I am not stating with certitude that key word loading those URLs is what caused them to reach the top. And if you go back, you will see that I did not make that claim in the beginning (you even said something to the effect of was I admitting that I did not think it helped SE algorithms). By saying I wasn't sure it influenced algorithms, that is all I meant (that I wasn't sure). I believe it is one element that may well help influence the math, but I'm NOT sure.

But that is immaterial because I am certain BEYOND any DOUBT that it DOES help ranking to put keywords in… if for no other reason than the simple fact that it causes the listing to be a tiny bit more likely to get clicked on. The more times your link is clicked on Google, the more times google see your link as being VOTED as a valid search result… thus the faster it will climb.

Lay aside wether or not keyword loaded URLs influence rankings/indexing via internal SE algorithms. Lay it aside, it is completely beside the point.

Now here is the point. Look again at that enlarged picture. Notice the use of BOLD.

Why do people make things bold? It is done with the intent of drawing attention to important things. The eye is DRAWN to things which are bold. You know instinctively know that that which is bold is IMPORTANT.

Notice that GOOGLE BOLDS OUT KEY WORDS IN URLS. You cannot argue with that. It is there, plane as the nose on my face.

If a human eye is drawn to something, it is more likely to see that which it is drawn to. In this case, a bolded out URL holds the potential to be significant. Even if it is seen just for a millisecond, it is seen. If it is clicked on, it is a voted on. The more votes it receives, the higher it climbs.

Fact: Most people don't look at the google links before clicking.

Somebody needs to send a memo over to google. Come on. Seriously?

Kidding aside, try to answer "NO" just one time to any of these questions with a straight face:
• Is BOLD used to draw the attention of a reader to significant things?
• Does Google use BOLD to draw attention to significant things?
• Does Google bold out words from searches in the listings?
• Does Google bold out words from searches in the URLs of listings?

If you answered "YES" to every question above, then you are a sane, logical person. Sadly, it also means that I have made my point and properly loading keywords into URLs does indeed help to generate traffic from google. Not only does google show those keyword loaded URLs, they take the time to bold them out so that the searcher will notice them. Clearly google LIKES keyword loaded urls. And, thus, your "core belief" turns out to be just that, a belief. Time to park it next to the Barney Rubble's jalopy. :)

If you answered "NO" to any of those questions, then your argument is similar to a creationist claiming the earth is only 4000 years old. :P

I'm sorry, I have to kid you one more time:

People need their core belief’s to be challenged, how else are they supposed to maintain their faith?

Thankfully, belief doesn't play a part here. It's science and numbers.

^ I just find it a bit ironic that science and solid facts can unravel so easy. :laugh:

I really hope you are not mad at me, I'm trying not to come off as a jerk. I've been cracking up as I write. Listen, I feel a bit like a skinny kid who got bullied into a corner, then gave 'em what for. Please recall that I tried to avoid all this! :)

28 Dec 2012, 9:10 PM
#26
feznizzle avatar

feznizzle

Totally Zenned

Join Date:
Apr 2010
Posts:
900
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

With regard to this experiment... I have a website where I have failed miserably with regard to SEO, maybe we could use it? The site is a ZC 139h w/ Ceon URI manager (paid version), sitemap xml, etc.

The reason I think it might be good fodder for the experiment is that it has more than 11000 products, plenty of things to play with. It's one cart. On one server. With one main domain.

Seems to me that those factors eliminate a lot of variables that would come into play by hosting a bunch of sites (some servers would be faster than others; some domains will be older than others; bla bla bla).

Anyway, maybe it could server as an experiment unto itself? To run concurrently with other experiments?

It is a live site, but it is mine exclusively. And I really do not care about it. The site used to (earlier version, think it was zc originally) do about 10 sales a week. Now I'm lucky to do 3 a month.

Here it is (take out the ): med-p****arts.com

Again, before you get critical about my SEO efforts on this site, please note that I am well aware that it is a glaring SEO flop!

29 Dec 2012, 10:18 AM
#27
rodg avatar

rodg

Deceased

Join Date:
Jan 2007
Location:
Australia
Posts:
6,263
Plugin Contributions:
4

Re: Hoping to pick the brains of some old pros!

Feznizzle:

Ok Rod, let me take a run at convincing you once and for all that keyword loading of URLs is THE correct way to go.

Firstly I'll repeat my stance that URL contents are** insignificant** with todays SE technology. I don't want you thinking I've changed my view with what I'm going to say (repeat/reword) next.

My arguement is, and always has been, that using URL rewrites/redirects for this purpose is totally and completely ineffective and will cause more problems than it could ever hope to solve. This is NOT a valid method of SEO optimisation.

So before progressing further, I think we need to agree that 'keyword loading' by the site/store itself (when generating its links/URL's) is not the same as keyword loading via URL rewrites/redirects performed by the server. Again, it is THIS ASPECT of 'SEO' that I'm against/opposed to.
I don't have any other issue with SEO, which again, I'll repeat is a legitimate/valid service/operation.

My 'secondary' argument is the significance (or lack of) of URL content regardless of whether store generated or apache generated.

Feznizzle:

But first, let me address 2 points.
POINT #1
I notice that you do not disagree that pretty much every search you run, there are at least one or two listings on page 1 with keyword loaded URLs.

I don't dissagree with that at all... in fact I'd expect it to be this way.. because to me it shows the insignificance of the URL contents... If it were significant then I'd be expecting the opposite in that most of the top listings would have the keyword loaded URL's.

Why isn't this the case? Why are the non-loaded URL's not only taking top spots, but also in larger numbers? To me this says plain and simple.. URL contents are insignificant.

Feznizzle:

What percentage of sites are dynamic/db driven versus static? My guess is that the VAST majority of sites are still static. Yet when you do random google searches, at least one or two listings with keyword loaded url's show up.

For the sake of arguement, I'll also agree with this... Lets be generous and say only 10% of the sites have been URL optimised via keyword loading.
We now perform a random search, and based on your findings at least one or two listings with keyword loaded url's show up, on a page of 25?
That equates to between 4-8%. Statistically speaking, if the URL had no effect whatsoever, we should still be seeing ~10% of the listings with the URL keywords, and 90% without. So, why are we only seeing between 4-8%?
If anything this is showing us that the keyword loaded URL's aren't doing as well as they should be, even with all else being equal.

For statistical equilibrium we need to assume 15%-20% of all sites are using keyword loaded URL's.

Feznizzle:

The point of this exercise isn't that some loaded URLs are showing up, it's that tons of them ARE showing up everywhere.

Yes, but statistically, not with any greater frequency than the proportionate amount of non keyword loaded URL's.

If you wish to convince me the value of keyword loading URL's I going to need to see a LOT more keyword loaded sites taking top spots, not the statistically below average that we both seem to agree we are seeing.

Feznizzle:

This is probably because the God of all SE's is commanding webmasters to so (this is a pdf from GOOGLE):
http://tinyurl.com/google-doc-SEO

I draw your attention to Commandment #8 (page 8): Thou Shalt Load Thine URLs with Key Words

I was wondering when someone was going to point me to this document.

I should have forwarned you, I get a great kick at finding errors and omissions in official documents, text books, etc.

I've even found errors in one of the better known networking text books (by Cisco, of all people). It took me ages to convince my lecturers that the book was wrong. The error (a rather serious one at that) was corrected in an updated release. :)

What I'm saying here is that I don't take all text at its word, regardless who wrote it. I've learned to verify facts from several independent sources before accepting it as gospel. I ask that you should also keep a somewhat open mind.

Oh, I'm writing this without even knowing if there are errors or omissions anyway... I hope I don't find any for the 'obvious' reasons. <g>

OK, from page#8.

Creating descriptive categories and filenames for the documents on
your website can not only help you keep your site better organized,

Agreed. Common sense!.


but it could also lead to better crawling of your documents by
search engines.

Note... it states "but it could"... not "but it DOES", or "but it WILL"

Surely, if it DOES, then Google would say IT WILL.


Also, it can create easier, "friendlier" URLs for those
that want to link to your content.

Agreed..... but as I often argue, 'Human Friendly' and 'SE friendly' are more often than not at odds with each other.

If the discussion is about whether HUMANS preferring keyword loaded URL's rather than cryptic URL's then this is a no brainer. Humans do prefer the keywords (if/when they see them). The SE's don't give toss either way.


Google is good at crawling all types of URL structures, even if they're
quite complex,

IOW, Google treats/crawls all URL's equally.....


but spending the time to make your URLs as simple as
possible for both users and search engines can help.

Note the wording again... 'it can help. Not it 'does' help.

If Google aren't confident enough to say whether this WILL help or not, then I'm certainly not going to be confident about it either.

The rest of Page#8 goes on to discuss the human aspect, which really has nothing to do with SE rankings.

If this is the best you have to offer then I'm afraid you are reading far to much into what this text is actually saying.
Yes, I realise that I may be nitpicking of the wording here, but please understand that I have a very technical academic background, and certain words and phrases such as 'will, won't, must, must not, should and should not' do have very explicit meanings. One of the first things I learned when creating technical docs is to avoid using words like 'may' or 'could' because they lead to misinterpretation and/or doubt as to what is being stated.

Feznizzle:

By the way, Google clearly states that they expect dynamic redirects.

And rightly so. Dynamic redirects have been a valid part of the HTML specs since way back with HTTP V1.1 (?). It wouldn't be much of a SE if it couldn't follow them.

Google doesn't say the redirects are good for SEO.... at least not in anything I've read.

Don't forget, it is still for you to show that this SEO is good and benefical... not for me to show it isn't.

Feznizzle:

This fact seems to undo at least half of what you say here, completely invalidating your conclusion:

How so? I stated that in this case Google will index the site directed TO, not the one directed FROM. I also stated that this is very inneficent, and it is best to NOT have the redirect in the first place.

I don't see how this invalida

Feznizzle:

POINT #2
One of the deciding factors, many, many moons ago when I used to use Adwords heavily, was the fact that the amount bid on a particular keyword was not the only factor that governed ad placement. In other words, I could bid 10¢ per click and still receive more traffic than somebody bidding 50¢.

I was hoping/expecting to leave Adwords out of this because of the $$$ factor alone. Given enough funds any site can appear at the top of google for any search query... it is hardly grounds for a discussion about SEO, and a long way from the aspect of SEO that we are currently discussing.

Feznizzle:

The reason given by Google was that they ranked ads the same way they ranked organic listings. Ads that generated more traffic (clicks) was deemed more relevant and pushed to the top.

OK, I'll accpet this claim. I still don't see how it is going to be relevant to our experiments though.

Feznizzle:

This is a basic extension to PageRank theory. First allow other sites to vote via links (PageRank), creating a suggested group of results for a search. Then allow traffic generated to act as a vote, thus altering the original ranking via popularity of specific listings as they correlate to specific searches.

Again, I don't see how this is going to be relevent to our experiments. We are not trying to prove the benefit of SEO, we all agree it is generally a good thing.
We are trying to prove the benefits (or lack thereof) of UTL rewriting as a valid method of SEO. Let's not lose site of this fact.

Feznizzle:

Ok, here is my 'big point' that I was going to post the other night (forgive the tone, this was written before you softened the edges of your original opinion).
Very simple. Just two little images of one search for the words: "Your argument is invalid"

Yoiks.... Methinks you just blew it. ... read on....

Feznizzle:

The fact that keywords show up in URLs is not an accident, is not a coincidence, is not a conspiracy against guys who've been around since the web was delivered over strings attached to tin cans.

Agreed.... It is infact something to be expected because the SE's do not differentiate a keyword loaded URL from a non keyword loaded URL... to Google, they are simply URL's... nothing more, nothing less.

Feznizzle:

This is very simply the correct way to do things in the world of SEO.

No, it has nothing to do with SEO whatsoever. If it WERE SEO relevent then all the top spots would be keyword loaded. This simply isn't the case.

Feznizzle:

Many intelligent programmers use the strategy

Programmer != Network Engineer.
Programmer != Server administrator.

I happen to have Nationally recognised certifaction for all three. :)

Feznizzle:

. Just because everyone jumps off the bridge doesn't mean you should.

This is coming from someone that has followed the crowd and jumped on the bandwagon that wrongly beleives URL rewriting is beneficial to SEO.

I'm not the person that does what everyone else does... I'm the antagnosist.. I will speak against the majority if/when I beleive I'm right.

Feznizzle:

But come on! You certainly cannot claim to be the keeper of all things Google, shake your fist at the sea, and simply dismiss everyone as stupid because they disagree with you. Can you?.

Err, Hmmm.. Well, Yes, yes I can.
Well, no, I don't dismiss everyone as being 'stupid' because I don't beleive that is the case. Misinformed a more apt term I'd use.

Feznizzle:

I don't think that is your position. Rather, I think you would prefer this be built into ZC as a core function. I'm with you on that.

Yes, I would like to see this as a core function, but NOT for any misplaced SEO reasons, but simply for the HUMAN factor.

Also, IF keyword packing URL has any benefit at all, it is only via this method that it can possible work. It cannot work via redirects.

Feznizzle:

But that is immaterial because I am certain BEYOND any DOUBT that it DOES help ranking to put keywords in…

Good, at least we are still in opposite sides of the fence... I really wasn't happy when you seemed to back off and stated you wasn't really sure.

Feznizzle:

if for no other reason than the simple fact that it causes the listing to be a tiny bit more likely to get clicked on. The more times your link is clicked on Google, the more times google see your link as being VOTED as a valid search result… thus the faster it will climb.

I've no idea what the 'weights' are for any given click, but for our experiments to be valid we will have to agree to avoid needless clicks to our own product so as to avoid this bias.

Feznizzle:

Lay aside wether or not keyword loaded URLs influence rankings/indexing via internal SE algorithms. Lay it aside, it is completely beside the point.

No, that is one of the main points. Is keyword packing beneficial or not.

Feznizzle:

Now here is the point. Look again at that enlarged picture. Notice the use of BOLD.
Why do people make things bold? It is done with the intent of drawing attention to important things. The eye is DRAWN to things which are bold. You know instinctively know that that which is bold is IMPORTANT.

Notice that GOOGLE BOLDS OUT KEY WORDS IN URLS. You cannot argue with that. It is there, plane as the nose on my face.

Guess what. I am going to argue with that... No, not about the bold being to draw attention, but the comment GOOGLE BOLDS OUT KEY WORDS IN URLS.

The thing is, those aren't KEY WORDS that are being bolded... Those are the words in the SEARCH QUERY.
You'll also notice that they aren't just bolded in the URL's, but are in fact bolded in every place they are found on the results page(s).

It really doesn't tell us a thing about the 'key words in URL' being of any benefit in regards to the page ranking. In fact it really doesn't tell us anything useful at all.. it is just a means to draw the users attention to the portion(s) of the pages that contain their search query.
If you see it as anything more than that you are reading far more into it than there is.

Feznizzle:

If a human eye is drawn to something, it is more likely to see that which it is drawn to. In this case, a bolded out URL holds the potential to be significant. Even if it is seen just for a millisecond, it is seen. If it is clicked on, it is a voted on. The more votes it receives, the higher it climbs.

OK, but the same could be said if/when a person clicks on the link because the 'preview' line has been bolded due to the same match (regardless of whether it is also in the URL or not), so once again, this isn't telling us anything about whether the keyword packed URL has any bearing on ranking.

Feznizzle:

Kidding aside, try to answer "NO" just one time to any of these questions with a straight face:
• Is BOLD used to draw the attention of a reader to significant things?
• Does Google use BOLD to draw attention to significant things?
• Does Google bold out words from searches in the listings?
• Does Google bold out words from searches in the URLs of listings?

If you answered "YES" to every question above, then you are a sane, logical person.

Oh goodie. Apparently I'm a sane, logical person.

Feznizzle:

Sadly, it also means that I have made my point and properly loading keywords into URLs does indeed help to generate traffic from google.

Alas, the only point you've made is in somehow making a connection between "key words" and "search query highlighting" when there is no such connection.

Feznizzle:

Not only does google show those keyword loaded URLs, they take the time to bold them out so that the searcher will notice them. Clearly google LIKES keyword loaded urls.

This is a HUGE leap you are making.... It is clear that Goolge likes to highlight the search queries that it is given by making them BOLD in the results, but it does this no matter where the search query is found on the page. IF it only highlighted the ones in the URL's then you might possibly be onto something, but that isn't the case... it ISN'T highlighing the keywords as you state... it is highlighting the search query. Please tell me that you understand the difference.

Feznizzle:

And, thus, your "core belief" turns out to be just that, a belief. Time to park it next to the Barney Rubble's jalopy. :)

Alas, it has been you that has made this leap between two disparate items and you have done so to back up your belief. It say nothing about mine/

Feznizzle:

If you answered "NO" to any of those questions, then your argument is similar to a creationist claiming the earth is only 4000 years old. :P

I thought it was 6000? 4000bc + 200ad ... unimportant.... just another fairytale, just like URL rewrites being a good thing <g>

Feznizzle:

I'm sorry, I have to kid you one more time:
^ I just find it a bit ironic that science and solid facts can unravel so easy. :laugh:

I've provided a little statistal data at the start of this post that seem to prove my stance.

you still haven;t unravlelld anything... Shame about that huge leap of faith you needed to make though... But then again, without this, your comments don't hold water do they :)

Feznizzle:

I really hope you are not mad at me, I'm trying not to come off as a jerk. I've been cracking up as I write. Listen, I feel a bit like a skinny kid who got bullied into a corner, then gave 'em what for. Please recall that I tried to avoid all this! :)

Not mad at all.... Not even frustrated (which is quite surprising for me)... In other words, I'm still really enjoying this discussion... I really hope you are too.

Cheers
Rod

ps. Sorry about all the typos.

NOTE: We are sorry that Rod is no longer with us.
We are grateful for all his contributions to the Zen Cart community.

Ozpost - The Ultimate Shipping module for Australian Merchants. Click these links for its Homepage, or Download from Zen-Cart.com.

29 Dec 2012, 1:04 PM
#28
rodg avatar

rodg

Deceased

Join Date:
Jan 2007
Location:
Australia
Posts:
6,263
Plugin Contributions:
4

Re: Hoping to pick the brains of some old pros!

Feznizzle:

With regard to this experiment... I have a website where I have failed miserably with regard to SEO, maybe we could use it? The site is a ZC 139h w/ Ceon URI manager (paid version), sitemap xml, etc.

I've no objection if you use this site... I'd much rather you used one where you are happy with the SEO though.. I don't want you coming back after being proven wrong claiming it wasn't a fair test because you didn't optimise it correctly.

Feznizzle:

The reason I think it might be good fodder for the experiment is that it has more than 11000 products, plenty of things to play with. It's one cart. On one server. With one main domain.

I suspect the sheer quantity of products on this site will give it the edge over my proposed test site, which has less than 100 products, purely from the number of links available back to itself. Will I use this as an 'excuse' if my store does badly though? I think that will depend on how badly, and exactly what results we see. If there is clear evidence that the results have been swayed by this I do have a client site also with about 10000 products which I'll seek permission to add the test product.
Incidentally, this client's site also doesn't use any form of redirection/rewrites, but is constantly in the top 3 of google and averages about $15,000 in sales per month.

Feznizzle:

Seems to me that those factors eliminate a lot of variables that would come into play by hosting a bunch of sites (some servers would be faster than others; some domains will be older than others; bla bla bla).

I agree that the age of the site may have some relevance in the rankings, but my gut feeling that all else being equal, this will also be insignificant.

I vehemently disagree that the server speed would have any effect on ranking though. It would be a mistake for Google to even take this into account because speed shouldn't have bearing on the relevancy of any given search query.

Feznizzle:

Anyway, maybe it could server as an experiment unto itself? To run concurrently with other experiments?

Sure, why not?

Feznizzle:

It is a live site, but it is mine exclusively. And I really do not care about it. The site used to (earlier version, think it was zc originally) do about 10 sales a week. Now I'm lucky to do 3 a month.

FWIW, our store only does about two sales a year (excluding ozpost subscriptions). This is mainly because it has a lot of ######## that no one wants.. it's mainly used for testing and development rather than trying to sell stuff (we haven't added any new products for over two years), so even though it is a 'dead' site, I'm still confident enough to expect it to make top billing with the debobulator.

What makes me so confident? Easy. Google doesn't play 'favourites'. It's prime purpose in life is to provide the most relevant results for any given search query.
This being the case, unless there are other significant factors at play, the test item in my site (no SEO) should get pretty much the same ranking as any other site for this same item, keyword packed URL's included.

If/when we find this isn't the case... for example, if my store is nowhere to be seen, then we'll be able to look deeper and perhaps understand why, and hopefully even isolate the component that has made the difference.

Feznizzle:

Again, before you get critical about my SEO efforts on this site, please note that I am well aware that it is a glaring SEO flop!

The only SEO component of interest (at this stage) is the URL redirection/rewriting.

Almost all other SEO components are based on well known and proven technology, namely GOOD CONTENT (equal ground in our test case), Meta-Tags (created specifically for SEO), Sensible filenames (common sense, takes care of the human element. Not applicable in our test case). Sensible 'page' titles (not to be confused with URL's).

I've stated many times, I've no idea how much 'weight' google gives to any given 'item of interest', but if these 4 things alone don't make up at least 90% of the total possible score/ranking for any given search result, I think it'll be time for me to hang up my hat and take up gardening. All other SEO techniques combined will make the final 10% difference.. IOW, they could sway the balance between two otherwise equal rankings. (note, I'm not willing to say whether the sway will be for or against any given site).

In short, I'm still not expecting the keyword loaded URL's to provide any significant or noticeable benefit.

Cheers
Rod

NOTE: We are sorry that Rod is no longer with us.
We are grateful for all his contributions to the Zen Cart community.

Ozpost - The Ultimate Shipping module for Australian Merchants. Click these links for its Homepage, or Download from Zen-Cart.com.

29 Dec 2012, 1:41 PM
#29
feznizzle avatar

feznizzle

Totally Zenned

Join Date:
Apr 2010
Posts:
900
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

Good, very glad to hear you are enjoying the conversation. I am too, otherwise I would have quit as soon as I realized the conversation went private.

Ya know, sometimes I find it easier to respect a believer than an atheist… though my personal leanings are toward the ladder. In my younger days I was motived to question people of faith on both sides of a coin. At some point a believer will surrender without a change of mind, usually defaulting to something about faith. At that point, they tell me to @#$ off, take my science and shove it. Looking back now, I can tell I never convinced a believer to question. And, having matured a bit, I'm glad of that.

Atheists, on the other hand, usually default to something about the impossibility of proving a negative and tell me that it is my responsibility to prove that Santa is real, not theirs to prove he is not. But they never get that there is a third possibility.

Why do I respect believers more? Because at least there is the acknowledgement of the human condition (impossibility of perfect knowledge). BTW, you mentioned something about that previously, meant to clarify (aka perfect information). It's a principal in economics, comes from game theory. As applied, basically at a certain level, human understanding becomes irrelevant.

Google "perfect information". Just look for the first bolded out URL. Wikipedia made sure to include the key words in the URL to make it easy to find! ;-)

Anyway, a believer will acknowledge his/her own limitations, fall back on faith. An atheist (translation: believer in denial) will argue until blue in the face that the known religions on earth are 99.9999% improbable, therefore there is no God. Further, this logic can be applied to the jillions of stars out and the bijillions of planets circulating those stars times the percentage of possibility of life, times the percentage of sentient beings on those planets, times the percentage of those beings who have developed faiths. times the KNOWN percentage of correct religions on our planet to arrive at a statistical impossibility that we don't already know EVERYTHING there is to know.

So a believer believes honestly. An atheist believes no truth exists outside of his/her constructs. Though the atheist is probably correct, I can't help a bit of resentment in that atheist generally can't see themselves in the math.

In economics, imperfect information points out a few GAPING wholes in mathematical formulas used to predict/understand markets… from micro to macro. There are tidy little formulas to predict small things like the individual value of a company all the way up to large market fluctuations. But there is no way to know every possible outcome (right down to whether or not the market will behave rationally). Is Apple over valued, especially since the beloved one has departed? Maybe. But what if AppleTV comes out this year? Will people buy or sell based upon information in a rational way? You can hedge bets, maybe, but not even the Oracle of Omaha is 100% right.

In religion (particularly the devout belief in no higher order directive), I use it to point out that an atheist's conclusion is on equal par with a Harry Krishna.

++++

I am now officially surrendering on this aspect of our conversation because it appears we are at an impasse. You seem to have completely missed my point, I'm not sure further words could clarify.

Before I lie down, let point out a few parting thoughts. I promise to read any response you may post, but I will not respond any further (other than to say that my client asked me to do it, therefore I do it).

First, you may have an editorial-ish background, I do not, sorry. I read to understand, not to pick apart (unless ideas presented warrant). It seems kind of unfair to use semantics to pick apart a point, especially since this is anything but an academic setting.

You spend a lot of time distinguishing between KEY WORDS and SEARCH QUERY. This seems pretty silly to me. The point of intelligent SEO is understanding key words and applying them. The most important keywords describe the content of a page. The most important of those keywords are likely to appear in the title so that the reader knows what to expect. Those are also the same words that a reasonably intelligent person would want in the URL. To a searcher, the keywords are those chosen for the query.

So to my way of thinking (and pretty much everyone else):
KEYWORDS vs SEARCH QUERY = POTATO vs POTATO

Second, google does in fact bold out search query words (read: keywords) in URLs (from the pdf):

Lastly, remember that the URL to a document is displayed as part of a search result in Google, below the document's title and snippet. Like the title and snippet, words in the URL on the search result appear in bold if they appear in the user's query (3).

Definition from glossary (also from the pdf):

An HTML tag denoting strong emphasis. According to standard, it will indicate emphasis through use of bold print.

I guess that is all I have. I hereby vow never to engage in this conversation again on ZC forums. I have enjoyed it, helped me to think it thru… but it is too much work! :)

I will, however, happily engage you will experimenting. You have alluded to worry that I might engage in activities that would tilt the balance and skew results. But that defeats the point of experimenting. If I wanted to prove a point rather than discover some deeper fact… well, what would be the benefit to me? I find you an engaging fella, but there are more pleasant ways to ##################e. :)

Ok, I'm done. Maybe we can start another thread to invite broader participation? You can incorporate this conversation via a link in case somebody wants to punish themselves by reading our blatherings. :)

29 Dec 2012, 1:48 PM
#30
feznizzle avatar

feznizzle

Totally Zenned

Join Date:
Apr 2010
Posts:
900
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

With regard to site speed, here is an old article. It is a bit dated, but nevertheless still relevant:
http://tinyurl.com/not-sure-why

I am more or less in agreement with you on the 4 things that you find most important, though not sure enough to give percentages. Certainly server speed is probably pretty minor. :)

EDITED: for some reason ZC forum did not allow me to post the original url

29 Dec 2012, 2:12 PM
#32
rodg avatar

rodg

Deceased

Join Date:
Jan 2007
Location:
Australia
Posts:
6,263
Plugin Contributions:
4

Re: Hoping to pick the brains of some old pros!

Re: Site speed & rankings.

I humbly stand corrected on this one. Totally surprised, but further enlightened.

Thanks for the links.

Now, back to the URL rewriting..... This is still the big one I have 'issues' with, simply because it still causes more people far more problems than any benefits they hope to gain.

Cheers
Rod

NOTE: We are sorry that Rod is no longer with us.
We are grateful for all his contributions to the Zen Cart community.

Ozpost - The Ultimate Shipping module for Australian Merchants. Click these links for its Homepage, or Download from Zen-Cart.com.

29 Dec 2012, 3:16 PM
#33
rodg avatar

rodg

Deceased

Join Date:
Jan 2007
Location:
Australia
Posts:
6,263
Plugin Contributions:
4

Re: Hoping to pick the brains of some old pros!

Feznizzle:

Ya know, sometimes I find it easier to respect a believer than an atheist…

Shoot... We'd been doing so well at not making this a 'religious' issue.
Sorry, but I'm not going to respond to this part of the conversation, other than the 'proof' issue.

Like it or not, if you (or anyone) is going to promote a technology as a service then it MUST be something that can be proven, otherwise it's just 'snake oil'.

The purpose of this ongoing thread is to see if we CAN prove the technology. If it can't be proven, then it must be assume it simply does work.
This is logic. Not faith.

Feznizzle:

First, you may have an editorial-ish background, I do not, sorry. I read to understand, not to pick apart (unless ideas presented warrant). It seems kind of unfair to use semantics to pick apart a point, especially since this is anything but an academic setting.

So now you are implying that the people running Google and writing their documents are NOT academics, and therefore aren't aware of what you consider 'semantics'.

I wasn't having a go at your misinterpreting what was written, I was having a go at the google doc writer for not being confident enough to be able to state "it will" rather than "it may".
As I said previously, if THEY aren't confident enough to say "will" then I'm not going to say it either, and NEITHER SHOULD YOU (but you do anyway).

Feznizzle:

You spend a lot of time distinguishing between KEY WORDS and SEARCH QUERY. This seems pretty silly to me.

I just knew you wouldn't 'get' that one.

Try this. A 'key word' is something the site owner thinks is important to drive visitors to the site.
A 'search query' is what the consumer enters into the search engine when looking for a particular term, phrase or item.

You are assuming these are both the same. In many cases they are, in many other cases they aren't.

Example: Keyword loaded URL
http://mysite.com/teddy_bear/
The keywords obviously 'teddy' and 'bear'

Consumer enters search query 'Yogi Bear'

Results returned should include results like

http://mysite.com/teddy_bear/
plus a few lines of text which may or may not include teddy bear

and


http://mysite.com/index.php?main_page=product_info&cPath=2_9&products_id=124
plus a few more lines of text, assumedly containing Yogi and Bear.

Happy so far? OK, now lets add the BOLDING that Google would apply:

http://mysite.com/teddy_bear/
plus a few lines of text which may or may not include teddy bear

and

http://mysite.com/index.php?main_page=product_info&cPath=2_9&products_id=124
plus a few more lines of text, assumedly containing** Yogi **and Bear.

Now 'Teddy', which we have established as being a KEY WORD is not being bolded. The reason is obvious, it is NOT part of our SEARCH QUERY.
On the other hand, Yogi, which has NOT been established as a KEY WORD IS bolded because it IS part of the SEARCH QUERY.

If you are unable to make this distinction, then there really is no such thing as a 'key word' at all.. it is all just words... a merchant can't really add a keyword unless they know in advance what a person is searching for, and it is clear that google isn't highlighting 'key words' it is just highlighting the search query.

Feznizzle:

So to my way of thinking (and pretty much everyone else):
KEYWORDS vs SEARCH QUERY = POTATO vs POTATO

If you search for something like "this that and the other", none of which could possibly be considered as a 'key word', Google still returns a ton of results, all nicely bolded where the search query matches. Do you still say that this is the 'same as bolding key words', are would you be happy saying it has simply highlighted the search query?

Not that this really matters,,, the bolding is for human benefit only. A factor we wish to eliminate as part of our testing.

Feznizzle:

Second, google does in fact bold out search query words (read: keywords) in URLs (from the pdf):

I never stated anything different. It also does the same on the non URL portions of the page(s) shown. All this is showing is that the 'keywords' (or if you prefer. search query) in the URL's are being treated no differently than the search query being found elsewhere on the page. This is a point in my favour, not yours.
You would have scored this point if it were ONLY the URL keywords being bolded, because this would indicate the URL contents were ranked differently than the same keywords used in page content.

Feznizzle:

I guess that is all I have. I hereby vow never to engage in this conversation again on ZC forums. I have enjoyed it, helped me to think it thru… but it is too much work! :)

LOL. Take this on the chin..... but IF this SEO technique was as effective as the experts would have us beleive, it would be EASY to prove. It really shouldn't be hard to demonstrate a technology that works (it is a lot harder to demonstrate that it doesn't work).

Feznizzle:

I will, however, happily engage you will experimenting.

Good.... I was afraid you were going to chicken out.... Can't say I'd blame you....

Feznizzle:

You have alluded to worry that I might engage in activities that would tilt the balance and skew results.

Not so, at least not intentionally. You'd pointed out a thing ot two that may skew the results, and I suggested that being the case we should avoid doing it.

Feznizzle:

But that defeats the point of experimenting.

Agreed.

Having said that though, I'm still so confident in the Google algorithms to produce a level playing field where all else aside 'content is king' that even if either of us did try to skew the results it would probably backfire on us anyway, so I'm not too fussed.

Anyway, like you, I'm pretty much all talked out here... The experiments will probably give us both a lot more to discuss and consider... so I hope to remain pretty quiet until them,.....

Cheers
Rod

NOTE: We are sorry that Rod is no longer with us.
We are grateful for all his contributions to the Zen Cart community.

Ozpost - The Ultimate Shipping module for Australian Merchants. Click these links for its Homepage, or Download from Zen-Cart.com.

29 Dec 2012, 4:11 PM
#34
feznizzle avatar

feznizzle

Totally Zenned

Join Date:
Apr 2010
Posts:
900
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

If you search for something like "this that and the other", none of which could possibly be considered as a 'key word', Google still returns a ton of results, all nicely bolded where the search query matches. Do you still say that this is the 'same as bolding key words', are would you be happy saying it has simply highlighted the search query?

Not that this really matters,,, the bolding is for human benefit only. A factor we wish to eliminate as part of our testing.

You have completely missed my point. You are a smart guy. I don't expect you to go back and reread the post where I outlined my main train of thought (post begins with knocking out 2 quick points b4 gettin to the point)... but I really wish you would. Really. And if you do, just let points you disagree with slide for the moment, take it as a whole.

If you don't see "human factor" as integral to what I said, then I have failed to effectively use the English language. Not being sarcastic. I have many short comings, writing is but one.

+++++

Setting that aside, I am actually interested in your point of view. Very. Though it has less to do with why I use this strategy.

Please lay out ALL control aspects of what you would like to do:
Page title, content
Meta?
Sitemap xml?
etc, etc

Once we are sure of the task, let's pick a time that works mutually (morning down under, evening in the sunshine state, whatever).

I'd like to see the first go as very minimal.

30 Dec 2012, 2:18 AM
#35
sph avatar

sph

Totally Zenned

Join Date:
Jan 2006
Posts:
1,556
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

RodG,

Looks like my informal test was fairly recent, July 2011. Found my post at this earlier CEON thread (posts 2846, 2848):

http://www.zen-cart.com/showthread.php?117613-Ceon-URI-Mapping-%28SEO%29&p=1044614&highlight=sph#post1044614

Hmmmm. My recollection is that CEON went ahead of the product on my dynamic url site (my main site, prommart.com). Perhaps it did later. Also, I looked at the "way back maching" and apparently my best seasons were when using the Aqua Shade template (basically, Classic), and not Ic-Atheme as I mostly remember. But as Confucius says: a short pencil is better than a long memory. :)

But I've had more recent cases where same product on Ultimate SEO URL was ahead of dynamic url. But then sometimes the dynamic url is ahead.

Here's something. Last season a competitor took out an exact match domain for individual dresses, using WP that eventually went to her ecommerce site. Now, I considered that a classic, black-hat doorway page, but it worked for her as she was often number one on Google. But I took one dress and gave it the best product description I have ever done. And while on dynamic urls, my dress on prommart.com was often number one above her. If not, then number two. So I know that dynamic urls can work.

Beating or holding my own with a dynamic url vs an exact match domain name with WP and re-written url.

Unfortunately, this dress turned out to be a bomb and nobody was looking for it.... LOL

Re duplicate content in the experiment, I meant that in my scenario with both sites having virtually idential content, it might be considered duplicate content to google.

I do think that re-written urls can look spammy, but that may be because the category names and product names are themselves spammy. I'd guess that even with dynamic urls, spammy category/product names aren't good.

Later on I realized that my idea of having idential sites, the only difference being the url, wouldn't work. Because the very domain name could be a ranking factor and it would be difficult getting identical domain names. Only possibility would be two carts on subdomains with nearly identical subdomain names: cart-1 and cart-2

Wouldn't mind joining the experiment if offer is open. I would consider putting the test product on my main site, prommart.com. I have a couple of secondary carts not really doing anything. Then I have a couple of test zencarts and should still have a Prestashop and Opencart, both relatively recent versions. But I recently restored my computer back to factor settings so would have to hunt them down online.

I am neither for nor against re-writing. Just what works best for search engines and customers. And myself. While on dynamic urls I missed being able to read my own links when looking at them. This may have been my main reason for going back to re-writing.

In the end it has to be a business decision. Just as it is with the cart itself. As much as we love Zencart, should a head-to-head experiment show Prestashop or Opencart superior for search engines and customer satisfaction, that would seem to settle it.

Steve
prommart.com

30 Dec 2012, 2:31 AM
#36
sph avatar

sph

Totally Zenned

Join Date:
Jan 2006
Posts:
1,556
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

Feznizzle:

Hi SPH,

Thanks for joining the conversation. I'm with you, it would be very nice if ZC would handle key word loading of URLs internally as a core function. Fingers crossed… maybe it will be in the next big version?

Yes, I suspect that if zencart came with rewriting that this "religious war" would mostly die out. Rewriting then would presumably be reliable with zencart upgrades and mods; little worry about breaking things. I think that 90+% of zencart users would turn it on and that the dynamic url only devotees would have a tougher time convincing people not to turn re-writing on.

But I doubt that built-in re-writing will happen anytime soon.

You're welcome re Easy Populate 4. Since using it I've come to prefer it.

Steve
prommart.com

30 Dec 2012, 3:01 AM
#37
sph avatar

sph

Totally Zenned

Join Date:
Jan 2006
Posts:
1,556
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

RodG:

Also, although we are aiming for a 'controlled' test, I believe that just the product and description should be enough to prove/disprove the claims, after all, our search query is going to be simply the product name, and page layout shouldn't have any effect on the results.

I've long suspected that the template just may be a factor. And especially in the case of templates such as Apple Zen, which is "source ordered", putting the product name and description toward the top of view source, and the categories at the bottom. But I suppose that almost any template can achieve this, using only a right-side column.

I've tested Apple Zen and still have it loaded to go on fashion-mart.biz. It is probably the fastest template I've ever tested, and I've tested many. But I never got to the point to testing is source ordered actually works. I even think from time to time about posting on that question.

But I suppose it depends upon what "content" you want to feed first to google: category links or product name and description.

Steve
prommart.com

30 Dec 2012, 1:43 PM
#38
rodg avatar

rodg

Deceased

Join Date:
Jan 2007
Location:
Australia
Posts:
6,263
Plugin Contributions:
4

Re: Hoping to pick the brains of some old pros!

Guys, I'll respond to some of the other post tomorrow, but meanwhile, I'm starting to see where a lot of the confusion (and misunderstanding) is coming from.

SPH:

Yes, I suspect that if zencart came with rewriting that this "religious war" would mostly die out..

Assuming you SEO guys have the same thinking as evidenced here, then we can possibly end this discussion here and now without any testing whatsover (although I still think it'll be good to do because it will help solidify the facts.

I'm not sure how I can say this without risking offence, but the comment "if zencart came with rewriting" doesn't make any sense whatsoever.
The fact that you stated it is showing is that you misunderstand how redirects/rewrites actually work.

Please try to not be offended, because this bites deep into the core of the discussion. If you are unable to distinguish the difference between a 'redirect' and the site created URL's then this** entire discussion** becomes pointless.

Simply put.
A redirect/rewrite is a function of the server software. (eg: Apache)
The site created URL's are a function of the application (eg: ZenCart).

When a request is made to a server, it is the server itself that gets the 1st chance at processing the request (not the application). If the server performs a redirect it is a 'bad thing'. It comes at a price... additional DNS lookups, etc ,etc.

If this same request doesn't hit the server redirect, the application simply serves the data requested. It makes no difference if the URL is user friendly or a complex dynamic one. In either case there will be NO redirection taking place.

What this means is that there are three possible scenario's

A) http://mysite.com/user_freindly_URL (With Apache redirect to http://mysite.com/complex_URL)
B) http://mysite.com/user_freindly_URL (Application generated)
C) http://mysite.com/complex_URL (Application generated)

Scenario A) BAD NEWS. AVOID. This is what I'm trying to get across to you (and others).

Scenario's B & C: Both GOOD. It's just that one is more human friendly than the other.

It appears to me, that you SEO guys are discussing the advantages/disadvantages between scenario's B & C only (which is always going to be a subject of hot debate because the difference it makes is arguably negligible).

It also appears that you are considering scenario A) as being the same or equal to scenario B). This is wrong. This is what I'm trying to elaborate on.
These scenarios are NOT equal, but virtually every 'SEO expert' I've spoken to so far fails to distinguish the difference.

So, here's the problem, you guys have 'learned' that 'human friendly' URL's are 'good' for search engine rankings. I'll not really argue this point, mainly on account of the fact that I kinda like seeing them too, So, what do you do? You realise ZenCart doesn't generate these 'friendly' URL's but since they are apparently a "good thing to have", you end up 'faking' them with the redirects... In other words, you are taking scenario 'A' and attempting to make it behave like scenario 'B'.

You are quite happy doing this, and you think you are doing the right thing, because it is now possible for a 'human' to enter the friendly URL in their browser and end up on the page with the complex URL. All appears fine and dandy, but lets take a closer look at what goes on 'behind the scenes' with these three scenarios. Firstly from the 'human' perspective.

A) Person enters http://mysite.com/user_freindly_URL into their browser, the browser sends the request to the server, the server sends a 301 redirect (with complex URL) back to the client, the client performs a second request to the server using the complex URL, as there are no further redirects the server passes the request to the application (eg, ZenCart) which then sends the final page back to the client. (You can't tell me that this doesn't come without some sort of hit on performance).

B) Person enters http://mysite.com/user_freindly_URL into their browser, the browser sends the request to the server, the server doesn't see any redirects and passes the request over to the application (eg ZenCart) which sends the requested page back to the client. Clean and simple.

C) Person enters http://mysite.com/complex_URL into their browser, the browser sends the request to the server, the server doesn't see any redirects and passes the request over to the application (eg ZenCart) which sends the requested page back to the client. Clean and simple.

Now, lets look at the same scenarios from Googles' perspective, but before doing so, let us also make a wild (and probably invalid) assumption that google has managed to index these sites and will show them all when someone does a search for 'Wurlitzer Debobulator'. These are the search results:

C) http://mysite.com/complex_URL
B) http://mysite.com/user_freindly_URL
A) You tell me... Will it be http://mysite.com/user_freindly_URL, the redirected http://mysite.com/complex_URL or Both?

Please give a reason/explanation for your answer.

By this time, it should be quite obvious that scenario A is a completely different kettle of fish than scenario's B & C. Not only is there a doubling of the amount of processing, and a doubling of requests needed to serve any given page, but there is also going to be some doubt (for some people) as to which URL will be presented by Google to the person doing the search.

If this discussion is about the benefits/gains between scenarios' B & C then I'm quite happy to end it here, and I'll even concede that all else being equal site B will probably come out ahead.

The thing is, we are not discussing the merits/differences of B & C simply because redirects don't enter the equation. Therefore this discussion needs to concentrate on the benefits/pitfalls/differences between scenarios' A & B only.

Do we all agree on this so far? (I hope so, else we really are all wasting each others time).

I'm going to leave this one here until I receive your responses to the question above, because that'll enable me to determine which direction I need to take and ensure we are all still on the same page, and discussing the same issue.

Cheers
Rod

NOTE: We are sorry that Rod is no longer with us.
We are grateful for all his contributions to the Zen Cart community.

Ozpost - The Ultimate Shipping module for Australian Merchants. Click these links for its Homepage, or Download from Zen-Cart.com.

30 Dec 2012, 2:50 PM
#39
rodg avatar

rodg

Deceased

Join Date:
Jan 2007
Location:
Australia
Posts:
6,263
Plugin Contributions:
4

Re: Hoping to pick the brains of some old pros!

Feznizzle:

You have completely missed my point..

Always possible, especially with some of these longer postings.

Feznizzle:

You are a smart guy. I don't expect you to go back and reread the post where I outlined my main train of thought .

I consider myself pretty smart too, which is why I did go back and re-read you post. :)

Feznizzle:

If you don't see "human factor" as integral to what I said, then I have failed to effectively use the English language. Not being sarcastic. I have many short comings, writing is but one..

Perhaps the misunderstanding was with my response rather than my not appreciating the 'human factor'.

I do fully understand and appreciate how the human factor can change the playing field, its just that as much as possible I'm trying to keep this aspect out of it... at least for the time being.. It's not that I don't consider it important (quite the opposite), but I'm really really trying to get down to the core issue, and as per the message I just posted, I think that I for one am finally coming to understand where the confusion is coming from.

Feznizzle:

Setting that aside, I am actually interested in your point of view. Very.

As I am yours, and if your thinking is along the same lines as Steve's, which I think is probably the case, I'm hoping my slight change of tact will give yet another perspective on things... Could it really be that this 'religious war'. like almost any other 'religious war', is nothing more than the result of us misunderstanding each other and talking cross purposes? IOW, you are discussing the merits/differences between my scenario B & C whilst I'm discussing the pitfalls/differences between scenario's A & B.

Did ANY of us even realise that there were the three scenario's at play here..... (Strange as it may sound, this didn't even occur to me until just now).

Feznizzle:

Please lay out ALL control aspects of what you would like to do:
Page title, content

I plan on doing nothing special with the page title.. It'll be whatever ZenCart generates by default.
I have no objection if you wish to manipulate this (I don't think it'll have an effect on what I hope to prove).

Content. I've done nothing in this regard. Thought it might be something you'd like to do... Otherwise my plan was to go to a text generating type site to create a 'random' product description. I'd suggest a paragraph or two rather than a sentence or two.

Feznizzle:

Meta?

Again, I plan to let ZenCart take care of this.
If you wish to create you own I've no objections. Again, I don't think it'll have any effect on what I hope to prove

Feznizzle:

Sitemap xml?

Our store is already 'known' by google, and since ZenCart uses a logical file structure and navigation I don't think Google have any problems in picking up the new product. As such, I don't use (or plan to use) a sitemap.
I've no objections if you do.

Feznizzle:

etc, etc

Simply put. I believe that Zen Cart does such a good job without any additional help that I plan on doing nothing special other than adding the product to the store and waiting for the results. I also have faith that the Google code is advanced enough to produce a level playing field whereby 'content rules', and the attempts to fool the system by creating Human Friendly URL's via rewrites or redirects (which is what you guys will need to do) will either be inneffective (at best) and possibly even detrimental (at worst).

IF I can prove this, then I'll have achieved what I've set out to do.
This isn't quite the same as my original intent, which was to have you 'prove' the benefits of URL rewrites/redirects (although if valid, should still be evident in the test results), but since having my 'revelation' about the 3 possible scenarios (rather than 2) I believe it will now be possible to demonstrate why scenario 'A' cannot work in the way intended for SEO purposes.

Feznizzle:

Once we are sure of the task, let's pick a time that works mutually (morning down under, evening in the sunshine state, whatever).

As long as we count morning out, I'll be ready to go whenever its most convenient for you... I don't think we need to be minute, or even hour accurate.. as long as its on the same day/afternoon/evening.

Feznizzle:

I'd like to see the first go as very minimal.

That is certainly my intent. As for you SEO guys, I say 'bring it on'. Give it all you've got (if you like). However, having said that, for your benefit I would suggest you start it simple, and if you don't get the results you need, start adding on the various SEO components to see how well you can get it. This'll help give you/us a better idea of what actually works and what has little to no effect.

Meanwhile, I'll keep my site 'zencart standard' - which will help for comparison purposes as you make your changes. Does that all seem reasonable?

I don't know about you, but I'm actually feeling a little excited about this... I know what I'm expecting to find/see but whether I'll see it or not is another matter.
I'm going to be happy regardless of the results.

Cheers
Rod.

ps. Well done guys.... this discussion is going FAR better than I hoped or expected. I was sure we'd be reduced to insulting each other and losing our tempers by now. To say I'm impressed is an understatement. It's seems ages since I've had a really good debate and exchange of ideas. I'm loving it. I truely hope you feel the same way. My only 'regret' is that it takes up a lot of time.... something that none of us probably have much to spare.

NOTE: We are sorry that Rod is no longer with us.
We are grateful for all his contributions to the Zen Cart community.

Ozpost - The Ultimate Shipping module for Australian Merchants. Click these links for its Homepage, or Download from Zen-Cart.com.

30 Dec 2012, 5:26 PM
#40
feznizzle avatar

feznizzle

Totally Zenned

Join Date:
Apr 2010
Posts:
900
Plugin Contributions:
0

Re: Hoping to pick the brains of some old pros!

My understanding of your view has gone thru phases. First I thought you believed there was zero point to keyword loading and that it was just a method for unscrupulous developers to extract more money from clients (sadly true, more often than not). Then I thought this softened to "well its not as effective as it used to be, therefore it is a pointless endeavor." Then I thought it was a performance issue (very valid, worthy of experiment) based upon redirect penalties, something I felt was overcome by 1) the way google treats those URLs (bold) and 2) the human factor. Those two reasons being those that explain why I implement the strategy by default, feeling that they override all else (again a reasoned position IMHO, worthy of experiment).

I was pretty confused when you didn't seem aware of that google factors server performance in, which to me would surely factor in all the 301 bouncing around (which I thought was your objection). Anyway, now I feel like I have a solid grasp on your position, thanks.

A) Person enters http://mysite.com/user_freindly_URL into their browser, the browser sends the request to the server, the server sends a 301 redirect (with complex URL) back to the client, the client performs a second request to the server using the complex URL, as there are no further redirects the server passes the request to the application (eg, ZenCart) which then sends the final page back to the client. (You can't tell me that this doesn't come without some sort of hit on performance).

This outlines my understanding of what was happening. This has always been a concern to me, especially since google announced a few years ago that they do indeed factor in performance as relevant to indexing.

Ceon URI defaults to 301, which is what google suggests in that PDF. I would be shocked if google indexed duplicate content.

Though I personally (GREATLY) prefer user friendly URLs but would never waste a thought on it, were it not for a desire to insert keywords into the URL. So to your 3 possibles scenarios, I would be delighted to add a 4th:
D) http://mysite.com/complex_URL_with_keywords (Application generated) (similar to the way this forum appears to work; this, I think, is what SPH was suggesting)

++++

Anyway, I'm with SPH on this. For me, this is simply business. I simply don't care what works, I just want to know.

I plan to be methodical with what switches I turn on. URL rewrites go hand and hand, so I will use them together. But I will add stuff as we go.

I don't think we should standardize the content, but I do think some things should be controlled:
1. I think we should all use EZ page rather than a product page (eliminates question of category).
2. Think we should all add the page to the footer of our site (rather then bury it completely).
3. I like the 2 or 3 paragraph limitation, but think we should standardize keyword reference:
• "Wurlitzer Debobulator" should be referenced together 3 times.
• "Wurlitzer" and "Debobulator" should each be referenced twice, not together.
4. No images (for now)

This will give us a mildly standardized page without appearing to duplicate each other. I'm still pretty concerned about the age of the domains and the established SE reputation of the sites and the performance of various servers… but anyway, I'm in.

Another thought, does anybody have access to an active adwords account? Later on, I'd like to see how that effects organic listings. I've always secretly believed it pushes your natural link further down (having a bad net effect on organic efforts). Sorry, totally random but I am very curious about that because I am contemplating using Adwords to jumpstart sales for one of my clients.

How does Jan 2 sound to you guys? Just stick it up when you find a moment...