I’m currently working on a clients website that uses CactuShop. While it claims to be ‘SEO Optimised’, it isn’t the most search engine friendly pieces of OSC shopping cart software I’ve worked with. Although it’s not the worst either.
While working through the site and seeing what can be improved on and what couldn’t, I noticed an option called ‘fakelinks’ which replaces ugly URLs with ‘search engine friendly’ ones. Sounds like a great SEO tool right? Wrong.
1. There are still ugly URLs for the same pages appearing throughout the site via breadcrumb links, featured items, related products and CactuShop don’t seem to think this is an issue:
Why aren’t all URLs friendly?
Some links in CactuShop such as those from home page featured items, related product links, search results, etc. may still be of the parameterized type rather than the staticlooking ‘friendly’ type. This is not a bug. It is not necessary for all links to be friendly; it is enough that Google and other search engines have a route right the way through to products that is ‘friendly’. The worst that can happen is for search engines not to follow these remaining ‘unfriendly’ links; but it doesn’t matter since every product can be reached anyway through an alternative ‘friendly’ route.
Actually it does matter. Ever heard of duplicate content? Ever heard of crosslinking between related pages? The use of breadcrumbs, related products, etc is only of SEO benefit as long as they all point to the same URL. What your pages are actually doing is sending internal ‘votes’ to duplicate content. I repeat, it does matter.
2. If the site has been live for a while Google will already be aware of the duplicate content and be filtering out a lot of the pagesbut those pages are in the index, some of which may rank. CactuShop is an ASP application and requires a Microsoft platform on which to run. This makes 301 redirects difficult or impossible. So if we activate ‘fakelinks’, what happens to the original URLs. Are they 404′d? Do they still exist as yet more duplicate content? Whatever happens to them it’s not good because we can’t make them redirect to the new URLs. A BIG no-no when changing URL structures (and a fast way to take a nose dive in any existing rankings).
3. The generated xml Google Sitemap includes all those duplicate unfriendly URLs. Submitting a sitemap to Google is saying “these are all the pages I want you to crawl and index”. But we don’t want those ugly duplicate versions indexing, do we?
So, that’s my fakelink rant over with. There are a number of other things I don’t like that I’m trying to make my way around but this one annoyed me.
If you have worked with CactusShop I would be interested in hearing your experiences and perhaps getting a few tips. Feel free to comment.
UPDATE August 2010: I opted not to use the fakelinks option. Instead I stuck with the dynamic url structure and focussed on the URLS that are used within the main navigation. I reduced visible links to ‘duplicate’ URLs and increased internal links to main (navigation) URLs with internal crosslinking between relevant pages. There was a noteable increase in ranks and visits in the months folowing this change.
In addition, if you use webmaster tools, you can tell Google to ignore certain parameters:
Dynamic parameters (for example, session IDs, source, or language) in your URLs can result in many different URLs all pointing to essentially the same content. For example, http://www.example.com/dresses?sid=12395923 might point to the same content as http://www.example.com/dresses. You can specify whether you want Google to ignore up to 15 specific parameters in your URL. This can result in more efficient crawling and fewer duplicate URLs, while helping to ensure that the information you need is preserved. (Note: While Google takes suggestions into account, we don’t guarantee that we’ll follow them in every case.)
On of the parameters used in duplicate pages is strPageHistory (e.g. used in ‘try these categories’ links). Please be careful before using this feature. Suggesting Google ignores this parameter if at that point it is the only URL version in the SERPs for a particular product might not be a good move. You need to build authority for the URL you WANT to appear first.
Not used Cactushop but OScommerce has a similar problem, think they called them “easylinks” or something. All they do is allow the same content to be accessed via different URLs, rather than replacing and overwriting older versions.
Proof, if proof be needbe, that good SEO and good web design don’t always come as a package.
Hi, I work with Cactushop on a daily basis as it’s our main base for all of our ecommerce customers. While I do think it is a very good ecommerce platform, easily one of the most feature rich I have come across, I have come head to head with all of the problems you list.
There are ways around all of them, although it does require a fair bit of recoding the site in order to achieve your goals.
I think the reason they have left out friendly links for every link is mostly down to laziness. The function for building these dynamic links requires that you pass in the category hierarchy of the product that you are linking to. In places where you are cross linking products or linking from a sitemap the categories have not been pulled. The solution is to make your own database call wherever it’s not putting friendly links and pass all of the data into the dynamic links function.
If duplicate content is an issue for you though you should be aware that even taking this approach will cause duplicate content if you are sharing your product across multiple categories as the user will always have two paths to the product.
I’ve found the best solution (providing the site is new and does not already have pages ranked) is to alter the link building function so that it never puts the full path of categories in the link it only ever puts the last category and the product name, you’ll also need to change the numerics that it adds to the end of the link so that they are always the same. It is painful but does work.
Finally the 301 issue. I’ve read conflicting views on the method I take but for the most part it seems to work for me. Try creating a new table in your database called 301 (or something to that effect) with two columns, oldIURL and newURL, in the 404 page before anything else is done, make a call to this table and pass in the URL the user is looking for. If the URL is found in the table have the 404 page write a 301 redirect to the page directing the user/search engine to the new page on the site.
My view is that the above works but as I say I have read that as your site has gone to the 404 page it’s already returning a success to the browser.
I’d love to write all of the above in more detail but I’m a bit short on time. I’m new to this site but I’ll keep an eye on it and reply to this thread if I can be of any help to you.
Regards
Kev Ballard
hya kev, if there was an award for the best comment to date you would have it. Thanks so much for sharing your experiences and possible solutions.
Unfortunately I don’t have access to the database. For now, I’ve taken a few steps to reduce the duplicate URLs. The more info button for products on category pages generates one of them, I changed that to a text link so it uses the same link as the image. I’ve removed the viewing history and ‘customer also bought’ lists as these too generated duplicates. For the type of site it is I’m not sure that was particularly beneficial anyway, more of a gimic if anything.
I havent submitted an XML map yet (the existing file isn’t the right format). The developers have assured me they can create one that builds URLs from the navigation and one level down, filtering out the duplicate ones. If it doesn’t do what they say then I’ll have to build and maintain it manually. Hopefully then google will have a better idea which URLs we want indexing and perhaps will help with Googles own dup content filtering.
For now I’ll have to assume I can’t have 301 redirects, which means we can’t have fakelinks either because I still don’t know if the old URLs get rewritten when fakelinks is activated (no response from the developer on that score).
Thanks again for your response.
If they have trouble creating the XML sitemap for you, try using gsitecrawler its a desktop app that spiders the site creating a sitemap, the good thing about it is you can set up rules for pages it should ignore. When I use it I tend to tell it to ignore anything with .asp in the URL then you can be sure you are only getting the friendly URLs.
Sorry if thats telling you something you already know, just thought it could be an idea if they cant generate a good sitemap for you.
If I can be of any further help let me know, it’s sad but I think I know Cactushop just about inside out.
Kev
Hi, I work for Cactusoft and a few comments.
Firstly, the content duplication issue is well dealt with by Google.
http://googlewebmastercentral.blogspot.com/2008/09/demystifying-duplicate-content-penalty.html
In short, Google will recognize duplicate content (pages accessible by multiple URLs, including ‘unfriendly’ ones) and pool the pagerank for each. This is from Google’s own webmaster blog in 2008, and I’ve not seen anything from them since to suggest in 2010 this is no longer the case. I am somewhat curious about many SE beliefs that contradict the SEs official statements on the matter – whether these are based on real evidence or that they get repeated enough to become ‘law’ regardless. If there is real evidence, I’m open to it, but generally the justifications given are normally of the type ‘everyone knows this’ or ‘it’s established fact’ or rather too commonly than should be the case ‘I know someone who works for Google’. Our Dubai office is in the same building as Google, but sadly even that does not give one the inside track on their algorithms.
That said, we’ve been working for the past 2-3 years on a new cart in ASP.NET (www.kartris.com). This has a number of SEO improvements. With regard to ‘friendly’ URLs, this is implemented throughout now. We also use the ‘canonical’ tag, which deals with the very legitimate reasons why you may have the same page accessible with different URLs. It avoids the overhead of 301 redirects, instead embedding the ‘preferred’ URL in a tag within each page. All major search engines support this now.
By not 301 redirecting, you can keep page hierarchy in the address bar and have the correct breadcrumb navigation, and still tell the SEs which URL this page should be indexed as.
The ASP.NET work has taken up 2-3 years of our time, so while we may be guilty of many things, I hope it clarifies that ‘laziness’ is not one of them
Hey Paul, I need to respond to your comments about duplicate content.
First, my opinions about duplicate content are based on a good many years worth of experience with onsite SEO. Experience doesn’t mean reading something in a forum, from another SEO or by an SE and then regurgitating it, experience means hands on work, applying changes and observing the affects. I don’t know anyone who works at Google and even if I did I would never be so lame to use it. Most people who say that tend to be bullshitters and usually mean their Adwords manager or a Google engineer they exchanged a few LOL’s with online.
I never said anything about a penalty and I am somewhat curious about why people still keep going on about penalties and bans when in reality it takes something really naughty to achieve (duplicate content caused by dynamic shopping carts is not going to get a penalty). This paragraph talks about why duplicate content matters:
If you were to compare 2 websites identical in every way except for multiple URLs:
Website 2 will fair better and I stand by that regardless of what you have read elsewhere. Don’t ask me to send you a link that backs this up because I’ve already said my views are based on actual experience. I recommend that you try and test this yourself rather than going off what you have read.
Thing is Paul, where SEO is concerned there are practices that ‘will do’ and there are practices that ‘will do better’. When a website is competing for ranks, they want the practices that ‘will do better’ because they can make one hell of a difference.
I work with small businesses and am a big fan of cart software that enables them to have an ecommerce website on a low budget. I have worked with Magento too (which in my opinion, with certain modules/add-ons is better than cactushop for SE friendliness) I do appreciate how difficult it is to build software that meets a pedantic SEO’s requirements
If I end up working with a client that uses your new software I’ll be sure to give my feedback, it sounds promising. As for the canonical tag it’s still just a band aid. While it may be supported it doesn’t mean that a website that uses canonical tag performs the same as a website that uses other methods that avoid linking to multiple URLs for the same content. All things considered, time and cost, I can see why the canonical tag may be the feasible option in many cases.
Thanks for commenting
I came across this interesting thread – I’m currently upgrading my Cactushop site to Kartris.
Kay, your argument misunderstand the demands of a large ecommerce site. Imagine I have URLs like these
http://www.mysite.com/iphone/chargers/
http://www.mysite.com/itouch/chargers/
If I have a charger that works with both iphone and itouch (i’ll call it the “widget X50″), good ecommerce software should enable me to be it in both.
http://www.mysite.com/iphone/chargers/widget-x50
http://www.mysite.com/itouch/chargers/widget-x50
Same product – two URLs. How would YOU handle that?
Fortunately google DOES handle this. It will see both URLs and pool the page rank if you use a canonical tag, using the URL you’ve told it to. The canonical tag is NOT a sticking plaster – its an essential SEO tool to help this pooling.
I agree that Kartris is way better than Cactushop in this regard (that is part of the reason I’m upgrading my site). But that doesn’t change the incorrect assertions made in this thread!
As for Magento – don’t get me started on that! Magento cops out on their demo and just has URLs like this…
http://demo.magentocommerce.com/electronics/cellphones
(which is fine)
but then links from this to
http://demo.magentocommerce.com/nokia-2610-phone.html
so in the above example the URL on magento would be
http://www.mysite.com/widget-x50.html
compared to
http://www.mysite.com/iphone/chargers/widget-x50
I’d suggest this second URL is much better as it includes “iphone” and “chargers” in the URL, rather than just the obscure product name sitting on the root.
Why not use Kartris – its an easy upgrade from Cactushop?.
Marcus. I understand large ecommerce sites very well. I’ve worked with a number of them. How many have YOU worked with? How many have YOU been responsible for the SEO for?
I think I’ve already said that if it were down to me that product pages wouldn’t have multiple URLs. There would be ONE URL.
Then your understanding of SEO is outdated. I can’t believe you are accusing me of making incorrect assertions when you are coming out with that sort of bullshit.
What planet are you on? It only becomes essential when the software produces multiple URLs. Sheesh! You are completely missing the point about how having multiple URLs reduces the potential strength for what should be ONE page.
It’s not an option right now to change the software. If it were, oddly enough, I’m not feeling very much love for Kartris right now…
[qt]Marcus. I understand large ecommerce sites very well. I’ve worked with a number of them. How many have YOU worked with? How many have YOU been responsible for the SEO for?[/qt]
I’ve been doing this since 1996 – 2 years before Google started (it was Altavista back then that we focussed on). But this seem to be a rather Ad Hominem argument you are making? – I wouldn’t claim that my years as a developer makes me right – I prefer facts.
[qt]Then your understanding of SEO is outdated. I can’t believe you are accusing me of making incorrect assertions when you are coming out with that sort of bullshit.[/qt]
Umm, I don’t know what to say. Calm down. Try to argue facts, rather than abuse. Some evidence would be nice.
Lets have an example of pooling. I used to work for these guys, so I know a little about how their ecommerce system works. The URL is there just for SEO – so long as the number is there at the end (similar to Cactus-shop and most carts I believe)
http://www.amazon.com/Passage-Justin-Cronin/dp/0345504968
(page rank 4)
Note that it gets PR4 even with no direct links to it (they all have parameters). Thats the canonical tag at work!
Now lets abuse the URL some more.
http://www.amazon.com/i-can-put-anything-here/dp/0345504968
(page rank 4)
How is this PR4? I just made up that URL – it isn’t linked anywhere? No links to anything like that URL should mean no PR right?
Wrong!
Google is much cleverer than most anyone gives it credit for. Google prefers each page to have its own URL – its obvious – they only have to index it once in that case, rather than indexing several and pooling them.
But at present the PR is pooled if you use the canonical tag. So you get the benefits of a good URL and intuitive site URL structure PLUS no PR penalty. That might change of course, but at present that is the state of affairs.
Rather than talking about how long you’ve been doing this, how big your ecommerce experience is (Amazon?) or accusing others of talking “bullshit”, its better to keep a cool head, and look at the facts. I’m a scientist by training – that’s what I try to do – separate they myths with evidence of what happens in the real world!!
Marcus, it was you who called my understanding of ecommerce sites into question and it was you who brought up how long you have been working online, not me. So it appears you are debating with yourself on that score. I know SEOs with less than a years experience who get it totally. I also know others who have years experience but because their specialist skill is perhaps another area, their understanding of SEO is fundamental. Just ‘working on’ a website does not make you responsible for the SEO or aware of all the elements that contribute towards it Marcus. I asked you how many sites you have been responsible for the SEO for. I.e. actually working towards the ranks, monitoring and reporting on an ongoing basis. Observing how changes affect the serps and visibility. Developing a site to be as SE friendly as possible (bearing in mind whatever restrictions are in place) is one thing, the ongoing SEO and understanding what activities actually contribute to the improvements/failings is quite another.
‘Pagerank pooling’ – SEO is NOT all about pagerank. Just because all ‘pagerank’ is pooled to the canonical URL does not make it equal to a website that uses ONE url instance throughout. Sites like amazon and wikipedia are large, complex, have serious domain strength and attract inbound links from external locations without trying. Just because something works for one site, doesnt mean it’s appropriate for another. You can’t use amazon as an example unless it’s a like-for-like model and how many other websites have the luxury of a site and domain such as that as a starting point?
Sometimes readers perceive the writers temperament from how they themselves are feeling. Perhaps it’s you who is not calm? I can assure you I am
If you think I’m being abusive simply because I used the word ‘bullshit’ then you should try taking your pagerank discussion to an SEO forum. I’ll try again with a nicer response:
URLs with a directory structure USED TO BE beneficial and there was a time I did recommend this structure. It is true that keywords in a domain name remain strong but it is debateable how much strength the keywords in a filename hold these days and from what I’ve been observing it’s less than it was. Today, from MY observations, the best URL is a short URL (at best from the root) rather than having them sit in various ‘keyword-rich’ folders as if to benefit the SEO (using keywords on the page copy works as well, if not better). And a page with the same content is best served with one URL visible throughout the website (to robots/crawlers and humans). The pagerank pooling effect that using the canonical tag produces is a great solution for certain complex sites that for whatever reason are not able to serve a ‘single URL’. But this does not mean it is the best solution for all websites (in the real world!!).
As a developer since 1996, why are using cart software anyway? Are you simply a user of the software as you originally insinuated, or are you a Cactusoft developer?
Kay,
I am a Cactushop developer (ie a user of their software, not an employee). I’ve also worked with various free open source systems, most recently Magento. I’ve built quite a few sites using Cactushop over the years, and I was probably one of the people who kept bugging them about getting SEO friendly URLs supported! Prior to working for a smaller development house I worked as a consultant for various large companies, mostly in the US (and their Irish subsidiaries) where I worked on site architecture, which included in part SEO. I didn’t directly monitor search ranks etc, but people reported to me who did.
I’m interested in your comments that you used to prefer keywords in domains, but don’t now. We’ve seen no major difference, against keywords in folder or the page name, but have seen them suffer where in neither.
So these are pretty much the same from SEO point of view
http://test.com/electronics_cellphones_nokia2160.html
http://test.com/electronics/cellphones/nokia2160.html
but both are better than this
http://test.com/nokia2160.html
Perhaps the sites we work on are quite different in nature?
With respect to design of URLs there is also a usability issue for customers, that is often ignored by those focussed only on SEO. Many like to trim URLs, eg
I follow search engine link to
sitedomain.com/electronics/cellphones/nokia/330.htm
you’d be surprised how many users like to manually edit the address bar to eg
sitedomain.com/electronics/cellphones/nokia/
and
sitedomain.com/electronics/cellphones/
so they can see a wider range of products. Studies show a surprising number of people do this. Sites should support it if possible.
So when designing the architecture of a site we weren’t just considering SEO (since my background entailed SEO, but not exclusively, usability etc was also important). Primarily you should design for people!!! Its important as an SEOer to consider people and not just bots. This is in part why google et all had to support canonical tag… without it we’d have to move away from a structure like this, taking usability away from those real people who like to edit addresses.
SEO is a factor in URL design, but people should come before bots. And this is why much SEO out there is flawed – since they are SEOers only, and don’t consider usability for people sufficiently – its a miopic approach that ignores the primary users of the site – the customer. I’ve no doubt more developers need to brush up on SEO knowledge, but at the same time SEOers need to broaden their understanding of usability studies, and other issues of site design.
Incidentally Amazon doesn’t structure in this way – largely because the URLs would be impossibly long!
All the best
Marcus
thanks for coming back Marcus
Referring back to the url structure and using keywords. There was a time when I would have agreed that both:
test.com/electronics_cellphones_nokia2160.html
test.com/electronics/cellphones/nokia2160.html
are better than test.com/nokia2160.html
Using those examples I say the order of preference is:
1. test.com/electronics_cellphones_nokia2160.html – IMO a short url is best (ie not within a folder structure and sitting at the root). Keywords in urls used to give quite a boost. With search engines getting smarter (and more websites using keywords in urls) it’s still good (for SEO AND usability) but isn’t the boost it used to be.
2. test.com/nokia2160.html – this is a short url and the onpage content will take care of a wider range of search terms. From a usability point of view the visitor doesnt need to see ‘electronics_cellphones’ to know that nokia is a mobile phone.
3. test.com/electronics/cellphones/nokia2160.html – as i said above I dont think the directory structure url helps seo at all anymore, i actually think it hinders it.
Having a different page for the same product so it sits in different directories targetting various keywords reminds me of when I would happily produce static ‘SEO information pages’ that went something like this:
Nokia mobile phones
Nokia cell phones
Mobile phone Nokia
Cellphone Nokia
and so on
Horrendous content but they did well. They achieved ranks for website where their dynamic pages failed because the search engines couldnt crawl them or couldnt decide what the content was. These days those ‘information pages’ would be laughed at and instead the SEO focus should be on the main product page.
It’s my opinion that a product that uses a single url, pref at root but kept short (ie not within a complex directory structure) has the most potential for building authority and therefore rank.
I agree very much with a lot of what you say. A website that is built JUST for SEO would be the most boring site ever and I cant see the conversion rate being that great. It’s tough balancing the SEO with user experience and I try to do that. I make a lot of SEO sacrifices in favour of budget, time, effort and other potential benefits of doing it in a less SEO friendly way. In my earlier days as an SEO small changes made a HUGE difference and I was a little more dogmatic.
I work with small businesses now and for many a bespoke website is not an option, they dont have the budget. I’m very open to compromise and will do my best with what I’ve got. I’m actually a fan of OSC and cart software. It’s just sometimes i get a little bit pedantic and will give minor issues more debating time than they are worth
Read the above comments with interest.
The problem with much of the SEO industry is that it lacks any scientific rigour. Instead, one is asked to trust someone on the basis that they have X number of years doing this and are an ‘expert’. No one with a background in empiricism from engineering or science is going to accept justifications like this. You need to be able to prove any theory with hard evidence to back it up.
If you are going to claim that URL format X is better than Y, then you have to be ready to provide testable and verifiable evidence. Fortunately, it is very easy to do (which makes it all the more regrettable that 99% of SEOs prefer to rely on ‘it is because I say so and I have lots of experience’).
1. Set up a site.
2. Create two or more separate pages with varying characteristics whose SEO effects you wish to test (in the case of friendly URLs, identical content, different URLs).
3. Wait a suitable period and then observe the results in Google.
4. Repeat enough times to eliminate chance or other factors from the result.
Only then can you state with confidence that URL format X is better than URL format Y. And you need to be able to display those results to others, who can verify your methodology. Just telling the world is not enough. You need to provide the data, and the URLs such that any doubting reader can check it for themselves or look for flaws in your methods that might have slanted the results.
I would thoroughly recommend the book ‘Bad Science’ by Ben Goldacre if you are in any doubt about the unreliability of anecdotal evidence and the need for good scientific method.
Now, interestingly if you do search in Google for ‘basingstokefurter’ (a made up sausage that features in our CactuShop demo data), the top 10 results all seem to point to NON-friendly URLs, with parameters. Exactly the kind we are told Google does not like. At least at the time of me writing this (of course, Google may change its mind in the future).
http://www.google.com/search?hl=en&q=basingstokefurter&btnG=Search&aq=f&aqi=&aql=&oq=&gs_rfai=
Now I confess I don’t know why this is the case.
But the point is that as an engineer, the evidence is god. It is fact because the evidence supports it, though the ‘why’ is still open for discussion.
I am quite sure that this might upset some SEOs, as there are ‘articles of faith’ in any religion and to challenge those is sacrilege. And I don’t seek to undermine anyone’s professional integrity as I am quite sure that many SEOs fully believe what they are saying. But belief and fact are not necessarily the same thing.
You can argue about beliefs, but facts speak for themselves.
I’m familiar with the book but I fail to see why this relates or why you think I doubt testing. I don’t doubt testing I just dont publish all my experiments and tests. This is a blog where I state opinions which you are free to take on board or not, it is not an academic site.
I’m not someone who makes a living just ‘telling other people what to do’ with no actual experience. Look around you there are plenty of people who only repeat and regurgitate what they have read and what they ‘think’ is evidence, these people are often self proclaimed ‘doctors’ and ‘gurus’. I’m simply a marketing consultant with a blog who shares the odd opinion (of which I’m entitled to do) based on my own experience (which includes my own tests!). You are free to disagree. But if I ever decide to become a self proclaimed doctor of SEO or guru who does nothing but provides interesting reading material and scientific studies I will no longer be an SEO because I would not have the time to actually DO any SEO. And therefore will have no business talking about it.
Now, let’s look at your ‘tests’.
I completely agree with you that you should test and I do many myself. Many SEO tests actually require more than just two different pages. You have missed out several considerations somewhere between 2&3 including how you ensure each of those pages get indexed and are ‘promoted’ equally. Many SEO comparison tests need to be carried out on completely seperate domains that are equal in every way.
I’m not sure what point you are trying to make. The post was about the Cactushop fakelink option. I do not use fakelinks for my clients website since it introduces MORE urls. Instead I have reduced the number of URLs it produces. I would rather have dynamic looking urls without keywords than multiple urls for the same product/category.
Well, it’s buggar all to do with URL structure. Google has been able to index ‘unfriendly’ urls for a few years now. Results appear for your made up phrase because the content you added this phrase to are indexed by Google. Stick that phrase anywhere. As long as the content is indexed it will appear in the SERPs (when searched for). Wait a few days, when your comment is indexed my blog will appear in the results too! Although, you will probably disregard my response to that one since it is only my opinion and I’m not providing you with any hard evidence.
I get rather a lot of hits to this page for cactusoft, cactushop and kartris terms. And I also get emails from people asking for advice, which suggests that it would be better if instead of continuing this tit for tat discussion which emerged from my critism of ONE element of cactushop, perhaps it should become a little more constructive. So how about we end with something helpful?
An email I received today is from someone who wants to know whether he should stick with cactushop v6, upgrade to kartis or dump cactusoft altogether. Paul, would you mind listing the main reasons someone should upgrade from cactushop v6 to Kartris (and specifically include any added SEO benefits)?
With respect, how should one determine the value of your opinions when you don’t provide any evidence for them? What makes your opinions worth more than mine or any other experienced web professional who may disagree with you? You could lend weight to your statements if you provided the evidence you say you have.
You failed to understand the point I made regarding ‘basingstokefurter’. Sure those pages with unfriendly links contain content which is what Google indexes. But the fact remains that those pages are accessible with both friendly and unfriendly URLs, but Google clearly prefers the unfriendly ones in 10 out of 10 cases on the first page of results.
In engineering or medicine, something like this that contradicts an accepted wisdom (and ‘friendly URLs’ are considered essential by virtually all SEOs) would be accepted in an open-minded way. Further research would attempt to confirm the findings, and then to explain why this might be the case. But empirical evidence trumps theory, ALWAYS.
I hope that SEOs can become less like religious zealots and a little more like scientists and engineers, being open-minded and keen to know the truth rather than justification for beliefs for which there may be little evidence other than personal feelings.
Turning to Kartris, since you asked… it does a number of things differently from CactuShop. It uses friendly URLs throughout, with links from non-category pages also being friendly. Category and product pages have a canonical URL tag which is unique, so duplicate content issues can be dealt with by SEs.
The text part of friendly URLs can be customized or completely changed.
Categories, products, custom pages and knowledgebase articles give full control over page title, meta-description and keywords fields.
Like I said, I am not convinced at all that friendly URLs significantly aid SE performance in any way given the evidence on searching for common CactuShop demo products. However, I do think they have a better aesthetic quality, which makes them worthwhile regardless.
We get asked a lot about SEO, and whether there are specialists we’d recommend. I know one guy who is very good, but has been somewhat out of the loop recently. My typical advice would be to search google for ‘seo’ and the guys at the top are probably the ones who’ve had most success. If they can do it for themselves, then it’s far easier to convince me they can do it for me.
SEO is ever changing and testing grounds change too. Evidence mostly becomes out of date or contaminated the moment it is published. I could publish current evidence on clients sites but that wouldn’t be wise. If you as the reader do not value my opinion you can simply keep moving. Or as you have done, add a comment which gets published so it can be read alongside. Anyone else is free to make up their own mind but, as I’m a firm believer that you should never believe what you read, they can only really understand or agree/disagree with anything I or my commenters say if they have tried it themselves.
Looking at the first result all internal links I can find point to the dynamic URL. Perhaps somewhere on the site there is a ‘friendly’ URL but when Google is deciding which URL to rank it considers the site structure and internal linking too. This is NOT evidence that Google prefers dynamic URLs, not by a long shot. Nor is it evidence that dynamic URLs can rank competitively. It is simply evidence that Google indexes them.
But that is beside the point. I’ve already stated Google has no problems indexing and ranking those URLs. When I first posted about the Cactushop fakelinks option, I expected it would rewrite ALL the URLs that are linked to across the website, not simply throw an additional URL into the pot which is what my issue was/is if you go back and read the original post.
Well of course the people at the top will be very successful in making money selling SEO but have you any idea how much time, effort and man power it takes to rank for ‘SEO’? If you have the budget to pay to replicate that then good for you but there aren’t many small businesses who could afford it. Funnily enough if you said that to your Cactushop/Kartris clients and they took your advice and signed up with one of those top ranking SEO companies I’d bet one of their first recommendations would be to dump cart software. Just sayin
I show potential customers work, ranks and results I’ve achieved for current clients and share their contact details so they can have a chat. It works for me
THanks for breaking down the SEO improvements with Kartris. Have recently got another client onboard that uses Cactushop. It’s a small business with small budget so we may have to stick with Cactushop for the time being but it is possible they will consider an upgrade. Whether that is to Kartris or Magento I’m not sure yet will just have to see how it goes and see what the current developer thinks. Thanks for your comments